{
  "best_metric": 1.57305241,
  "best_model_checkpoint": "/home/ubuntu/s2/output_qwen05B_2_epochs/qwen2_5-0_5b-instruct/v0-20250319-182107/checkpoint-68000",
  "epoch": 1.7250126839167934,
  "eval_steps": 1000,
  "global_step": 68000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "acc": 0.6415714,
      "epoch": 2.536783358701167e-05,
      "grad_norm": 8.8125,
      "learning_rate": 2.536783358701167e-09,
      "loss": 1.76253867,
      "memory(GiB)": 10.73,
      "step": 1,
      "train_speed(iter/s)": 0.0206
    },
    {
      "acc": 0.60247374,
      "epoch": 0.00012683916793505834,
      "grad_norm": 7.8125,
      "learning_rate": 1.2683916793505834e-08,
      "loss": 1.97762632,
      "memory(GiB)": 13.65,
      "step": 5,
      "train_speed(iter/s)": 0.098412
    },
    {
      "acc": 0.57780371,
      "epoch": 0.0002536783358701167,
      "grad_norm": 7.125,
      "learning_rate": 2.536783358701167e-08,
      "loss": 2.08873425,
      "memory(GiB)": 13.65,
      "step": 10,
      "train_speed(iter/s)": 0.186774
    },
    {
      "acc": 0.56816006,
      "epoch": 0.000380517503805175,
      "grad_norm": 7.34375,
      "learning_rate": 3.80517503805175e-08,
      "loss": 2.17052231,
      "memory(GiB)": 17.81,
      "step": 15,
      "train_speed(iter/s)": 0.266638
    },
    {
      "acc": 0.57780371,
      "epoch": 0.0005073566717402334,
      "grad_norm": 9.0625,
      "learning_rate": 5.073566717402334e-08,
      "loss": 2.07844791,
      "memory(GiB)": 17.81,
      "step": 20,
      "train_speed(iter/s)": 0.338851
    },
    {
      "acc": 0.57653689,
      "epoch": 0.0006341958396752917,
      "grad_norm": 8.125,
      "learning_rate": 6.341958396752917e-08,
      "loss": 2.13973999,
      "memory(GiB)": 17.81,
      "step": 25,
      "train_speed(iter/s)": 0.405014
    },
    {
      "acc": 0.58652143,
      "epoch": 0.00076103500761035,
      "grad_norm": 13.625,
      "learning_rate": 7.6103500761035e-08,
      "loss": 2.11944504,
      "memory(GiB)": 17.81,
      "step": 30,
      "train_speed(iter/s)": 0.465447
    },
    {
      "acc": 0.58463202,
      "epoch": 0.0008878741755454084,
      "grad_norm": 8.0625,
      "learning_rate": 8.878741755454084e-08,
      "loss": 2.11370773,
      "memory(GiB)": 17.81,
      "step": 35,
      "train_speed(iter/s)": 0.52101
    },
    {
      "acc": 0.59612646,
      "epoch": 0.0010147133434804667,
      "grad_norm": 8.4375,
      "learning_rate": 1.0147133434804667e-07,
      "loss": 2.06100388,
      "memory(GiB)": 17.81,
      "step": 40,
      "train_speed(iter/s)": 0.57192
    },
    {
      "acc": 0.5892786,
      "epoch": 0.001141552511415525,
      "grad_norm": 8.25,
      "learning_rate": 1.1415525114155251e-07,
      "loss": 2.03739967,
      "memory(GiB)": 17.81,
      "step": 45,
      "train_speed(iter/s)": 0.619351
    },
    {
      "acc": 0.60097218,
      "epoch": 0.0012683916793505834,
      "grad_norm": 8.5,
      "learning_rate": 1.2683916793505834e-07,
      "loss": 2.05330925,
      "memory(GiB)": 17.81,
      "step": 50,
      "train_speed(iter/s)": 0.663375
    },
    {
      "acc": 0.58988309,
      "epoch": 0.0013952308472856417,
      "grad_norm": 8.5625,
      "learning_rate": 1.3952308472856418e-07,
      "loss": 2.16794739,
      "memory(GiB)": 17.81,
      "step": 55,
      "train_speed(iter/s)": 0.704311
    },
    {
      "acc": 0.56837025,
      "epoch": 0.0015220700152207,
      "grad_norm": 9.0,
      "learning_rate": 1.5220700152207e-07,
      "loss": 2.12977352,
      "memory(GiB)": 17.81,
      "step": 60,
      "train_speed(iter/s)": 0.742433
    },
    {
      "acc": 0.58301125,
      "epoch": 0.0016489091831557584,
      "grad_norm": 8.0625,
      "learning_rate": 1.6489091831557585e-07,
      "loss": 2.06642132,
      "memory(GiB)": 17.81,
      "step": 65,
      "train_speed(iter/s)": 0.778083
    },
    {
      "acc": 0.5904233,
      "epoch": 0.0017757483510908167,
      "grad_norm": 7.375,
      "learning_rate": 1.7757483510908168e-07,
      "loss": 2.0742939,
      "memory(GiB)": 17.81,
      "step": 70,
      "train_speed(iter/s)": 0.811541
    },
    {
      "acc": 0.56817775,
      "epoch": 0.001902587519025875,
      "grad_norm": 8.9375,
      "learning_rate": 1.9025875190258752e-07,
      "loss": 2.16684647,
      "memory(GiB)": 17.81,
      "step": 75,
      "train_speed(iter/s)": 0.84281
    },
    {
      "acc": 0.59556479,
      "epoch": 0.0020294266869609334,
      "grad_norm": 8.75,
      "learning_rate": 2.0294266869609335e-07,
      "loss": 2.00218239,
      "memory(GiB)": 17.81,
      "step": 80,
      "train_speed(iter/s)": 0.872192
    },
    {
      "acc": 0.56787491,
      "epoch": 0.0021562658548959918,
      "grad_norm": 8.0,
      "learning_rate": 2.1562658548959918e-07,
      "loss": 2.19605064,
      "memory(GiB)": 23.68,
      "step": 85,
      "train_speed(iter/s)": 0.900011
    },
    {
      "acc": 0.5814374,
      "epoch": 0.00228310502283105,
      "grad_norm": 7.84375,
      "learning_rate": 2.2831050228310502e-07,
      "loss": 2.06067905,
      "memory(GiB)": 23.68,
      "step": 90,
      "train_speed(iter/s)": 0.92634
    },
    {
      "acc": 0.57568212,
      "epoch": 0.0024099441907661084,
      "grad_norm": 7.3125,
      "learning_rate": 2.409944190766109e-07,
      "loss": 2.16754417,
      "memory(GiB)": 23.7,
      "step": 95,
      "train_speed(iter/s)": 0.951167
    },
    {
      "acc": 0.58423882,
      "epoch": 0.0025367833587011668,
      "grad_norm": 8.125,
      "learning_rate": 2.536783358701167e-07,
      "loss": 2.08557796,
      "memory(GiB)": 23.7,
      "step": 100,
      "train_speed(iter/s)": 0.974621
    },
    {
      "acc": 0.57016425,
      "epoch": 0.002663622526636225,
      "grad_norm": 8.25,
      "learning_rate": 2.6636225266362255e-07,
      "loss": 2.16158848,
      "memory(GiB)": 23.7,
      "step": 105,
      "train_speed(iter/s)": 0.996991
    },
    {
      "acc": 0.59947433,
      "epoch": 0.0027904616945712835,
      "grad_norm": 6.875,
      "learning_rate": 2.7904616945712836e-07,
      "loss": 2.02864761,
      "memory(GiB)": 23.7,
      "step": 110,
      "train_speed(iter/s)": 1.018037
    },
    {
      "acc": 0.59076071,
      "epoch": 0.002917300862506342,
      "grad_norm": 8.0,
      "learning_rate": 2.917300862506342e-07,
      "loss": 2.14251175,
      "memory(GiB)": 23.7,
      "step": 115,
      "train_speed(iter/s)": 1.038235
    },
    {
      "acc": 0.59897242,
      "epoch": 0.0030441400304414,
      "grad_norm": 7.28125,
      "learning_rate": 3.0441400304414e-07,
      "loss": 2.0877779,
      "memory(GiB)": 23.7,
      "step": 120,
      "train_speed(iter/s)": 1.057373
    },
    {
      "acc": 0.59436555,
      "epoch": 0.0031709791983764585,
      "grad_norm": 8.3125,
      "learning_rate": 3.170979198376459e-07,
      "loss": 2.01410065,
      "memory(GiB)": 23.7,
      "step": 125,
      "train_speed(iter/s)": 1.075761
    },
    {
      "acc": 0.58036852,
      "epoch": 0.003297818366311517,
      "grad_norm": 7.90625,
      "learning_rate": 3.297818366311517e-07,
      "loss": 2.07639065,
      "memory(GiB)": 23.7,
      "step": 130,
      "train_speed(iter/s)": 1.093303
    },
    {
      "acc": 0.59717317,
      "epoch": 0.003424657534246575,
      "grad_norm": 7.34375,
      "learning_rate": 3.4246575342465755e-07,
      "loss": 2.04616165,
      "memory(GiB)": 23.7,
      "step": 135,
      "train_speed(iter/s)": 1.110006
    },
    {
      "acc": 0.59098282,
      "epoch": 0.0035514967021816335,
      "grad_norm": 9.6875,
      "learning_rate": 3.5514967021816336e-07,
      "loss": 2.03467312,
      "memory(GiB)": 23.7,
      "step": 140,
      "train_speed(iter/s)": 1.125831
    },
    {
      "acc": 0.58623438,
      "epoch": 0.003678335870116692,
      "grad_norm": 9.0625,
      "learning_rate": 3.678335870116692e-07,
      "loss": 2.08755283,
      "memory(GiB)": 23.7,
      "step": 145,
      "train_speed(iter/s)": 1.1411
    },
    {
      "acc": 0.59764867,
      "epoch": 0.00380517503805175,
      "grad_norm": 10.5,
      "learning_rate": 3.8051750380517503e-07,
      "loss": 2.01638031,
      "memory(GiB)": 23.7,
      "step": 150,
      "train_speed(iter/s)": 1.155741
    },
    {
      "acc": 0.58929129,
      "epoch": 0.0039320142059868085,
      "grad_norm": 10.8125,
      "learning_rate": 3.932014205986809e-07,
      "loss": 2.06034012,
      "memory(GiB)": 23.7,
      "step": 155,
      "train_speed(iter/s)": 1.169957
    },
    {
      "acc": 0.59738331,
      "epoch": 0.004058853373921867,
      "grad_norm": 7.96875,
      "learning_rate": 4.058853373921867e-07,
      "loss": 2.05501957,
      "memory(GiB)": 23.7,
      "step": 160,
      "train_speed(iter/s)": 1.183313
    },
    {
      "acc": 0.58028755,
      "epoch": 0.004185692541856925,
      "grad_norm": 9.0,
      "learning_rate": 4.1856925418569256e-07,
      "loss": 2.15213394,
      "memory(GiB)": 23.7,
      "step": 165,
      "train_speed(iter/s)": 1.195829
    },
    {
      "acc": 0.57383642,
      "epoch": 0.0043125317097919835,
      "grad_norm": 7.9375,
      "learning_rate": 4.3125317097919837e-07,
      "loss": 2.14222946,
      "memory(GiB)": 23.7,
      "step": 170,
      "train_speed(iter/s)": 1.208079
    },
    {
      "acc": 0.57469215,
      "epoch": 0.004439370877727042,
      "grad_norm": 10.75,
      "learning_rate": 4.4393708777270423e-07,
      "loss": 2.18012199,
      "memory(GiB)": 23.7,
      "step": 175,
      "train_speed(iter/s)": 1.22011
    },
    {
      "acc": 0.57810321,
      "epoch": 0.0045662100456621,
      "grad_norm": 7.65625,
      "learning_rate": 4.5662100456621004e-07,
      "loss": 2.16264229,
      "memory(GiB)": 23.7,
      "step": 180,
      "train_speed(iter/s)": 1.231613
    },
    {
      "acc": 0.60413814,
      "epoch": 0.0046930492135971585,
      "grad_norm": 8.5,
      "learning_rate": 4.693049213597159e-07,
      "loss": 2.05986958,
      "memory(GiB)": 23.7,
      "step": 185,
      "train_speed(iter/s)": 1.242767
    },
    {
      "acc": 0.57760043,
      "epoch": 0.004819888381532217,
      "grad_norm": 6.9375,
      "learning_rate": 4.819888381532218e-07,
      "loss": 2.09638023,
      "memory(GiB)": 23.7,
      "step": 190,
      "train_speed(iter/s)": 1.25346
    },
    {
      "acc": 0.59249134,
      "epoch": 0.004946727549467275,
      "grad_norm": 8.1875,
      "learning_rate": 4.946727549467275e-07,
      "loss": 2.11143036,
      "memory(GiB)": 23.7,
      "step": 195,
      "train_speed(iter/s)": 1.263822
    },
    {
      "acc": 0.58348241,
      "epoch": 0.0050735667174023336,
      "grad_norm": 7.875,
      "learning_rate": 5.073566717402334e-07,
      "loss": 2.09154167,
      "memory(GiB)": 23.7,
      "step": 200,
      "train_speed(iter/s)": 1.273627
    },
    {
      "acc": 0.58108397,
      "epoch": 0.005200405885337392,
      "grad_norm": 8.3125,
      "learning_rate": 5.200405885337392e-07,
      "loss": 2.10265102,
      "memory(GiB)": 23.7,
      "step": 205,
      "train_speed(iter/s)": 1.283285
    },
    {
      "acc": 0.58467541,
      "epoch": 0.00532724505327245,
      "grad_norm": 8.375,
      "learning_rate": 5.327245053272451e-07,
      "loss": 2.09133244,
      "memory(GiB)": 23.7,
      "step": 210,
      "train_speed(iter/s)": 1.292618
    },
    {
      "acc": 0.57472029,
      "epoch": 0.005454084221207509,
      "grad_norm": 7.65625,
      "learning_rate": 5.454084221207509e-07,
      "loss": 2.10947704,
      "memory(GiB)": 23.7,
      "step": 215,
      "train_speed(iter/s)": 1.301665
    },
    {
      "acc": 0.58943138,
      "epoch": 0.005580923389142567,
      "grad_norm": 7.34375,
      "learning_rate": 5.580923389142567e-07,
      "loss": 2.02641792,
      "memory(GiB)": 23.7,
      "step": 220,
      "train_speed(iter/s)": 1.310391
    },
    {
      "acc": 0.59376535,
      "epoch": 0.005707762557077625,
      "grad_norm": 7.875,
      "learning_rate": 5.707762557077626e-07,
      "loss": 2.04560623,
      "memory(GiB)": 23.7,
      "step": 225,
      "train_speed(iter/s)": 1.318825
    },
    {
      "acc": 0.58068542,
      "epoch": 0.005834601725012684,
      "grad_norm": 7.71875,
      "learning_rate": 5.834601725012684e-07,
      "loss": 2.09772301,
      "memory(GiB)": 23.7,
      "step": 230,
      "train_speed(iter/s)": 1.326981
    },
    {
      "acc": 0.597399,
      "epoch": 0.005961440892947742,
      "grad_norm": 7.78125,
      "learning_rate": 5.961440892947743e-07,
      "loss": 2.00362549,
      "memory(GiB)": 23.7,
      "step": 235,
      "train_speed(iter/s)": 1.334921
    },
    {
      "acc": 0.57693577,
      "epoch": 0.0060882800608828,
      "grad_norm": 7.125,
      "learning_rate": 6.0882800608828e-07,
      "loss": 2.16377869,
      "memory(GiB)": 23.7,
      "step": 240,
      "train_speed(iter/s)": 1.342669
    },
    {
      "acc": 0.59407139,
      "epoch": 0.006215119228817859,
      "grad_norm": 9.0,
      "learning_rate": 6.215119228817859e-07,
      "loss": 2.07984734,
      "memory(GiB)": 23.7,
      "step": 245,
      "train_speed(iter/s)": 1.350127
    },
    {
      "acc": 0.58396721,
      "epoch": 0.006341958396752917,
      "grad_norm": 7.65625,
      "learning_rate": 6.341958396752918e-07,
      "loss": 2.2032732,
      "memory(GiB)": 23.7,
      "step": 250,
      "train_speed(iter/s)": 1.357356
    },
    {
      "acc": 0.56820168,
      "epoch": 0.006468797564687975,
      "grad_norm": 8.5,
      "learning_rate": 6.468797564687976e-07,
      "loss": 2.147155,
      "memory(GiB)": 23.7,
      "step": 255,
      "train_speed(iter/s)": 1.364239
    },
    {
      "acc": 0.58067222,
      "epoch": 0.006595636732623034,
      "grad_norm": 8.375,
      "learning_rate": 6.595636732623034e-07,
      "loss": 2.05872955,
      "memory(GiB)": 23.7,
      "step": 260,
      "train_speed(iter/s)": 1.371152
    },
    {
      "acc": 0.59502158,
      "epoch": 0.006722475900558092,
      "grad_norm": 7.90625,
      "learning_rate": 6.722475900558092e-07,
      "loss": 2.07336369,
      "memory(GiB)": 23.7,
      "step": 265,
      "train_speed(iter/s)": 1.377754
    },
    {
      "acc": 0.60438747,
      "epoch": 0.00684931506849315,
      "grad_norm": 8.4375,
      "learning_rate": 6.849315068493151e-07,
      "loss": 2.00989799,
      "memory(GiB)": 23.7,
      "step": 270,
      "train_speed(iter/s)": 1.384326
    },
    {
      "acc": 0.5757288,
      "epoch": 0.006976154236428209,
      "grad_norm": 7.75,
      "learning_rate": 6.97615423642821e-07,
      "loss": 2.07700272,
      "memory(GiB)": 23.7,
      "step": 275,
      "train_speed(iter/s)": 1.39051
    },
    {
      "acc": 0.58312683,
      "epoch": 0.007102993404363267,
      "grad_norm": 8.125,
      "learning_rate": 7.102993404363267e-07,
      "loss": 2.09542122,
      "memory(GiB)": 23.7,
      "step": 280,
      "train_speed(iter/s)": 1.396509
    },
    {
      "acc": 0.59664135,
      "epoch": 0.007229832572298325,
      "grad_norm": 8.375,
      "learning_rate": 7.229832572298326e-07,
      "loss": 1.99681587,
      "memory(GiB)": 23.7,
      "step": 285,
      "train_speed(iter/s)": 1.402414
    },
    {
      "acc": 0.61559401,
      "epoch": 0.007356671740233384,
      "grad_norm": 8.0625,
      "learning_rate": 7.356671740233384e-07,
      "loss": 2.007934,
      "memory(GiB)": 23.7,
      "step": 290,
      "train_speed(iter/s)": 1.40824
    },
    {
      "acc": 0.59848299,
      "epoch": 0.007483510908168442,
      "grad_norm": 8.375,
      "learning_rate": 7.483510908168443e-07,
      "loss": 2.0704546,
      "memory(GiB)": 23.7,
      "step": 295,
      "train_speed(iter/s)": 1.413935
    },
    {
      "acc": 0.59056811,
      "epoch": 0.0076103500761035,
      "grad_norm": 7.34375,
      "learning_rate": 7.610350076103501e-07,
      "loss": 2.03679428,
      "memory(GiB)": 30.44,
      "step": 300,
      "train_speed(iter/s)": 1.419217
    },
    {
      "acc": 0.59642339,
      "epoch": 0.007737189244038559,
      "grad_norm": 8.4375,
      "learning_rate": 7.737189244038559e-07,
      "loss": 2.08682041,
      "memory(GiB)": 30.44,
      "step": 305,
      "train_speed(iter/s)": 1.424615
    },
    {
      "acc": 0.59266996,
      "epoch": 0.007864028411973617,
      "grad_norm": 7.5,
      "learning_rate": 7.864028411973618e-07,
      "loss": 1.98927956,
      "memory(GiB)": 30.44,
      "step": 310,
      "train_speed(iter/s)": 1.429741
    },
    {
      "acc": 0.58240619,
      "epoch": 0.007990867579908675,
      "grad_norm": 7.40625,
      "learning_rate": 7.990867579908676e-07,
      "loss": 2.15537796,
      "memory(GiB)": 30.44,
      "step": 315,
      "train_speed(iter/s)": 1.434744
    },
    {
      "acc": 0.59544802,
      "epoch": 0.008117706747843734,
      "grad_norm": 7.9375,
      "learning_rate": 8.117706747843734e-07,
      "loss": 2.0775713,
      "memory(GiB)": 30.44,
      "step": 320,
      "train_speed(iter/s)": 1.439629
    },
    {
      "acc": 0.56318431,
      "epoch": 0.008244545915778792,
      "grad_norm": 7.34375,
      "learning_rate": 8.244545915778793e-07,
      "loss": 2.14091702,
      "memory(GiB)": 30.44,
      "step": 325,
      "train_speed(iter/s)": 1.444485
    },
    {
      "acc": 0.57702899,
      "epoch": 0.00837138508371385,
      "grad_norm": 7.1875,
      "learning_rate": 8.371385083713851e-07,
      "loss": 2.12820282,
      "memory(GiB)": 30.44,
      "step": 330,
      "train_speed(iter/s)": 1.449223
    },
    {
      "acc": 0.58454962,
      "epoch": 0.008498224251648909,
      "grad_norm": 8.5,
      "learning_rate": 8.49822425164891e-07,
      "loss": 2.14468822,
      "memory(GiB)": 30.44,
      "step": 335,
      "train_speed(iter/s)": 1.453945
    },
    {
      "acc": 0.58777819,
      "epoch": 0.008625063419583967,
      "grad_norm": 7.84375,
      "learning_rate": 8.625063419583967e-07,
      "loss": 2.06611099,
      "memory(GiB)": 30.44,
      "step": 340,
      "train_speed(iter/s)": 1.458459
    },
    {
      "acc": 0.59058614,
      "epoch": 0.008751902587519025,
      "grad_norm": 6.65625,
      "learning_rate": 8.751902587519026e-07,
      "loss": 2.08925419,
      "memory(GiB)": 30.44,
      "step": 345,
      "train_speed(iter/s)": 1.462888
    },
    {
      "acc": 0.57748685,
      "epoch": 0.008878741755454084,
      "grad_norm": 7.9375,
      "learning_rate": 8.878741755454085e-07,
      "loss": 2.08326321,
      "memory(GiB)": 30.44,
      "step": 350,
      "train_speed(iter/s)": 1.467172
    },
    {
      "acc": 0.59131088,
      "epoch": 0.009005580923389142,
      "grad_norm": 9.75,
      "learning_rate": 9.005580923389143e-07,
      "loss": 2.08673401,
      "memory(GiB)": 30.44,
      "step": 355,
      "train_speed(iter/s)": 1.471447
    },
    {
      "acc": 0.58826323,
      "epoch": 0.0091324200913242,
      "grad_norm": 7.375,
      "learning_rate": 9.132420091324201e-07,
      "loss": 1.99175434,
      "memory(GiB)": 30.44,
      "step": 360,
      "train_speed(iter/s)": 1.475559
    },
    {
      "acc": 0.59775381,
      "epoch": 0.009259259259259259,
      "grad_norm": 8.3125,
      "learning_rate": 9.259259259259259e-07,
      "loss": 2.05985851,
      "memory(GiB)": 30.44,
      "step": 365,
      "train_speed(iter/s)": 1.479608
    },
    {
      "acc": 0.57983561,
      "epoch": 0.009386098427194317,
      "grad_norm": 7.65625,
      "learning_rate": 9.386098427194318e-07,
      "loss": 2.09405727,
      "memory(GiB)": 30.44,
      "step": 370,
      "train_speed(iter/s)": 1.483702
    },
    {
      "acc": 0.58051119,
      "epoch": 0.009512937595129375,
      "grad_norm": 7.21875,
      "learning_rate": 9.512937595129377e-07,
      "loss": 2.1734066,
      "memory(GiB)": 30.44,
      "step": 375,
      "train_speed(iter/s)": 1.487584
    },
    {
      "acc": 0.59764881,
      "epoch": 0.009639776763064434,
      "grad_norm": 8.875,
      "learning_rate": 9.639776763064435e-07,
      "loss": 2.01658096,
      "memory(GiB)": 30.44,
      "step": 380,
      "train_speed(iter/s)": 1.491261
    },
    {
      "acc": 0.58592234,
      "epoch": 0.009766615930999492,
      "grad_norm": 9.375,
      "learning_rate": 9.766615930999493e-07,
      "loss": 2.06452904,
      "memory(GiB)": 30.44,
      "step": 385,
      "train_speed(iter/s)": 1.494978
    },
    {
      "acc": 0.58475637,
      "epoch": 0.00989345509893455,
      "grad_norm": 7.84375,
      "learning_rate": 9.89345509893455e-07,
      "loss": 2.07754631,
      "memory(GiB)": 30.44,
      "step": 390,
      "train_speed(iter/s)": 1.498661
    },
    {
      "acc": 0.59650869,
      "epoch": 0.010020294266869609,
      "grad_norm": 7.71875,
      "learning_rate": 1.002029426686961e-06,
      "loss": 2.06810417,
      "memory(GiB)": 30.44,
      "step": 395,
      "train_speed(iter/s)": 1.502239
    },
    {
      "acc": 0.57616796,
      "epoch": 0.010147133434804667,
      "grad_norm": 10.4375,
      "learning_rate": 1.0147133434804667e-06,
      "loss": 2.12413635,
      "memory(GiB)": 30.44,
      "step": 400,
      "train_speed(iter/s)": 1.505829
    },
    {
      "acc": 0.58666806,
      "epoch": 0.010273972602739725,
      "grad_norm": 6.9375,
      "learning_rate": 1.0273972602739727e-06,
      "loss": 2.087327,
      "memory(GiB)": 30.44,
      "step": 405,
      "train_speed(iter/s)": 1.509305
    },
    {
      "acc": 0.59116058,
      "epoch": 0.010400811770674784,
      "grad_norm": 8.5,
      "learning_rate": 1.0400811770674785e-06,
      "loss": 2.09962368,
      "memory(GiB)": 30.44,
      "step": 410,
      "train_speed(iter/s)": 1.512594
    },
    {
      "acc": 0.6074688,
      "epoch": 0.010527650938609842,
      "grad_norm": 7.78125,
      "learning_rate": 1.0527650938609842e-06,
      "loss": 1.99229374,
      "memory(GiB)": 30.44,
      "step": 415,
      "train_speed(iter/s)": 1.515805
    },
    {
      "acc": 0.59611673,
      "epoch": 0.0106544901065449,
      "grad_norm": 9.0,
      "learning_rate": 1.0654490106544902e-06,
      "loss": 2.07241859,
      "memory(GiB)": 30.44,
      "step": 420,
      "train_speed(iter/s)": 1.518971
    },
    {
      "acc": 0.58405657,
      "epoch": 0.010781329274479959,
      "grad_norm": 7.21875,
      "learning_rate": 1.078132927447996e-06,
      "loss": 2.06975822,
      "memory(GiB)": 30.44,
      "step": 425,
      "train_speed(iter/s)": 1.522202
    },
    {
      "acc": 0.58463049,
      "epoch": 0.010908168442415017,
      "grad_norm": 7.25,
      "learning_rate": 1.0908168442415017e-06,
      "loss": 2.05938606,
      "memory(GiB)": 30.44,
      "step": 430,
      "train_speed(iter/s)": 1.525413
    },
    {
      "acc": 0.58507853,
      "epoch": 0.011035007610350075,
      "grad_norm": 7.96875,
      "learning_rate": 1.1035007610350077e-06,
      "loss": 2.08135872,
      "memory(GiB)": 30.44,
      "step": 435,
      "train_speed(iter/s)": 1.528495
    },
    {
      "acc": 0.59827957,
      "epoch": 0.011161846778285134,
      "grad_norm": 8.6875,
      "learning_rate": 1.1161846778285134e-06,
      "loss": 2.00571327,
      "memory(GiB)": 30.44,
      "step": 440,
      "train_speed(iter/s)": 1.531595
    },
    {
      "acc": 0.60118923,
      "epoch": 0.011288685946220192,
      "grad_norm": 6.96875,
      "learning_rate": 1.1288685946220194e-06,
      "loss": 2.01260796,
      "memory(GiB)": 30.44,
      "step": 445,
      "train_speed(iter/s)": 1.534584
    },
    {
      "acc": 0.5814682,
      "epoch": 0.01141552511415525,
      "grad_norm": 7.0,
      "learning_rate": 1.1415525114155251e-06,
      "loss": 2.06783485,
      "memory(GiB)": 30.44,
      "step": 450,
      "train_speed(iter/s)": 1.537517
    },
    {
      "acc": 0.58993053,
      "epoch": 0.011542364282090309,
      "grad_norm": 7.46875,
      "learning_rate": 1.154236428209031e-06,
      "loss": 2.06729431,
      "memory(GiB)": 30.44,
      "step": 455,
      "train_speed(iter/s)": 1.540385
    },
    {
      "acc": 0.59743342,
      "epoch": 0.011669203450025367,
      "grad_norm": 6.65625,
      "learning_rate": 1.1669203450025369e-06,
      "loss": 2.02055702,
      "memory(GiB)": 30.44,
      "step": 460,
      "train_speed(iter/s)": 1.543118
    },
    {
      "acc": 0.59117823,
      "epoch": 0.011796042617960426,
      "grad_norm": 7.375,
      "learning_rate": 1.1796042617960426e-06,
      "loss": 2.09539967,
      "memory(GiB)": 30.44,
      "step": 465,
      "train_speed(iter/s)": 1.545787
    },
    {
      "acc": 0.57558403,
      "epoch": 0.011922881785895484,
      "grad_norm": 7.375,
      "learning_rate": 1.1922881785895486e-06,
      "loss": 2.16062622,
      "memory(GiB)": 30.44,
      "step": 470,
      "train_speed(iter/s)": 1.548442
    },
    {
      "acc": 0.6126267,
      "epoch": 0.012049720953830542,
      "grad_norm": 6.125,
      "learning_rate": 1.2049720953830543e-06,
      "loss": 1.98560028,
      "memory(GiB)": 30.44,
      "step": 475,
      "train_speed(iter/s)": 1.551007
    },
    {
      "acc": 0.59188223,
      "epoch": 0.0121765601217656,
      "grad_norm": 7.3125,
      "learning_rate": 1.21765601217656e-06,
      "loss": 2.07525215,
      "memory(GiB)": 30.44,
      "step": 480,
      "train_speed(iter/s)": 1.553644
    },
    {
      "acc": 0.58786926,
      "epoch": 0.012303399289700659,
      "grad_norm": 5.21875,
      "learning_rate": 1.230339928970066e-06,
      "loss": 2.05112915,
      "memory(GiB)": 30.44,
      "step": 485,
      "train_speed(iter/s)": 1.55618
    },
    {
      "acc": 0.60624104,
      "epoch": 0.012430238457635717,
      "grad_norm": 6.4375,
      "learning_rate": 1.2430238457635718e-06,
      "loss": 2.00564232,
      "memory(GiB)": 30.44,
      "step": 490,
      "train_speed(iter/s)": 1.55861
    },
    {
      "acc": 0.59575095,
      "epoch": 0.012557077625570776,
      "grad_norm": 8.75,
      "learning_rate": 1.2557077625570776e-06,
      "loss": 2.08506622,
      "memory(GiB)": 30.44,
      "step": 495,
      "train_speed(iter/s)": 1.56104
    },
    {
      "acc": 0.58726721,
      "epoch": 0.012683916793505834,
      "grad_norm": 7.875,
      "learning_rate": 1.2683916793505835e-06,
      "loss": 2.02813663,
      "memory(GiB)": 30.44,
      "step": 500,
      "train_speed(iter/s)": 1.563386
    },
    {
      "acc": 0.57700849,
      "epoch": 0.012810755961440892,
      "grad_norm": 7.625,
      "learning_rate": 1.2810755961440893e-06,
      "loss": 2.07331543,
      "memory(GiB)": 30.44,
      "step": 505,
      "train_speed(iter/s)": 1.565667
    },
    {
      "acc": 0.59316711,
      "epoch": 0.01293759512937595,
      "grad_norm": 7.5625,
      "learning_rate": 1.2937595129375953e-06,
      "loss": 2.10827484,
      "memory(GiB)": 30.44,
      "step": 510,
      "train_speed(iter/s)": 1.567896
    },
    {
      "acc": 0.58322258,
      "epoch": 0.013064434297311009,
      "grad_norm": 7.03125,
      "learning_rate": 1.306443429731101e-06,
      "loss": 2.11230087,
      "memory(GiB)": 30.44,
      "step": 515,
      "train_speed(iter/s)": 1.57008
    },
    {
      "acc": 0.60459247,
      "epoch": 0.013191273465246067,
      "grad_norm": 6.15625,
      "learning_rate": 1.3191273465246068e-06,
      "loss": 1.98551025,
      "memory(GiB)": 30.44,
      "step": 520,
      "train_speed(iter/s)": 1.572359
    },
    {
      "acc": 0.5908164,
      "epoch": 0.013318112633181126,
      "grad_norm": 7.3125,
      "learning_rate": 1.3318112633181127e-06,
      "loss": 2.03188496,
      "memory(GiB)": 30.44,
      "step": 525,
      "train_speed(iter/s)": 1.574537
    },
    {
      "acc": 0.61041069,
      "epoch": 0.013444951801116184,
      "grad_norm": 7.375,
      "learning_rate": 1.3444951801116185e-06,
      "loss": 1.97797852,
      "memory(GiB)": 30.44,
      "step": 530,
      "train_speed(iter/s)": 1.576773
    },
    {
      "acc": 0.60151291,
      "epoch": 0.013571790969051242,
      "grad_norm": 7.3125,
      "learning_rate": 1.3571790969051243e-06,
      "loss": 1.9924551,
      "memory(GiB)": 30.44,
      "step": 535,
      "train_speed(iter/s)": 1.578905
    },
    {
      "acc": 0.59988222,
      "epoch": 0.0136986301369863,
      "grad_norm": 6.875,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 2.02952709,
      "memory(GiB)": 30.44,
      "step": 540,
      "train_speed(iter/s)": 1.581017
    },
    {
      "acc": 0.58190384,
      "epoch": 0.013825469304921359,
      "grad_norm": 7.6875,
      "learning_rate": 1.382546930492136e-06,
      "loss": 2.10898304,
      "memory(GiB)": 30.44,
      "step": 545,
      "train_speed(iter/s)": 1.58318
    },
    {
      "acc": 0.58921618,
      "epoch": 0.013952308472856417,
      "grad_norm": 7.84375,
      "learning_rate": 1.395230847285642e-06,
      "loss": 2.10822105,
      "memory(GiB)": 30.44,
      "step": 550,
      "train_speed(iter/s)": 1.5853
    },
    {
      "acc": 0.5891572,
      "epoch": 0.014079147640791476,
      "grad_norm": 7.3125,
      "learning_rate": 1.4079147640791477e-06,
      "loss": 2.01958981,
      "memory(GiB)": 30.44,
      "step": 555,
      "train_speed(iter/s)": 1.587307
    },
    {
      "acc": 0.60352993,
      "epoch": 0.014205986808726534,
      "grad_norm": 7.6875,
      "learning_rate": 1.4205986808726534e-06,
      "loss": 1.99034004,
      "memory(GiB)": 30.44,
      "step": 560,
      "train_speed(iter/s)": 1.589387
    },
    {
      "acc": 0.59245887,
      "epoch": 0.014332825976661592,
      "grad_norm": 10.125,
      "learning_rate": 1.4332825976661594e-06,
      "loss": 2.06775417,
      "memory(GiB)": 30.44,
      "step": 565,
      "train_speed(iter/s)": 1.591357
    },
    {
      "acc": 0.5993598,
      "epoch": 0.01445966514459665,
      "grad_norm": 9.4375,
      "learning_rate": 1.4459665144596652e-06,
      "loss": 1.99099979,
      "memory(GiB)": 30.44,
      "step": 570,
      "train_speed(iter/s)": 1.593288
    },
    {
      "acc": 0.61712952,
      "epoch": 0.014586504312531709,
      "grad_norm": 6.59375,
      "learning_rate": 1.458650431253171e-06,
      "loss": 1.91461067,
      "memory(GiB)": 38.53,
      "step": 575,
      "train_speed(iter/s)": 1.595075
    },
    {
      "acc": 0.58643556,
      "epoch": 0.014713343480466767,
      "grad_norm": 7.21875,
      "learning_rate": 1.4713343480466769e-06,
      "loss": 2.13691216,
      "memory(GiB)": 38.53,
      "step": 580,
      "train_speed(iter/s)": 1.596917
    },
    {
      "acc": 0.59246416,
      "epoch": 0.014840182648401826,
      "grad_norm": 7.25,
      "learning_rate": 1.4840182648401826e-06,
      "loss": 2.11300983,
      "memory(GiB)": 38.53,
      "step": 585,
      "train_speed(iter/s)": 1.598892
    },
    {
      "acc": 0.58048639,
      "epoch": 0.014967021816336884,
      "grad_norm": 6.0,
      "learning_rate": 1.4967021816336886e-06,
      "loss": 2.03772526,
      "memory(GiB)": 38.53,
      "step": 590,
      "train_speed(iter/s)": 1.60065
    },
    {
      "acc": 0.57663441,
      "epoch": 0.015093860984271942,
      "grad_norm": 6.34375,
      "learning_rate": 1.5093860984271944e-06,
      "loss": 2.04765453,
      "memory(GiB)": 38.53,
      "step": 595,
      "train_speed(iter/s)": 1.602525
    },
    {
      "acc": 0.60374422,
      "epoch": 0.015220700152207,
      "grad_norm": 6.1875,
      "learning_rate": 1.5220700152207001e-06,
      "loss": 1.97475071,
      "memory(GiB)": 38.53,
      "step": 600,
      "train_speed(iter/s)": 1.604246
    },
    {
      "acc": 0.60834856,
      "epoch": 0.015347539320142059,
      "grad_norm": 6.4375,
      "learning_rate": 1.534753932014206e-06,
      "loss": 1.9379118,
      "memory(GiB)": 38.53,
      "step": 605,
      "train_speed(iter/s)": 1.605993
    },
    {
      "acc": 0.59056187,
      "epoch": 0.015474378488077117,
      "grad_norm": 6.375,
      "learning_rate": 1.5474378488077118e-06,
      "loss": 2.02729664,
      "memory(GiB)": 38.53,
      "step": 610,
      "train_speed(iter/s)": 1.607765
    },
    {
      "acc": 0.60268025,
      "epoch": 0.015601217656012176,
      "grad_norm": 7.0625,
      "learning_rate": 1.5601217656012176e-06,
      "loss": 2.02083397,
      "memory(GiB)": 38.53,
      "step": 615,
      "train_speed(iter/s)": 1.609482
    },
    {
      "acc": 0.61031318,
      "epoch": 0.015728056823947234,
      "grad_norm": 7.03125,
      "learning_rate": 1.5728056823947236e-06,
      "loss": 1.95255795,
      "memory(GiB)": 38.53,
      "step": 620,
      "train_speed(iter/s)": 1.61111
    },
    {
      "acc": 0.59041319,
      "epoch": 0.015854895991882292,
      "grad_norm": 6.3125,
      "learning_rate": 1.5854895991882293e-06,
      "loss": 2.07051506,
      "memory(GiB)": 38.53,
      "step": 625,
      "train_speed(iter/s)": 1.612804
    },
    {
      "acc": 0.59723597,
      "epoch": 0.01598173515981735,
      "grad_norm": 6.96875,
      "learning_rate": 1.5981735159817353e-06,
      "loss": 2.05251141,
      "memory(GiB)": 38.53,
      "step": 630,
      "train_speed(iter/s)": 1.614459
    },
    {
      "acc": 0.60010281,
      "epoch": 0.01610857432775241,
      "grad_norm": 9.8125,
      "learning_rate": 1.610857432775241e-06,
      "loss": 2.0439394,
      "memory(GiB)": 38.53,
      "step": 635,
      "train_speed(iter/s)": 1.616059
    },
    {
      "acc": 0.58903327,
      "epoch": 0.016235413495687467,
      "grad_norm": 7.15625,
      "learning_rate": 1.6235413495687468e-06,
      "loss": 2.03954334,
      "memory(GiB)": 38.53,
      "step": 640,
      "train_speed(iter/s)": 1.61776
    },
    {
      "acc": 0.59707794,
      "epoch": 0.016362252663622526,
      "grad_norm": 6.5625,
      "learning_rate": 1.6362252663622528e-06,
      "loss": 2.03261719,
      "memory(GiB)": 38.53,
      "step": 645,
      "train_speed(iter/s)": 1.619225
    },
    {
      "acc": 0.58319554,
      "epoch": 0.016489091831557584,
      "grad_norm": 6.40625,
      "learning_rate": 1.6489091831557585e-06,
      "loss": 2.04252071,
      "memory(GiB)": 38.53,
      "step": 650,
      "train_speed(iter/s)": 1.620694
    },
    {
      "acc": 0.59543414,
      "epoch": 0.016615930999492642,
      "grad_norm": 6.0,
      "learning_rate": 1.6615930999492643e-06,
      "loss": 2.05091801,
      "memory(GiB)": 38.53,
      "step": 655,
      "train_speed(iter/s)": 1.622209
    },
    {
      "acc": 0.58364267,
      "epoch": 0.0167427701674277,
      "grad_norm": 6.125,
      "learning_rate": 1.6742770167427702e-06,
      "loss": 2.02139549,
      "memory(GiB)": 38.53,
      "step": 660,
      "train_speed(iter/s)": 1.623756
    },
    {
      "acc": 0.60236578,
      "epoch": 0.01686960933536276,
      "grad_norm": 6.15625,
      "learning_rate": 1.686960933536276e-06,
      "loss": 1.91847992,
      "memory(GiB)": 38.53,
      "step": 665,
      "train_speed(iter/s)": 1.625136
    },
    {
      "acc": 0.60077438,
      "epoch": 0.016996448503297817,
      "grad_norm": 6.21875,
      "learning_rate": 1.699644850329782e-06,
      "loss": 2.01930237,
      "memory(GiB)": 38.53,
      "step": 670,
      "train_speed(iter/s)": 1.626499
    },
    {
      "acc": 0.60120392,
      "epoch": 0.017123287671232876,
      "grad_norm": 7.03125,
      "learning_rate": 1.7123287671232877e-06,
      "loss": 2.00528069,
      "memory(GiB)": 38.53,
      "step": 675,
      "train_speed(iter/s)": 1.62787
    },
    {
      "acc": 0.5836062,
      "epoch": 0.017250126839167934,
      "grad_norm": 6.6875,
      "learning_rate": 1.7250126839167935e-06,
      "loss": 2.14715729,
      "memory(GiB)": 38.53,
      "step": 680,
      "train_speed(iter/s)": 1.6293
    },
    {
      "acc": 0.59741678,
      "epoch": 0.017376966007102992,
      "grad_norm": 6.15625,
      "learning_rate": 1.7376966007102994e-06,
      "loss": 2.00957355,
      "memory(GiB)": 38.53,
      "step": 685,
      "train_speed(iter/s)": 1.630563
    },
    {
      "acc": 0.60352869,
      "epoch": 0.01750380517503805,
      "grad_norm": 6.28125,
      "learning_rate": 1.7503805175038052e-06,
      "loss": 2.03679848,
      "memory(GiB)": 38.53,
      "step": 690,
      "train_speed(iter/s)": 1.631989
    },
    {
      "acc": 0.59241843,
      "epoch": 0.01763064434297311,
      "grad_norm": 8.625,
      "learning_rate": 1.7630644342973112e-06,
      "loss": 2.01553612,
      "memory(GiB)": 38.53,
      "step": 695,
      "train_speed(iter/s)": 1.633397
    },
    {
      "acc": 0.58871098,
      "epoch": 0.017757483510908167,
      "grad_norm": 5.78125,
      "learning_rate": 1.775748351090817e-06,
      "loss": 1.98986931,
      "memory(GiB)": 38.53,
      "step": 700,
      "train_speed(iter/s)": 1.63469
    },
    {
      "acc": 0.58268156,
      "epoch": 0.017884322678843226,
      "grad_norm": 6.125,
      "learning_rate": 1.7884322678843227e-06,
      "loss": 1.99861069,
      "memory(GiB)": 38.53,
      "step": 705,
      "train_speed(iter/s)": 1.636068
    },
    {
      "acc": 0.58590431,
      "epoch": 0.018011161846778284,
      "grad_norm": 7.21875,
      "learning_rate": 1.8011161846778286e-06,
      "loss": 2.05112705,
      "memory(GiB)": 38.53,
      "step": 710,
      "train_speed(iter/s)": 1.637263
    },
    {
      "acc": 0.58075576,
      "epoch": 0.018138001014713342,
      "grad_norm": 5.6875,
      "learning_rate": 1.8138001014713344e-06,
      "loss": 2.09609985,
      "memory(GiB)": 38.53,
      "step": 715,
      "train_speed(iter/s)": 1.63855
    },
    {
      "acc": 0.59707537,
      "epoch": 0.0182648401826484,
      "grad_norm": 6.78125,
      "learning_rate": 1.8264840182648401e-06,
      "loss": 1.99963741,
      "memory(GiB)": 38.53,
      "step": 720,
      "train_speed(iter/s)": 1.639946
    },
    {
      "acc": 0.58498487,
      "epoch": 0.01839167935058346,
      "grad_norm": 6.84375,
      "learning_rate": 1.8391679350583461e-06,
      "loss": 2.08599586,
      "memory(GiB)": 38.53,
      "step": 725,
      "train_speed(iter/s)": 1.641132
    },
    {
      "acc": 0.58994751,
      "epoch": 0.018518518518518517,
      "grad_norm": 7.03125,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 1.93367462,
      "memory(GiB)": 38.53,
      "step": 730,
      "train_speed(iter/s)": 1.64245
    },
    {
      "acc": 0.58264194,
      "epoch": 0.018645357686453576,
      "grad_norm": 6.03125,
      "learning_rate": 1.8645357686453578e-06,
      "loss": 2.06114101,
      "memory(GiB)": 38.53,
      "step": 735,
      "train_speed(iter/s)": 1.643655
    },
    {
      "acc": 0.59040194,
      "epoch": 0.018772196854388634,
      "grad_norm": 6.84375,
      "learning_rate": 1.8772196854388636e-06,
      "loss": 2.05004845,
      "memory(GiB)": 38.53,
      "step": 740,
      "train_speed(iter/s)": 1.644852
    },
    {
      "acc": 0.59944916,
      "epoch": 0.018899036022323693,
      "grad_norm": 6.46875,
      "learning_rate": 1.8899036022323693e-06,
      "loss": 1.94879971,
      "memory(GiB)": 38.53,
      "step": 745,
      "train_speed(iter/s)": 1.645978
    },
    {
      "acc": 0.58941746,
      "epoch": 0.01902587519025875,
      "grad_norm": 6.21875,
      "learning_rate": 1.9025875190258753e-06,
      "loss": 2.0135458,
      "memory(GiB)": 38.53,
      "step": 750,
      "train_speed(iter/s)": 1.647116
    },
    {
      "acc": 0.60376492,
      "epoch": 0.01915271435819381,
      "grad_norm": 6.34375,
      "learning_rate": 1.915271435819381e-06,
      "loss": 1.97635822,
      "memory(GiB)": 38.53,
      "step": 755,
      "train_speed(iter/s)": 1.648182
    },
    {
      "acc": 0.58112669,
      "epoch": 0.019279553526128868,
      "grad_norm": 5.6875,
      "learning_rate": 1.927955352612887e-06,
      "loss": 2.03324509,
      "memory(GiB)": 38.53,
      "step": 760,
      "train_speed(iter/s)": 1.649217
    },
    {
      "acc": 0.60946598,
      "epoch": 0.019406392694063926,
      "grad_norm": 7.0625,
      "learning_rate": 1.9406392694063926e-06,
      "loss": 2.03556252,
      "memory(GiB)": 38.53,
      "step": 765,
      "train_speed(iter/s)": 1.650329
    },
    {
      "acc": 0.5945179,
      "epoch": 0.019533231861998984,
      "grad_norm": 7.65625,
      "learning_rate": 1.9533231861998985e-06,
      "loss": 2.01756153,
      "memory(GiB)": 38.53,
      "step": 770,
      "train_speed(iter/s)": 1.651472
    },
    {
      "acc": 0.60753369,
      "epoch": 0.019660071029934043,
      "grad_norm": 6.1875,
      "learning_rate": 1.9660071029934045e-06,
      "loss": 2.01441383,
      "memory(GiB)": 38.53,
      "step": 775,
      "train_speed(iter/s)": 1.652488
    },
    {
      "acc": 0.57172837,
      "epoch": 0.0197869101978691,
      "grad_norm": 6.59375,
      "learning_rate": 1.97869101978691e-06,
      "loss": 2.08121719,
      "memory(GiB)": 38.53,
      "step": 780,
      "train_speed(iter/s)": 1.653552
    },
    {
      "acc": 0.61228228,
      "epoch": 0.01991374936580416,
      "grad_norm": 6.75,
      "learning_rate": 1.991374936580416e-06,
      "loss": 1.89397602,
      "memory(GiB)": 38.53,
      "step": 785,
      "train_speed(iter/s)": 1.654584
    },
    {
      "acc": 0.59674997,
      "epoch": 0.020040588533739218,
      "grad_norm": 6.625,
      "learning_rate": 2.004058853373922e-06,
      "loss": 2.0047966,
      "memory(GiB)": 38.53,
      "step": 790,
      "train_speed(iter/s)": 1.655673
    },
    {
      "acc": 0.60163665,
      "epoch": 0.020167427701674276,
      "grad_norm": 8.6875,
      "learning_rate": 2.016742770167428e-06,
      "loss": 1.93692665,
      "memory(GiB)": 38.53,
      "step": 795,
      "train_speed(iter/s)": 1.656734
    },
    {
      "acc": 0.60401855,
      "epoch": 0.020294266869609334,
      "grad_norm": 6.09375,
      "learning_rate": 2.0294266869609335e-06,
      "loss": 1.92346802,
      "memory(GiB)": 38.53,
      "step": 800,
      "train_speed(iter/s)": 1.657753
    },
    {
      "acc": 0.58779025,
      "epoch": 0.020421106037544393,
      "grad_norm": 6.84375,
      "learning_rate": 2.0421106037544395e-06,
      "loss": 2.02111969,
      "memory(GiB)": 38.53,
      "step": 805,
      "train_speed(iter/s)": 1.658819
    },
    {
      "acc": 0.60377512,
      "epoch": 0.02054794520547945,
      "grad_norm": 7.59375,
      "learning_rate": 2.0547945205479454e-06,
      "loss": 1.95429077,
      "memory(GiB)": 38.53,
      "step": 810,
      "train_speed(iter/s)": 1.659819
    },
    {
      "acc": 0.60720506,
      "epoch": 0.02067478437341451,
      "grad_norm": 6.78125,
      "learning_rate": 2.067478437341451e-06,
      "loss": 2.03128891,
      "memory(GiB)": 38.53,
      "step": 815,
      "train_speed(iter/s)": 1.660861
    },
    {
      "acc": 0.59849381,
      "epoch": 0.020801623541349568,
      "grad_norm": 7.65625,
      "learning_rate": 2.080162354134957e-06,
      "loss": 2.00123291,
      "memory(GiB)": 38.53,
      "step": 820,
      "train_speed(iter/s)": 1.661747
    },
    {
      "acc": 0.57835402,
      "epoch": 0.020928462709284626,
      "grad_norm": 5.96875,
      "learning_rate": 2.092846270928463e-06,
      "loss": 2.04284058,
      "memory(GiB)": 38.53,
      "step": 825,
      "train_speed(iter/s)": 1.662649
    },
    {
      "acc": 0.58629961,
      "epoch": 0.021055301877219684,
      "grad_norm": 6.71875,
      "learning_rate": 2.1055301877219685e-06,
      "loss": 2.1173254,
      "memory(GiB)": 38.53,
      "step": 830,
      "train_speed(iter/s)": 1.663679
    },
    {
      "acc": 0.60785379,
      "epoch": 0.021182141045154743,
      "grad_norm": 5.65625,
      "learning_rate": 2.1182141045154744e-06,
      "loss": 1.94242649,
      "memory(GiB)": 38.53,
      "step": 835,
      "train_speed(iter/s)": 1.664639
    },
    {
      "acc": 0.60013657,
      "epoch": 0.0213089802130898,
      "grad_norm": 8.3125,
      "learning_rate": 2.1308980213089804e-06,
      "loss": 1.96173286,
      "memory(GiB)": 47.04,
      "step": 840,
      "train_speed(iter/s)": 1.665496
    },
    {
      "acc": 0.59912806,
      "epoch": 0.02143581938102486,
      "grad_norm": 7.03125,
      "learning_rate": 2.143581938102486e-06,
      "loss": 1.94571953,
      "memory(GiB)": 47.04,
      "step": 845,
      "train_speed(iter/s)": 1.666446
    },
    {
      "acc": 0.61398377,
      "epoch": 0.021562658548959918,
      "grad_norm": 6.25,
      "learning_rate": 2.156265854895992e-06,
      "loss": 1.97966213,
      "memory(GiB)": 47.04,
      "step": 850,
      "train_speed(iter/s)": 1.667412
    },
    {
      "acc": 0.57985878,
      "epoch": 0.021689497716894976,
      "grad_norm": 7.4375,
      "learning_rate": 2.168949771689498e-06,
      "loss": 2.04382324,
      "memory(GiB)": 47.04,
      "step": 855,
      "train_speed(iter/s)": 1.668301
    },
    {
      "acc": 0.60781727,
      "epoch": 0.021816336884830034,
      "grad_norm": 6.90625,
      "learning_rate": 2.1816336884830034e-06,
      "loss": 2.01011772,
      "memory(GiB)": 47.04,
      "step": 860,
      "train_speed(iter/s)": 1.669175
    },
    {
      "acc": 0.58056746,
      "epoch": 0.021943176052765093,
      "grad_norm": 8.25,
      "learning_rate": 2.1943176052765094e-06,
      "loss": 1.98371544,
      "memory(GiB)": 47.04,
      "step": 865,
      "train_speed(iter/s)": 1.670081
    },
    {
      "acc": 0.59575796,
      "epoch": 0.02207001522070015,
      "grad_norm": 7.21875,
      "learning_rate": 2.2070015220700153e-06,
      "loss": 2.07190132,
      "memory(GiB)": 47.04,
      "step": 870,
      "train_speed(iter/s)": 1.670983
    },
    {
      "acc": 0.59166737,
      "epoch": 0.02219685438863521,
      "grad_norm": 8.1875,
      "learning_rate": 2.2196854388635213e-06,
      "loss": 2.0117609,
      "memory(GiB)": 47.04,
      "step": 875,
      "train_speed(iter/s)": 1.671883
    },
    {
      "acc": 0.60695286,
      "epoch": 0.022323693556570268,
      "grad_norm": 8.125,
      "learning_rate": 2.232369355657027e-06,
      "loss": 1.95530815,
      "memory(GiB)": 47.04,
      "step": 880,
      "train_speed(iter/s)": 1.672678
    },
    {
      "acc": 0.59405203,
      "epoch": 0.022450532724505326,
      "grad_norm": 7.96875,
      "learning_rate": 2.245053272450533e-06,
      "loss": 2.03537674,
      "memory(GiB)": 47.04,
      "step": 885,
      "train_speed(iter/s)": 1.673527
    },
    {
      "acc": 0.5937891,
      "epoch": 0.022577371892440384,
      "grad_norm": 7.15625,
      "learning_rate": 2.2577371892440388e-06,
      "loss": 2.01818981,
      "memory(GiB)": 47.04,
      "step": 890,
      "train_speed(iter/s)": 1.67439
    },
    {
      "acc": 0.61011105,
      "epoch": 0.022704211060375443,
      "grad_norm": 5.5,
      "learning_rate": 2.2704211060375443e-06,
      "loss": 2.00760136,
      "memory(GiB)": 47.04,
      "step": 895,
      "train_speed(iter/s)": 1.675213
    },
    {
      "acc": 0.60563164,
      "epoch": 0.0228310502283105,
      "grad_norm": 7.28125,
      "learning_rate": 2.2831050228310503e-06,
      "loss": 2.01051254,
      "memory(GiB)": 47.04,
      "step": 900,
      "train_speed(iter/s)": 1.676019
    },
    {
      "acc": 0.5876997,
      "epoch": 0.02295788939624556,
      "grad_norm": 6.6875,
      "learning_rate": 2.2957889396245563e-06,
      "loss": 2.02636986,
      "memory(GiB)": 47.04,
      "step": 905,
      "train_speed(iter/s)": 1.676833
    },
    {
      "acc": 0.58789148,
      "epoch": 0.023084728564180618,
      "grad_norm": 6.3125,
      "learning_rate": 2.308472856418062e-06,
      "loss": 2.01125336,
      "memory(GiB)": 47.04,
      "step": 910,
      "train_speed(iter/s)": 1.677702
    },
    {
      "acc": 0.60056543,
      "epoch": 0.023211567732115676,
      "grad_norm": 5.40625,
      "learning_rate": 2.3211567732115678e-06,
      "loss": 1.96869583,
      "memory(GiB)": 47.04,
      "step": 915,
      "train_speed(iter/s)": 1.67851
    },
    {
      "acc": 0.62972975,
      "epoch": 0.023338406900050734,
      "grad_norm": 5.21875,
      "learning_rate": 2.3338406900050737e-06,
      "loss": 1.89889717,
      "memory(GiB)": 47.04,
      "step": 920,
      "train_speed(iter/s)": 1.679346
    },
    {
      "acc": 0.59116406,
      "epoch": 0.023465246067985793,
      "grad_norm": 6.0,
      "learning_rate": 2.3465246067985793e-06,
      "loss": 2.0276123,
      "memory(GiB)": 47.04,
      "step": 925,
      "train_speed(iter/s)": 1.680174
    },
    {
      "acc": 0.60044823,
      "epoch": 0.02359208523592085,
      "grad_norm": 12.25,
      "learning_rate": 2.3592085235920852e-06,
      "loss": 1.98527832,
      "memory(GiB)": 47.04,
      "step": 930,
      "train_speed(iter/s)": 1.681039
    },
    {
      "acc": 0.60560637,
      "epoch": 0.02371892440385591,
      "grad_norm": 8.25,
      "learning_rate": 2.371892440385591e-06,
      "loss": 1.90980549,
      "memory(GiB)": 47.04,
      "step": 935,
      "train_speed(iter/s)": 1.681786
    },
    {
      "acc": 0.59694967,
      "epoch": 0.023845763571790968,
      "grad_norm": 6.46875,
      "learning_rate": 2.384576357179097e-06,
      "loss": 1.9686676,
      "memory(GiB)": 47.04,
      "step": 940,
      "train_speed(iter/s)": 1.682594
    },
    {
      "acc": 0.60201035,
      "epoch": 0.023972602739726026,
      "grad_norm": 7.84375,
      "learning_rate": 2.3972602739726027e-06,
      "loss": 1.98535118,
      "memory(GiB)": 47.04,
      "step": 945,
      "train_speed(iter/s)": 1.683453
    },
    {
      "acc": 0.59199324,
      "epoch": 0.024099441907661084,
      "grad_norm": 7.46875,
      "learning_rate": 2.4099441907661087e-06,
      "loss": 2.01176834,
      "memory(GiB)": 47.04,
      "step": 950,
      "train_speed(iter/s)": 1.684276
    },
    {
      "acc": 0.61499534,
      "epoch": 0.024226281075596143,
      "grad_norm": 10.625,
      "learning_rate": 2.4226281075596147e-06,
      "loss": 1.94725342,
      "memory(GiB)": 47.04,
      "step": 955,
      "train_speed(iter/s)": 1.685038
    },
    {
      "acc": 0.59902039,
      "epoch": 0.0243531202435312,
      "grad_norm": 8.6875,
      "learning_rate": 2.43531202435312e-06,
      "loss": 2.03876152,
      "memory(GiB)": 47.04,
      "step": 960,
      "train_speed(iter/s)": 1.68579
    },
    {
      "acc": 0.59024591,
      "epoch": 0.02447995941146626,
      "grad_norm": 6.46875,
      "learning_rate": 2.447995941146626e-06,
      "loss": 2.04958992,
      "memory(GiB)": 47.04,
      "step": 965,
      "train_speed(iter/s)": 1.686533
    },
    {
      "acc": 0.59412632,
      "epoch": 0.024606798579401318,
      "grad_norm": 5.84375,
      "learning_rate": 2.460679857940132e-06,
      "loss": 2.03471642,
      "memory(GiB)": 47.04,
      "step": 970,
      "train_speed(iter/s)": 1.687248
    },
    {
      "acc": 0.59009824,
      "epoch": 0.024733637747336376,
      "grad_norm": 6.59375,
      "learning_rate": 2.4733637747336377e-06,
      "loss": 2.05138016,
      "memory(GiB)": 47.04,
      "step": 975,
      "train_speed(iter/s)": 1.688001
    },
    {
      "acc": 0.62229795,
      "epoch": 0.024860476915271434,
      "grad_norm": 6.1875,
      "learning_rate": 2.4860476915271436e-06,
      "loss": 1.94331303,
      "memory(GiB)": 47.04,
      "step": 980,
      "train_speed(iter/s)": 1.688725
    },
    {
      "acc": 0.60050669,
      "epoch": 0.024987316083206493,
      "grad_norm": 7.0,
      "learning_rate": 2.4987316083206496e-06,
      "loss": 2.00850029,
      "memory(GiB)": 47.04,
      "step": 985,
      "train_speed(iter/s)": 1.689395
    },
    {
      "acc": 0.58993878,
      "epoch": 0.02511415525114155,
      "grad_norm": 6.625,
      "learning_rate": 2.511415525114155e-06,
      "loss": 2.06494789,
      "memory(GiB)": 47.04,
      "step": 990,
      "train_speed(iter/s)": 1.690082
    },
    {
      "acc": 0.58899655,
      "epoch": 0.02524099441907661,
      "grad_norm": 5.6875,
      "learning_rate": 2.5240994419076615e-06,
      "loss": 1.99828987,
      "memory(GiB)": 47.04,
      "step": 995,
      "train_speed(iter/s)": 1.690771
    },
    {
      "acc": 0.60925131,
      "epoch": 0.025367833587011668,
      "grad_norm": 7.9375,
      "learning_rate": 2.536783358701167e-06,
      "loss": 1.91453819,
      "memory(GiB)": 47.04,
      "step": 1000,
      "train_speed(iter/s)": 1.691461
    },
    {
      "epoch": 0.025367833587011668,
      "eval_acc": 0.5865617501675866,
      "eval_loss": 1.9484902620315552,
      "eval_runtime": 58.4022,
      "eval_samples_per_second": 109.071,
      "eval_steps_per_second": 27.276,
      "step": 1000
    },
    {
      "acc": 0.59601641,
      "epoch": 0.025494672754946726,
      "grad_norm": 6.65625,
      "learning_rate": 2.549467275494673e-06,
      "loss": 2.02587528,
      "memory(GiB)": 52.96,
      "step": 1005,
      "train_speed(iter/s)": 1.531002
    },
    {
      "acc": 0.60238309,
      "epoch": 0.025621511922881784,
      "grad_norm": 8.6875,
      "learning_rate": 2.5621511922881786e-06,
      "loss": 2.07746582,
      "memory(GiB)": 52.96,
      "step": 1010,
      "train_speed(iter/s)": 1.532202
    },
    {
      "acc": 0.59062929,
      "epoch": 0.025748351090816843,
      "grad_norm": 6.96875,
      "learning_rate": 2.5748351090816846e-06,
      "loss": 2.02643967,
      "memory(GiB)": 52.96,
      "step": 1015,
      "train_speed(iter/s)": 1.53335
    },
    {
      "acc": 0.59743261,
      "epoch": 0.0258751902587519,
      "grad_norm": 7.46875,
      "learning_rate": 2.5875190258751905e-06,
      "loss": 2.0047905,
      "memory(GiB)": 52.96,
      "step": 1020,
      "train_speed(iter/s)": 1.534553
    },
    {
      "acc": 0.61405964,
      "epoch": 0.02600202942668696,
      "grad_norm": 6.34375,
      "learning_rate": 2.6002029426686965e-06,
      "loss": 1.98388634,
      "memory(GiB)": 52.96,
      "step": 1025,
      "train_speed(iter/s)": 1.535725
    },
    {
      "acc": 0.5995717,
      "epoch": 0.026128868594622018,
      "grad_norm": 5.5625,
      "learning_rate": 2.612886859462202e-06,
      "loss": 1.96609459,
      "memory(GiB)": 52.96,
      "step": 1030,
      "train_speed(iter/s)": 1.536888
    },
    {
      "acc": 0.61300802,
      "epoch": 0.026255707762557076,
      "grad_norm": 7.03125,
      "learning_rate": 2.625570776255708e-06,
      "loss": 1.98041763,
      "memory(GiB)": 52.96,
      "step": 1035,
      "train_speed(iter/s)": 1.538054
    },
    {
      "acc": 0.59687238,
      "epoch": 0.026382546930492135,
      "grad_norm": 8.6875,
      "learning_rate": 2.6382546930492135e-06,
      "loss": 1.96327171,
      "memory(GiB)": 52.96,
      "step": 1040,
      "train_speed(iter/s)": 1.53921
    },
    {
      "acc": 0.61108966,
      "epoch": 0.026509386098427193,
      "grad_norm": 5.875,
      "learning_rate": 2.65093860984272e-06,
      "loss": 2.03963013,
      "memory(GiB)": 52.96,
      "step": 1045,
      "train_speed(iter/s)": 1.54031
    },
    {
      "acc": 0.61035624,
      "epoch": 0.02663622526636225,
      "grad_norm": 6.65625,
      "learning_rate": 2.6636225266362255e-06,
      "loss": 1.94076576,
      "memory(GiB)": 52.96,
      "step": 1050,
      "train_speed(iter/s)": 1.541463
    },
    {
      "acc": 0.59615335,
      "epoch": 0.02676306443429731,
      "grad_norm": 7.375,
      "learning_rate": 2.6763064434297314e-06,
      "loss": 1.9409626,
      "memory(GiB)": 52.96,
      "step": 1055,
      "train_speed(iter/s)": 1.542592
    },
    {
      "acc": 0.59662423,
      "epoch": 0.026889903602232368,
      "grad_norm": 5.96875,
      "learning_rate": 2.688990360223237e-06,
      "loss": 1.9571476,
      "memory(GiB)": 52.96,
      "step": 1060,
      "train_speed(iter/s)": 1.543681
    },
    {
      "acc": 0.60188313,
      "epoch": 0.027016742770167426,
      "grad_norm": 5.6875,
      "learning_rate": 2.701674277016743e-06,
      "loss": 1.95284061,
      "memory(GiB)": 52.96,
      "step": 1065,
      "train_speed(iter/s)": 1.544773
    },
    {
      "acc": 0.6035326,
      "epoch": 0.027143581938102485,
      "grad_norm": 6.75,
      "learning_rate": 2.7143581938102485e-06,
      "loss": 2.00576782,
      "memory(GiB)": 52.96,
      "step": 1070,
      "train_speed(iter/s)": 1.54589
    },
    {
      "acc": 0.59823966,
      "epoch": 0.027270421106037543,
      "grad_norm": 6.5,
      "learning_rate": 2.727042110603755e-06,
      "loss": 1.98063354,
      "memory(GiB)": 52.96,
      "step": 1075,
      "train_speed(iter/s)": 1.546951
    },
    {
      "acc": 0.580969,
      "epoch": 0.0273972602739726,
      "grad_norm": 5.90625,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 2.08416595,
      "memory(GiB)": 52.96,
      "step": 1080,
      "train_speed(iter/s)": 1.548025
    },
    {
      "acc": 0.61635456,
      "epoch": 0.02752409944190766,
      "grad_norm": 6.5,
      "learning_rate": 2.7524099441907664e-06,
      "loss": 1.94256172,
      "memory(GiB)": 52.96,
      "step": 1085,
      "train_speed(iter/s)": 1.549137
    },
    {
      "acc": 0.61907287,
      "epoch": 0.027650938609842718,
      "grad_norm": 5.65625,
      "learning_rate": 2.765093860984272e-06,
      "loss": 1.92250996,
      "memory(GiB)": 52.96,
      "step": 1090,
      "train_speed(iter/s)": 1.550195
    },
    {
      "acc": 0.58657169,
      "epoch": 0.027777777777777776,
      "grad_norm": 6.84375,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 1.98541336,
      "memory(GiB)": 52.96,
      "step": 1095,
      "train_speed(iter/s)": 1.551289
    },
    {
      "acc": 0.60796785,
      "epoch": 0.027904616945712835,
      "grad_norm": 6.8125,
      "learning_rate": 2.790461694571284e-06,
      "loss": 1.93136368,
      "memory(GiB)": 52.96,
      "step": 1100,
      "train_speed(iter/s)": 1.552339
    },
    {
      "acc": 0.60143771,
      "epoch": 0.028031456113647893,
      "grad_norm": 6.6875,
      "learning_rate": 2.80314561136479e-06,
      "loss": 1.98077202,
      "memory(GiB)": 52.96,
      "step": 1105,
      "train_speed(iter/s)": 1.5534
    },
    {
      "acc": 0.60933809,
      "epoch": 0.02815829528158295,
      "grad_norm": 7.03125,
      "learning_rate": 2.8158295281582954e-06,
      "loss": 1.92722111,
      "memory(GiB)": 52.96,
      "step": 1110,
      "train_speed(iter/s)": 1.554441
    },
    {
      "acc": 0.59950509,
      "epoch": 0.02828513444951801,
      "grad_norm": 6.40625,
      "learning_rate": 2.8285134449518014e-06,
      "loss": 2.03983688,
      "memory(GiB)": 52.96,
      "step": 1115,
      "train_speed(iter/s)": 1.555466
    },
    {
      "acc": 0.58854389,
      "epoch": 0.028411973617453068,
      "grad_norm": 6.78125,
      "learning_rate": 2.841197361745307e-06,
      "loss": 2.03416462,
      "memory(GiB)": 52.96,
      "step": 1120,
      "train_speed(iter/s)": 1.55649
    },
    {
      "acc": 0.58544197,
      "epoch": 0.028538812785388126,
      "grad_norm": 6.15625,
      "learning_rate": 2.8538812785388133e-06,
      "loss": 1.98238449,
      "memory(GiB)": 52.96,
      "step": 1125,
      "train_speed(iter/s)": 1.557527
    },
    {
      "acc": 0.60424638,
      "epoch": 0.028665651953323185,
      "grad_norm": 7.125,
      "learning_rate": 2.866565195332319e-06,
      "loss": 1.94077835,
      "memory(GiB)": 52.96,
      "step": 1130,
      "train_speed(iter/s)": 1.558518
    },
    {
      "acc": 0.60898528,
      "epoch": 0.028792491121258243,
      "grad_norm": 6.34375,
      "learning_rate": 2.879249112125825e-06,
      "loss": 1.92646523,
      "memory(GiB)": 52.96,
      "step": 1135,
      "train_speed(iter/s)": 1.559489
    },
    {
      "acc": 0.59700308,
      "epoch": 0.0289193302891933,
      "grad_norm": 6.96875,
      "learning_rate": 2.8919330289193303e-06,
      "loss": 2.03535538,
      "memory(GiB)": 52.96,
      "step": 1140,
      "train_speed(iter/s)": 1.560425
    },
    {
      "acc": 0.59102287,
      "epoch": 0.02904616945712836,
      "grad_norm": 5.3125,
      "learning_rate": 2.9046169457128363e-06,
      "loss": 1.96916809,
      "memory(GiB)": 52.96,
      "step": 1145,
      "train_speed(iter/s)": 1.561401
    },
    {
      "acc": 0.59929562,
      "epoch": 0.029173008625063418,
      "grad_norm": 6.53125,
      "learning_rate": 2.917300862506342e-06,
      "loss": 1.99325848,
      "memory(GiB)": 52.96,
      "step": 1150,
      "train_speed(iter/s)": 1.562381
    },
    {
      "acc": 0.59558825,
      "epoch": 0.029299847792998476,
      "grad_norm": 6.90625,
      "learning_rate": 2.9299847792998482e-06,
      "loss": 1.96991768,
      "memory(GiB)": 52.96,
      "step": 1155,
      "train_speed(iter/s)": 1.56342
    },
    {
      "acc": 0.61323395,
      "epoch": 0.029426686960933535,
      "grad_norm": 6.71875,
      "learning_rate": 2.9426686960933538e-06,
      "loss": 1.96245213,
      "memory(GiB)": 52.96,
      "step": 1160,
      "train_speed(iter/s)": 1.564392
    },
    {
      "acc": 0.60217309,
      "epoch": 0.029553526128868593,
      "grad_norm": 7.21875,
      "learning_rate": 2.9553526128868598e-06,
      "loss": 1.93401833,
      "memory(GiB)": 52.96,
      "step": 1165,
      "train_speed(iter/s)": 1.56536
    },
    {
      "acc": 0.59514866,
      "epoch": 0.02968036529680365,
      "grad_norm": 5.90625,
      "learning_rate": 2.9680365296803653e-06,
      "loss": 2.00032043,
      "memory(GiB)": 52.96,
      "step": 1170,
      "train_speed(iter/s)": 1.56618
    },
    {
      "acc": 0.60530767,
      "epoch": 0.02980720446473871,
      "grad_norm": 6.03125,
      "learning_rate": 2.9807204464738717e-06,
      "loss": 1.95743942,
      "memory(GiB)": 52.96,
      "step": 1175,
      "train_speed(iter/s)": 1.56715
    },
    {
      "acc": 0.61269002,
      "epoch": 0.029934043632673768,
      "grad_norm": 6.25,
      "learning_rate": 2.9934043632673772e-06,
      "loss": 1.88470802,
      "memory(GiB)": 52.96,
      "step": 1180,
      "train_speed(iter/s)": 1.568079
    },
    {
      "acc": 0.59085321,
      "epoch": 0.030060882800608826,
      "grad_norm": 6.125,
      "learning_rate": 3.006088280060883e-06,
      "loss": 1.98053436,
      "memory(GiB)": 52.96,
      "step": 1185,
      "train_speed(iter/s)": 1.569022
    },
    {
      "acc": 0.59253659,
      "epoch": 0.030187721968543885,
      "grad_norm": 6.8125,
      "learning_rate": 3.0187721968543887e-06,
      "loss": 1.98493576,
      "memory(GiB)": 55.86,
      "step": 1190,
      "train_speed(iter/s)": 1.56996
    },
    {
      "acc": 0.59364595,
      "epoch": 0.030314561136478943,
      "grad_norm": 6.65625,
      "learning_rate": 3.0314561136478947e-06,
      "loss": 1.98580132,
      "memory(GiB)": 55.86,
      "step": 1195,
      "train_speed(iter/s)": 1.570865
    },
    {
      "acc": 0.63210149,
      "epoch": 0.030441400304414,
      "grad_norm": 6.78125,
      "learning_rate": 3.0441400304414002e-06,
      "loss": 1.91288757,
      "memory(GiB)": 55.86,
      "step": 1200,
      "train_speed(iter/s)": 1.571737
    },
    {
      "acc": 0.59039435,
      "epoch": 0.03056823947234906,
      "grad_norm": 11.5,
      "learning_rate": 3.0568239472349066e-06,
      "loss": 2.07269115,
      "memory(GiB)": 55.86,
      "step": 1205,
      "train_speed(iter/s)": 1.572636
    },
    {
      "acc": 0.59262409,
      "epoch": 0.030695078640284118,
      "grad_norm": 7.4375,
      "learning_rate": 3.069507864028412e-06,
      "loss": 1.94299812,
      "memory(GiB)": 55.86,
      "step": 1210,
      "train_speed(iter/s)": 1.573537
    },
    {
      "acc": 0.60261555,
      "epoch": 0.030821917808219176,
      "grad_norm": 7.0625,
      "learning_rate": 3.082191780821918e-06,
      "loss": 1.97984161,
      "memory(GiB)": 55.86,
      "step": 1215,
      "train_speed(iter/s)": 1.574387
    },
    {
      "acc": 0.59777422,
      "epoch": 0.030948756976154235,
      "grad_norm": 7.4375,
      "learning_rate": 3.0948756976154237e-06,
      "loss": 1.96874218,
      "memory(GiB)": 55.86,
      "step": 1220,
      "train_speed(iter/s)": 1.575294
    },
    {
      "acc": 0.6122798,
      "epoch": 0.031075596144089293,
      "grad_norm": 5.78125,
      "learning_rate": 3.1075596144089297e-06,
      "loss": 1.97807961,
      "memory(GiB)": 55.86,
      "step": 1225,
      "train_speed(iter/s)": 1.576191
    },
    {
      "acc": 0.60784287,
      "epoch": 0.03120243531202435,
      "grad_norm": 6.25,
      "learning_rate": 3.120243531202435e-06,
      "loss": 1.99195366,
      "memory(GiB)": 55.86,
      "step": 1230,
      "train_speed(iter/s)": 1.577021
    },
    {
      "acc": 0.59283824,
      "epoch": 0.03132927447995941,
      "grad_norm": 6.375,
      "learning_rate": 3.1329274479959416e-06,
      "loss": 2.04663258,
      "memory(GiB)": 55.86,
      "step": 1235,
      "train_speed(iter/s)": 1.577879
    },
    {
      "acc": 0.60226316,
      "epoch": 0.03145611364789447,
      "grad_norm": 6.34375,
      "learning_rate": 3.145611364789447e-06,
      "loss": 1.97379227,
      "memory(GiB)": 55.86,
      "step": 1240,
      "train_speed(iter/s)": 1.578801
    },
    {
      "acc": 0.60626063,
      "epoch": 0.031582952815829526,
      "grad_norm": 6.84375,
      "learning_rate": 3.158295281582953e-06,
      "loss": 2.02573814,
      "memory(GiB)": 55.86,
      "step": 1245,
      "train_speed(iter/s)": 1.579659
    },
    {
      "acc": 0.6025857,
      "epoch": 0.031709791983764585,
      "grad_norm": 6.90625,
      "learning_rate": 3.1709791983764586e-06,
      "loss": 1.97974224,
      "memory(GiB)": 55.86,
      "step": 1250,
      "train_speed(iter/s)": 1.580496
    },
    {
      "acc": 0.60810347,
      "epoch": 0.03183663115169964,
      "grad_norm": 6.28125,
      "learning_rate": 3.183663115169965e-06,
      "loss": 1.98515491,
      "memory(GiB)": 55.86,
      "step": 1255,
      "train_speed(iter/s)": 1.58135
    },
    {
      "acc": 0.60704136,
      "epoch": 0.0319634703196347,
      "grad_norm": 5.875,
      "learning_rate": 3.1963470319634706e-06,
      "loss": 2.01439991,
      "memory(GiB)": 55.86,
      "step": 1260,
      "train_speed(iter/s)": 1.582191
    },
    {
      "acc": 0.60790958,
      "epoch": 0.03209030948756976,
      "grad_norm": 6.15625,
      "learning_rate": 3.2090309487569765e-06,
      "loss": 1.93412552,
      "memory(GiB)": 55.86,
      "step": 1265,
      "train_speed(iter/s)": 1.583018
    },
    {
      "acc": 0.60558281,
      "epoch": 0.03221714865550482,
      "grad_norm": 6.3125,
      "learning_rate": 3.221714865550482e-06,
      "loss": 2.01308784,
      "memory(GiB)": 55.86,
      "step": 1270,
      "train_speed(iter/s)": 1.583856
    },
    {
      "acc": 0.60381126,
      "epoch": 0.032343987823439876,
      "grad_norm": 5.5625,
      "learning_rate": 3.234398782343988e-06,
      "loss": 2.00786781,
      "memory(GiB)": 55.86,
      "step": 1275,
      "train_speed(iter/s)": 1.584656
    },
    {
      "acc": 0.6057766,
      "epoch": 0.032470826991374935,
      "grad_norm": 5.28125,
      "learning_rate": 3.2470826991374936e-06,
      "loss": 1.9677681,
      "memory(GiB)": 55.86,
      "step": 1280,
      "train_speed(iter/s)": 1.585491
    },
    {
      "acc": 0.6083662,
      "epoch": 0.03259766615930999,
      "grad_norm": 6.46875,
      "learning_rate": 3.259766615931e-06,
      "loss": 1.98414745,
      "memory(GiB)": 55.86,
      "step": 1285,
      "train_speed(iter/s)": 1.586271
    },
    {
      "acc": 0.61575689,
      "epoch": 0.03272450532724505,
      "grad_norm": 5.8125,
      "learning_rate": 3.2724505327245055e-06,
      "loss": 1.91510162,
      "memory(GiB)": 55.86,
      "step": 1290,
      "train_speed(iter/s)": 1.587107
    },
    {
      "acc": 0.59676428,
      "epoch": 0.03285134449518011,
      "grad_norm": 6.4375,
      "learning_rate": 3.2851344495180115e-06,
      "loss": 1.98412094,
      "memory(GiB)": 55.86,
      "step": 1295,
      "train_speed(iter/s)": 1.587839
    },
    {
      "acc": 0.59345059,
      "epoch": 0.03297818366311517,
      "grad_norm": 6.0,
      "learning_rate": 3.297818366311517e-06,
      "loss": 2.06957817,
      "memory(GiB)": 55.86,
      "step": 1300,
      "train_speed(iter/s)": 1.588593
    },
    {
      "acc": 0.61568213,
      "epoch": 0.033105022831050226,
      "grad_norm": 9.125,
      "learning_rate": 3.310502283105023e-06,
      "loss": 1.91323452,
      "memory(GiB)": 55.86,
      "step": 1305,
      "train_speed(iter/s)": 1.589424
    },
    {
      "acc": 0.60652704,
      "epoch": 0.033231861998985285,
      "grad_norm": 5.8125,
      "learning_rate": 3.3231861998985286e-06,
      "loss": 1.94738922,
      "memory(GiB)": 55.86,
      "step": 1310,
      "train_speed(iter/s)": 1.590254
    },
    {
      "acc": 0.62636309,
      "epoch": 0.03335870116692034,
      "grad_norm": 5.71875,
      "learning_rate": 3.335870116692035e-06,
      "loss": 1.83315792,
      "memory(GiB)": 55.86,
      "step": 1315,
      "train_speed(iter/s)": 1.591055
    },
    {
      "acc": 0.60056524,
      "epoch": 0.0334855403348554,
      "grad_norm": 6.34375,
      "learning_rate": 3.3485540334855405e-06,
      "loss": 1.95279179,
      "memory(GiB)": 55.86,
      "step": 1320,
      "train_speed(iter/s)": 1.591818
    },
    {
      "acc": 0.6080533,
      "epoch": 0.03361237950279046,
      "grad_norm": 7.15625,
      "learning_rate": 3.3612379502790465e-06,
      "loss": 1.84236565,
      "memory(GiB)": 55.86,
      "step": 1325,
      "train_speed(iter/s)": 1.592603
    },
    {
      "acc": 0.61878386,
      "epoch": 0.03373921867072552,
      "grad_norm": 8.125,
      "learning_rate": 3.373921867072552e-06,
      "loss": 1.9659586,
      "memory(GiB)": 55.86,
      "step": 1330,
      "train_speed(iter/s)": 1.593399
    },
    {
      "acc": 0.60569687,
      "epoch": 0.033866057838660577,
      "grad_norm": 5.9375,
      "learning_rate": 3.3866057838660584e-06,
      "loss": 2.02585487,
      "memory(GiB)": 55.86,
      "step": 1335,
      "train_speed(iter/s)": 1.594149
    },
    {
      "acc": 0.60939317,
      "epoch": 0.033992897006595635,
      "grad_norm": 6.1875,
      "learning_rate": 3.399289700659564e-06,
      "loss": 1.92616482,
      "memory(GiB)": 55.86,
      "step": 1340,
      "train_speed(iter/s)": 1.594921
    },
    {
      "acc": 0.62465162,
      "epoch": 0.03411973617453069,
      "grad_norm": 5.625,
      "learning_rate": 3.41197361745307e-06,
      "loss": 1.93077335,
      "memory(GiB)": 55.86,
      "step": 1345,
      "train_speed(iter/s)": 1.595681
    },
    {
      "acc": 0.59684696,
      "epoch": 0.03424657534246575,
      "grad_norm": 5.71875,
      "learning_rate": 3.4246575342465754e-06,
      "loss": 1.93622208,
      "memory(GiB)": 55.86,
      "step": 1350,
      "train_speed(iter/s)": 1.596412
    },
    {
      "acc": 0.6266459,
      "epoch": 0.03437341451040081,
      "grad_norm": 6.59375,
      "learning_rate": 3.4373414510400814e-06,
      "loss": 1.88011494,
      "memory(GiB)": 55.86,
      "step": 1355,
      "train_speed(iter/s)": 1.59716
    },
    {
      "acc": 0.61261072,
      "epoch": 0.03450025367833587,
      "grad_norm": 7.4375,
      "learning_rate": 3.450025367833587e-06,
      "loss": 1.95805855,
      "memory(GiB)": 55.86,
      "step": 1360,
      "train_speed(iter/s)": 1.597891
    },
    {
      "acc": 0.60318518,
      "epoch": 0.03462709284627093,
      "grad_norm": 6.15625,
      "learning_rate": 3.4627092846270933e-06,
      "loss": 1.96464386,
      "memory(GiB)": 55.86,
      "step": 1365,
      "train_speed(iter/s)": 1.598665
    },
    {
      "acc": 0.59814706,
      "epoch": 0.034753932014205985,
      "grad_norm": 5.96875,
      "learning_rate": 3.475393201420599e-06,
      "loss": 1.96369362,
      "memory(GiB)": 55.86,
      "step": 1370,
      "train_speed(iter/s)": 1.599384
    },
    {
      "acc": 0.60429707,
      "epoch": 0.03488077118214104,
      "grad_norm": 5.84375,
      "learning_rate": 3.488077118214105e-06,
      "loss": 1.91435337,
      "memory(GiB)": 55.86,
      "step": 1375,
      "train_speed(iter/s)": 1.600112
    },
    {
      "acc": 0.61875076,
      "epoch": 0.0350076103500761,
      "grad_norm": 5.65625,
      "learning_rate": 3.5007610350076104e-06,
      "loss": 1.95240536,
      "memory(GiB)": 55.86,
      "step": 1380,
      "train_speed(iter/s)": 1.60083
    },
    {
      "acc": 0.61646738,
      "epoch": 0.03513444951801116,
      "grad_norm": 6.15625,
      "learning_rate": 3.5134449518011164e-06,
      "loss": 1.91961479,
      "memory(GiB)": 55.86,
      "step": 1385,
      "train_speed(iter/s)": 1.601568
    },
    {
      "acc": 0.59834375,
      "epoch": 0.03526128868594622,
      "grad_norm": 5.8125,
      "learning_rate": 3.5261288685946223e-06,
      "loss": 2.00084267,
      "memory(GiB)": 55.86,
      "step": 1390,
      "train_speed(iter/s)": 1.602225
    },
    {
      "acc": 0.58981743,
      "epoch": 0.03538812785388128,
      "grad_norm": 6.0625,
      "learning_rate": 3.5388127853881283e-06,
      "loss": 2.05951519,
      "memory(GiB)": 55.86,
      "step": 1395,
      "train_speed(iter/s)": 1.60292
    },
    {
      "acc": 0.59085937,
      "epoch": 0.035514967021816335,
      "grad_norm": 5.84375,
      "learning_rate": 3.551496702181634e-06,
      "loss": 1.9258213,
      "memory(GiB)": 55.86,
      "step": 1400,
      "train_speed(iter/s)": 1.60363
    },
    {
      "acc": 0.59757051,
      "epoch": 0.03564180618975139,
      "grad_norm": 7.53125,
      "learning_rate": 3.56418061897514e-06,
      "loss": 2.03836746,
      "memory(GiB)": 55.86,
      "step": 1405,
      "train_speed(iter/s)": 1.604323
    },
    {
      "acc": 0.5950069,
      "epoch": 0.03576864535768645,
      "grad_norm": 6.84375,
      "learning_rate": 3.5768645357686453e-06,
      "loss": 2.00823231,
      "memory(GiB)": 55.86,
      "step": 1410,
      "train_speed(iter/s)": 1.605035
    },
    {
      "acc": 0.60393982,
      "epoch": 0.03589548452562151,
      "grad_norm": 5.78125,
      "learning_rate": 3.5895484525621517e-06,
      "loss": 1.97716637,
      "memory(GiB)": 55.86,
      "step": 1415,
      "train_speed(iter/s)": 1.605676
    },
    {
      "acc": 0.61806784,
      "epoch": 0.03602232369355657,
      "grad_norm": 7.21875,
      "learning_rate": 3.6022323693556573e-06,
      "loss": 1.90677795,
      "memory(GiB)": 55.86,
      "step": 1420,
      "train_speed(iter/s)": 1.606377
    },
    {
      "acc": 0.60860581,
      "epoch": 0.03614916286149163,
      "grad_norm": 6.46875,
      "learning_rate": 3.6149162861491632e-06,
      "loss": 1.98624191,
      "memory(GiB)": 55.86,
      "step": 1425,
      "train_speed(iter/s)": 1.607012
    },
    {
      "acc": 0.61801763,
      "epoch": 0.036276002029426685,
      "grad_norm": 5.625,
      "learning_rate": 3.6276002029426688e-06,
      "loss": 1.89556847,
      "memory(GiB)": 55.86,
      "step": 1430,
      "train_speed(iter/s)": 1.607647
    },
    {
      "acc": 0.59453616,
      "epoch": 0.03640284119736174,
      "grad_norm": 6.34375,
      "learning_rate": 3.6402841197361748e-06,
      "loss": 1.99719849,
      "memory(GiB)": 55.86,
      "step": 1435,
      "train_speed(iter/s)": 1.608291
    },
    {
      "acc": 0.61071548,
      "epoch": 0.0365296803652968,
      "grad_norm": 7.6875,
      "learning_rate": 3.6529680365296803e-06,
      "loss": 1.98920689,
      "memory(GiB)": 55.86,
      "step": 1440,
      "train_speed(iter/s)": 1.608964
    },
    {
      "acc": 0.62530737,
      "epoch": 0.03665651953323186,
      "grad_norm": 5.6875,
      "learning_rate": 3.6656519533231867e-06,
      "loss": 1.93359718,
      "memory(GiB)": 55.86,
      "step": 1445,
      "train_speed(iter/s)": 1.609581
    },
    {
      "acc": 0.59595771,
      "epoch": 0.03678335870116692,
      "grad_norm": 5.5625,
      "learning_rate": 3.6783358701166922e-06,
      "loss": 2.05783997,
      "memory(GiB)": 55.86,
      "step": 1450,
      "train_speed(iter/s)": 1.610219
    },
    {
      "acc": 0.62097073,
      "epoch": 0.03691019786910198,
      "grad_norm": 5.53125,
      "learning_rate": 3.691019786910198e-06,
      "loss": 1.9678236,
      "memory(GiB)": 55.86,
      "step": 1455,
      "train_speed(iter/s)": 1.610914
    },
    {
      "acc": 0.61242352,
      "epoch": 0.037037037037037035,
      "grad_norm": 6.46875,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 1.84934578,
      "memory(GiB)": 55.86,
      "step": 1460,
      "train_speed(iter/s)": 1.611589
    },
    {
      "acc": 0.60875711,
      "epoch": 0.03716387620497209,
      "grad_norm": 6.59375,
      "learning_rate": 3.7163876204972097e-06,
      "loss": 1.97669029,
      "memory(GiB)": 55.86,
      "step": 1465,
      "train_speed(iter/s)": 1.612199
    },
    {
      "acc": 0.60609407,
      "epoch": 0.03729071537290715,
      "grad_norm": 5.53125,
      "learning_rate": 3.7290715372907157e-06,
      "loss": 1.96448975,
      "memory(GiB)": 55.86,
      "step": 1470,
      "train_speed(iter/s)": 1.612835
    },
    {
      "acc": 0.60854063,
      "epoch": 0.03741755454084221,
      "grad_norm": 6.6875,
      "learning_rate": 3.7417554540842216e-06,
      "loss": 1.9614727,
      "memory(GiB)": 55.86,
      "step": 1475,
      "train_speed(iter/s)": 1.613463
    },
    {
      "acc": 0.6250205,
      "epoch": 0.03754439370877727,
      "grad_norm": 5.53125,
      "learning_rate": 3.754439370877727e-06,
      "loss": 1.88623543,
      "memory(GiB)": 55.86,
      "step": 1480,
      "train_speed(iter/s)": 1.614113
    },
    {
      "acc": 0.6044075,
      "epoch": 0.03767123287671233,
      "grad_norm": 6.90625,
      "learning_rate": 3.767123287671233e-06,
      "loss": 1.8949955,
      "memory(GiB)": 55.86,
      "step": 1485,
      "train_speed(iter/s)": 1.614737
    },
    {
      "acc": 0.60617695,
      "epoch": 0.037798072044647385,
      "grad_norm": 6.625,
      "learning_rate": 3.7798072044647387e-06,
      "loss": 1.95957718,
      "memory(GiB)": 55.86,
      "step": 1490,
      "train_speed(iter/s)": 1.615374
    },
    {
      "acc": 0.58851228,
      "epoch": 0.03792491121258244,
      "grad_norm": 7.71875,
      "learning_rate": 3.792491121258245e-06,
      "loss": 2.05935116,
      "memory(GiB)": 55.86,
      "step": 1495,
      "train_speed(iter/s)": 1.615982
    },
    {
      "acc": 0.59769611,
      "epoch": 0.0380517503805175,
      "grad_norm": 5.96875,
      "learning_rate": 3.8051750380517506e-06,
      "loss": 1.96287575,
      "memory(GiB)": 55.86,
      "step": 1500,
      "train_speed(iter/s)": 1.61658
    },
    {
      "acc": 0.60486722,
      "epoch": 0.03817858954845256,
      "grad_norm": 5.71875,
      "learning_rate": 3.817858954845256e-06,
      "loss": 1.93798161,
      "memory(GiB)": 55.86,
      "step": 1505,
      "train_speed(iter/s)": 1.617179
    },
    {
      "acc": 0.60161924,
      "epoch": 0.03830542871638762,
      "grad_norm": 6.9375,
      "learning_rate": 3.830542871638762e-06,
      "loss": 1.95662956,
      "memory(GiB)": 55.86,
      "step": 1510,
      "train_speed(iter/s)": 1.617764
    },
    {
      "acc": 0.60661659,
      "epoch": 0.03843226788432268,
      "grad_norm": 5.8125,
      "learning_rate": 3.843226788432268e-06,
      "loss": 1.90788918,
      "memory(GiB)": 55.86,
      "step": 1515,
      "train_speed(iter/s)": 1.618326
    },
    {
      "acc": 0.61499333,
      "epoch": 0.038559107052257735,
      "grad_norm": 9.75,
      "learning_rate": 3.855910705225774e-06,
      "loss": 1.90940704,
      "memory(GiB)": 55.86,
      "step": 1520,
      "train_speed(iter/s)": 1.618894
    },
    {
      "acc": 0.60315471,
      "epoch": 0.03868594622019279,
      "grad_norm": 5.65625,
      "learning_rate": 3.86859462201928e-06,
      "loss": 1.92785397,
      "memory(GiB)": 55.86,
      "step": 1525,
      "train_speed(iter/s)": 1.619498
    },
    {
      "acc": 0.59764004,
      "epoch": 0.03881278538812785,
      "grad_norm": 6.03125,
      "learning_rate": 3.881278538812785e-06,
      "loss": 1.95198708,
      "memory(GiB)": 55.86,
      "step": 1530,
      "train_speed(iter/s)": 1.620092
    },
    {
      "acc": 0.61831112,
      "epoch": 0.03893962455606291,
      "grad_norm": 6.34375,
      "learning_rate": 3.893962455606292e-06,
      "loss": 1.94009991,
      "memory(GiB)": 55.86,
      "step": 1535,
      "train_speed(iter/s)": 1.620668
    },
    {
      "acc": 0.62329388,
      "epoch": 0.03906646372399797,
      "grad_norm": 7.40625,
      "learning_rate": 3.906646372399797e-06,
      "loss": 1.85862293,
      "memory(GiB)": 55.86,
      "step": 1540,
      "train_speed(iter/s)": 1.621287
    },
    {
      "acc": 0.5830049,
      "epoch": 0.03919330289193303,
      "grad_norm": 6.15625,
      "learning_rate": 3.919330289193303e-06,
      "loss": 2.00897255,
      "memory(GiB)": 55.86,
      "step": 1545,
      "train_speed(iter/s)": 1.621806
    },
    {
      "acc": 0.5943553,
      "epoch": 0.039320142059868085,
      "grad_norm": 6.25,
      "learning_rate": 3.932014205986809e-06,
      "loss": 1.96419506,
      "memory(GiB)": 55.86,
      "step": 1550,
      "train_speed(iter/s)": 1.62245
    },
    {
      "acc": 0.62053938,
      "epoch": 0.03944698122780314,
      "grad_norm": 6.625,
      "learning_rate": 3.944698122780315e-06,
      "loss": 1.90389671,
      "memory(GiB)": 55.86,
      "step": 1555,
      "train_speed(iter/s)": 1.623037
    },
    {
      "acc": 0.60830789,
      "epoch": 0.0395738203957382,
      "grad_norm": 9.5,
      "learning_rate": 3.95738203957382e-06,
      "loss": 1.97414207,
      "memory(GiB)": 55.86,
      "step": 1560,
      "train_speed(iter/s)": 1.623615
    },
    {
      "acc": 0.61599102,
      "epoch": 0.03970065956367326,
      "grad_norm": 6.59375,
      "learning_rate": 3.970065956367327e-06,
      "loss": 1.9083416,
      "memory(GiB)": 55.86,
      "step": 1565,
      "train_speed(iter/s)": 1.624233
    },
    {
      "acc": 0.60456533,
      "epoch": 0.03982749873160832,
      "grad_norm": 6.09375,
      "learning_rate": 3.982749873160832e-06,
      "loss": 1.94531078,
      "memory(GiB)": 55.86,
      "step": 1570,
      "train_speed(iter/s)": 1.624847
    },
    {
      "acc": 0.58893738,
      "epoch": 0.03995433789954338,
      "grad_norm": 8.5,
      "learning_rate": 3.995433789954338e-06,
      "loss": 2.01982937,
      "memory(GiB)": 55.86,
      "step": 1575,
      "train_speed(iter/s)": 1.625461
    },
    {
      "acc": 0.61439967,
      "epoch": 0.040081177067478435,
      "grad_norm": 5.1875,
      "learning_rate": 4.008117706747844e-06,
      "loss": 1.96854458,
      "memory(GiB)": 55.86,
      "step": 1580,
      "train_speed(iter/s)": 1.626016
    },
    {
      "acc": 0.60537062,
      "epoch": 0.04020801623541349,
      "grad_norm": 7.65625,
      "learning_rate": 4.02080162354135e-06,
      "loss": 1.90504494,
      "memory(GiB)": 55.86,
      "step": 1585,
      "train_speed(iter/s)": 1.626596
    },
    {
      "acc": 0.62116857,
      "epoch": 0.04033485540334855,
      "grad_norm": 6.6875,
      "learning_rate": 4.033485540334856e-06,
      "loss": 1.89018879,
      "memory(GiB)": 55.86,
      "step": 1590,
      "train_speed(iter/s)": 1.627163
    },
    {
      "acc": 0.60072103,
      "epoch": 0.04046169457128361,
      "grad_norm": 5.4375,
      "learning_rate": 4.046169457128362e-06,
      "loss": 1.97833405,
      "memory(GiB)": 55.86,
      "step": 1595,
      "train_speed(iter/s)": 1.627719
    },
    {
      "acc": 0.5929966,
      "epoch": 0.04058853373921867,
      "grad_norm": 6.84375,
      "learning_rate": 4.058853373921867e-06,
      "loss": 1.9963623,
      "memory(GiB)": 55.86,
      "step": 1600,
      "train_speed(iter/s)": 1.628297
    },
    {
      "acc": 0.62433748,
      "epoch": 0.04071537290715373,
      "grad_norm": 5.75,
      "learning_rate": 4.071537290715373e-06,
      "loss": 1.83874855,
      "memory(GiB)": 55.86,
      "step": 1605,
      "train_speed(iter/s)": 1.628839
    },
    {
      "acc": 0.60330362,
      "epoch": 0.040842212075088785,
      "grad_norm": 7.78125,
      "learning_rate": 4.084221207508879e-06,
      "loss": 1.93582878,
      "memory(GiB)": 55.86,
      "step": 1610,
      "train_speed(iter/s)": 1.629402
    },
    {
      "acc": 0.60428982,
      "epoch": 0.040969051243023843,
      "grad_norm": 8.5625,
      "learning_rate": 4.096905124302385e-06,
      "loss": 1.9523243,
      "memory(GiB)": 55.86,
      "step": 1615,
      "train_speed(iter/s)": 1.629974
    },
    {
      "acc": 0.61570668,
      "epoch": 0.0410958904109589,
      "grad_norm": 6.0,
      "learning_rate": 4.109589041095891e-06,
      "loss": 1.8981863,
      "memory(GiB)": 55.86,
      "step": 1620,
      "train_speed(iter/s)": 1.630511
    },
    {
      "acc": 0.59638033,
      "epoch": 0.04122272957889396,
      "grad_norm": 6.25,
      "learning_rate": 4.122272957889397e-06,
      "loss": 2.06798515,
      "memory(GiB)": 55.86,
      "step": 1625,
      "train_speed(iter/s)": 1.631054
    },
    {
      "acc": 0.59134579,
      "epoch": 0.04134956874682902,
      "grad_norm": 7.875,
      "learning_rate": 4.134956874682902e-06,
      "loss": 1.95156212,
      "memory(GiB)": 55.86,
      "step": 1630,
      "train_speed(iter/s)": 1.631589
    },
    {
      "acc": 0.63578887,
      "epoch": 0.04147640791476408,
      "grad_norm": 6.28125,
      "learning_rate": 4.147640791476408e-06,
      "loss": 1.8266901,
      "memory(GiB)": 55.86,
      "step": 1635,
      "train_speed(iter/s)": 1.632146
    },
    {
      "acc": 0.60775814,
      "epoch": 0.041603247082699135,
      "grad_norm": 5.53125,
      "learning_rate": 4.160324708269914e-06,
      "loss": 1.98309479,
      "memory(GiB)": 55.86,
      "step": 1640,
      "train_speed(iter/s)": 1.632649
    },
    {
      "acc": 0.6237278,
      "epoch": 0.041730086250634194,
      "grad_norm": 5.625,
      "learning_rate": 4.17300862506342e-06,
      "loss": 1.82640667,
      "memory(GiB)": 55.86,
      "step": 1645,
      "train_speed(iter/s)": 1.633162
    },
    {
      "acc": 0.61207809,
      "epoch": 0.04185692541856925,
      "grad_norm": 6.21875,
      "learning_rate": 4.185692541856926e-06,
      "loss": 1.84851265,
      "memory(GiB)": 55.86,
      "step": 1650,
      "train_speed(iter/s)": 1.633689
    },
    {
      "acc": 0.61386757,
      "epoch": 0.04198376458650431,
      "grad_norm": 7.5625,
      "learning_rate": 4.198376458650432e-06,
      "loss": 1.88960037,
      "memory(GiB)": 55.86,
      "step": 1655,
      "train_speed(iter/s)": 1.634223
    },
    {
      "acc": 0.61731625,
      "epoch": 0.04211060375443937,
      "grad_norm": 6.96875,
      "learning_rate": 4.211060375443937e-06,
      "loss": 1.85162468,
      "memory(GiB)": 55.86,
      "step": 1660,
      "train_speed(iter/s)": 1.63477
    },
    {
      "acc": 0.59406047,
      "epoch": 0.04223744292237443,
      "grad_norm": 6.4375,
      "learning_rate": 4.223744292237444e-06,
      "loss": 2.05021172,
      "memory(GiB)": 55.86,
      "step": 1665,
      "train_speed(iter/s)": 1.635297
    },
    {
      "acc": 0.60849686,
      "epoch": 0.042364282090309485,
      "grad_norm": 5.8125,
      "learning_rate": 4.236428209030949e-06,
      "loss": 1.91598587,
      "memory(GiB)": 55.86,
      "step": 1670,
      "train_speed(iter/s)": 1.635803
    },
    {
      "acc": 0.599365,
      "epoch": 0.042491121258244544,
      "grad_norm": 7.28125,
      "learning_rate": 4.249112125824455e-06,
      "loss": 1.9791338,
      "memory(GiB)": 55.86,
      "step": 1675,
      "train_speed(iter/s)": 1.636318
    },
    {
      "acc": 0.6116539,
      "epoch": 0.0426179604261796,
      "grad_norm": 9.75,
      "learning_rate": 4.261796042617961e-06,
      "loss": 1.91320686,
      "memory(GiB)": 55.86,
      "step": 1680,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.63092976,
      "epoch": 0.04274479959411466,
      "grad_norm": 7.53125,
      "learning_rate": 4.274479959411467e-06,
      "loss": 1.83175011,
      "memory(GiB)": 55.86,
      "step": 1685,
      "train_speed(iter/s)": 1.637347
    },
    {
      "acc": 0.61188216,
      "epoch": 0.04287163876204972,
      "grad_norm": 5.75,
      "learning_rate": 4.287163876204972e-06,
      "loss": 1.84597168,
      "memory(GiB)": 55.86,
      "step": 1690,
      "train_speed(iter/s)": 1.637889
    },
    {
      "acc": 0.6056201,
      "epoch": 0.04299847792998478,
      "grad_norm": 6.125,
      "learning_rate": 4.299847792998479e-06,
      "loss": 1.87785301,
      "memory(GiB)": 55.86,
      "step": 1695,
      "train_speed(iter/s)": 1.638433
    },
    {
      "acc": 0.61106491,
      "epoch": 0.043125317097919835,
      "grad_norm": 7.4375,
      "learning_rate": 4.312531709791984e-06,
      "loss": 1.91492901,
      "memory(GiB)": 55.86,
      "step": 1700,
      "train_speed(iter/s)": 1.638945
    },
    {
      "acc": 0.59373679,
      "epoch": 0.043252156265854894,
      "grad_norm": 8.1875,
      "learning_rate": 4.32521562658549e-06,
      "loss": 1.9386467,
      "memory(GiB)": 55.86,
      "step": 1705,
      "train_speed(iter/s)": 1.639446
    },
    {
      "acc": 0.60725961,
      "epoch": 0.04337899543378995,
      "grad_norm": 6.65625,
      "learning_rate": 4.337899543378996e-06,
      "loss": 1.97910957,
      "memory(GiB)": 55.86,
      "step": 1710,
      "train_speed(iter/s)": 1.639955
    },
    {
      "acc": 0.58542371,
      "epoch": 0.04350583460172501,
      "grad_norm": 5.46875,
      "learning_rate": 4.350583460172502e-06,
      "loss": 1.97228317,
      "memory(GiB)": 55.86,
      "step": 1715,
      "train_speed(iter/s)": 1.640438
    },
    {
      "acc": 0.62461662,
      "epoch": 0.04363267376966007,
      "grad_norm": 6.6875,
      "learning_rate": 4.363267376966007e-06,
      "loss": 1.87988815,
      "memory(GiB)": 55.86,
      "step": 1720,
      "train_speed(iter/s)": 1.64093
    },
    {
      "acc": 0.60323882,
      "epoch": 0.04375951293759513,
      "grad_norm": 5.875,
      "learning_rate": 4.375951293759514e-06,
      "loss": 1.95035305,
      "memory(GiB)": 55.86,
      "step": 1725,
      "train_speed(iter/s)": 1.64145
    },
    {
      "acc": 0.61045351,
      "epoch": 0.043886352105530185,
      "grad_norm": 6.625,
      "learning_rate": 4.388635210553019e-06,
      "loss": 1.90547295,
      "memory(GiB)": 55.86,
      "step": 1730,
      "train_speed(iter/s)": 1.64195
    },
    {
      "acc": 0.60988708,
      "epoch": 0.044013191273465244,
      "grad_norm": 5.71875,
      "learning_rate": 4.401319127346525e-06,
      "loss": 1.92009602,
      "memory(GiB)": 55.86,
      "step": 1735,
      "train_speed(iter/s)": 1.642447
    },
    {
      "acc": 0.61460381,
      "epoch": 0.0441400304414003,
      "grad_norm": 8.875,
      "learning_rate": 4.414003044140031e-06,
      "loss": 1.88086433,
      "memory(GiB)": 55.86,
      "step": 1740,
      "train_speed(iter/s)": 1.642955
    },
    {
      "acc": 0.61463203,
      "epoch": 0.04426686960933536,
      "grad_norm": 6.40625,
      "learning_rate": 4.426686960933537e-06,
      "loss": 1.896945,
      "memory(GiB)": 55.86,
      "step": 1745,
      "train_speed(iter/s)": 1.643396
    },
    {
      "acc": 0.59492693,
      "epoch": 0.04439370877727042,
      "grad_norm": 6.28125,
      "learning_rate": 4.439370877727043e-06,
      "loss": 1.99612389,
      "memory(GiB)": 55.86,
      "step": 1750,
      "train_speed(iter/s)": 1.643863
    },
    {
      "acc": 0.60498633,
      "epoch": 0.04452054794520548,
      "grad_norm": 6.84375,
      "learning_rate": 4.4520547945205486e-06,
      "loss": 1.92270393,
      "memory(GiB)": 55.86,
      "step": 1755,
      "train_speed(iter/s)": 1.644334
    },
    {
      "acc": 0.62942419,
      "epoch": 0.044647387113140535,
      "grad_norm": 5.96875,
      "learning_rate": 4.464738711314054e-06,
      "loss": 1.85165215,
      "memory(GiB)": 55.86,
      "step": 1760,
      "train_speed(iter/s)": 1.644771
    },
    {
      "acc": 0.60646572,
      "epoch": 0.044774226281075594,
      "grad_norm": 6.0,
      "learning_rate": 4.47742262810756e-06,
      "loss": 1.96096783,
      "memory(GiB)": 55.86,
      "step": 1765,
      "train_speed(iter/s)": 1.64526
    },
    {
      "acc": 0.61349754,
      "epoch": 0.04490106544901065,
      "grad_norm": 5.4375,
      "learning_rate": 4.490106544901066e-06,
      "loss": 1.86078606,
      "memory(GiB)": 55.86,
      "step": 1770,
      "train_speed(iter/s)": 1.645699
    },
    {
      "acc": 0.61186399,
      "epoch": 0.04502790461694571,
      "grad_norm": 7.625,
      "learning_rate": 4.502790461694572e-06,
      "loss": 1.96916714,
      "memory(GiB)": 55.86,
      "step": 1775,
      "train_speed(iter/s)": 1.64612
    },
    {
      "acc": 0.60810881,
      "epoch": 0.04515474378488077,
      "grad_norm": 6.5625,
      "learning_rate": 4.5154743784880776e-06,
      "loss": 1.99163055,
      "memory(GiB)": 55.86,
      "step": 1780,
      "train_speed(iter/s)": 1.646604
    },
    {
      "acc": 0.61640735,
      "epoch": 0.04528158295281583,
      "grad_norm": 5.28125,
      "learning_rate": 4.5281582952815835e-06,
      "loss": 1.90900517,
      "memory(GiB)": 55.86,
      "step": 1785,
      "train_speed(iter/s)": 1.647036
    },
    {
      "acc": 0.61566954,
      "epoch": 0.045408422120750885,
      "grad_norm": 6.21875,
      "learning_rate": 4.540842212075089e-06,
      "loss": 1.86947441,
      "memory(GiB)": 55.86,
      "step": 1790,
      "train_speed(iter/s)": 1.647483
    },
    {
      "acc": 0.61108809,
      "epoch": 0.045535261288685944,
      "grad_norm": 6.65625,
      "learning_rate": 4.553526128868595e-06,
      "loss": 1.94466934,
      "memory(GiB)": 55.86,
      "step": 1795,
      "train_speed(iter/s)": 1.647932
    },
    {
      "acc": 0.60261269,
      "epoch": 0.045662100456621,
      "grad_norm": 4.65625,
      "learning_rate": 4.566210045662101e-06,
      "loss": 1.94084854,
      "memory(GiB)": 55.86,
      "step": 1800,
      "train_speed(iter/s)": 1.648346
    },
    {
      "acc": 0.63324766,
      "epoch": 0.04578893962455606,
      "grad_norm": 6.375,
      "learning_rate": 4.5788939624556065e-06,
      "loss": 1.78107758,
      "memory(GiB)": 55.86,
      "step": 1805,
      "train_speed(iter/s)": 1.648777
    },
    {
      "acc": 0.60749745,
      "epoch": 0.04591577879249112,
      "grad_norm": 5.3125,
      "learning_rate": 4.5915778792491125e-06,
      "loss": 1.91480827,
      "memory(GiB)": 55.86,
      "step": 1810,
      "train_speed(iter/s)": 1.649204
    },
    {
      "acc": 0.58367977,
      "epoch": 0.04604261796042618,
      "grad_norm": 8.3125,
      "learning_rate": 4.6042617960426185e-06,
      "loss": 2.02885132,
      "memory(GiB)": 55.86,
      "step": 1815,
      "train_speed(iter/s)": 1.649647
    },
    {
      "acc": 0.60988832,
      "epoch": 0.046169457128361235,
      "grad_norm": 7.0,
      "learning_rate": 4.616945712836124e-06,
      "loss": 1.90552158,
      "memory(GiB)": 55.86,
      "step": 1820,
      "train_speed(iter/s)": 1.650116
    },
    {
      "acc": 0.6232265,
      "epoch": 0.046296296296296294,
      "grad_norm": 7.5,
      "learning_rate": 4.62962962962963e-06,
      "loss": 1.86238213,
      "memory(GiB)": 55.86,
      "step": 1825,
      "train_speed(iter/s)": 1.650541
    },
    {
      "acc": 0.6233623,
      "epoch": 0.04642313546423135,
      "grad_norm": 7.125,
      "learning_rate": 4.6423135464231355e-06,
      "loss": 1.95918674,
      "memory(GiB)": 55.86,
      "step": 1830,
      "train_speed(iter/s)": 1.650936
    },
    {
      "acc": 0.63595634,
      "epoch": 0.04654997463216641,
      "grad_norm": 8.6875,
      "learning_rate": 4.6549974632166415e-06,
      "loss": 1.90061665,
      "memory(GiB)": 55.86,
      "step": 1835,
      "train_speed(iter/s)": 1.651376
    },
    {
      "acc": 0.63078909,
      "epoch": 0.04667681380010147,
      "grad_norm": 5.75,
      "learning_rate": 4.6676813800101475e-06,
      "loss": 1.84006805,
      "memory(GiB)": 55.86,
      "step": 1840,
      "train_speed(iter/s)": 1.651804
    },
    {
      "acc": 0.60643654,
      "epoch": 0.04680365296803653,
      "grad_norm": 6.3125,
      "learning_rate": 4.6803652968036534e-06,
      "loss": 1.89833241,
      "memory(GiB)": 55.86,
      "step": 1845,
      "train_speed(iter/s)": 1.65222
    },
    {
      "acc": 0.62370415,
      "epoch": 0.046930492135971585,
      "grad_norm": 6.3125,
      "learning_rate": 4.6930492135971586e-06,
      "loss": 1.85581799,
      "memory(GiB)": 55.86,
      "step": 1850,
      "train_speed(iter/s)": 1.652628
    },
    {
      "acc": 0.60888762,
      "epoch": 0.047057331303906644,
      "grad_norm": 7.03125,
      "learning_rate": 4.705733130390665e-06,
      "loss": 1.87097797,
      "memory(GiB)": 55.86,
      "step": 1855,
      "train_speed(iter/s)": 1.653063
    },
    {
      "acc": 0.59038219,
      "epoch": 0.0471841704718417,
      "grad_norm": 5.25,
      "learning_rate": 4.7184170471841705e-06,
      "loss": 1.92111855,
      "memory(GiB)": 55.86,
      "step": 1860,
      "train_speed(iter/s)": 1.653456
    },
    {
      "acc": 0.60695114,
      "epoch": 0.04731100963977676,
      "grad_norm": 6.28125,
      "learning_rate": 4.7311009639776765e-06,
      "loss": 1.93966103,
      "memory(GiB)": 55.86,
      "step": 1865,
      "train_speed(iter/s)": 1.653841
    },
    {
      "acc": 0.59265776,
      "epoch": 0.04743784880771182,
      "grad_norm": 6.5625,
      "learning_rate": 4.743784880771182e-06,
      "loss": 1.93526154,
      "memory(GiB)": 55.86,
      "step": 1870,
      "train_speed(iter/s)": 1.654252
    },
    {
      "acc": 0.60292034,
      "epoch": 0.04756468797564688,
      "grad_norm": 6.25,
      "learning_rate": 4.756468797564688e-06,
      "loss": 1.88684444,
      "memory(GiB)": 55.86,
      "step": 1875,
      "train_speed(iter/s)": 1.654652
    },
    {
      "acc": 0.62602434,
      "epoch": 0.047691527143581935,
      "grad_norm": 6.125,
      "learning_rate": 4.769152714358194e-06,
      "loss": 1.94059296,
      "memory(GiB)": 55.86,
      "step": 1880,
      "train_speed(iter/s)": 1.65503
    },
    {
      "acc": 0.62489433,
      "epoch": 0.047818366311516994,
      "grad_norm": 5.125,
      "learning_rate": 4.7818366311517e-06,
      "loss": 1.88831844,
      "memory(GiB)": 55.86,
      "step": 1885,
      "train_speed(iter/s)": 1.655445
    },
    {
      "acc": 0.61385436,
      "epoch": 0.04794520547945205,
      "grad_norm": 5.84375,
      "learning_rate": 4.7945205479452054e-06,
      "loss": 1.90837364,
      "memory(GiB)": 55.86,
      "step": 1890,
      "train_speed(iter/s)": 1.655874
    },
    {
      "acc": 0.62878113,
      "epoch": 0.04807204464738711,
      "grad_norm": 7.0,
      "learning_rate": 4.807204464738711e-06,
      "loss": 1.8503315,
      "memory(GiB)": 55.86,
      "step": 1895,
      "train_speed(iter/s)": 1.656288
    },
    {
      "acc": 0.61581697,
      "epoch": 0.04819888381532217,
      "grad_norm": 10.75,
      "learning_rate": 4.819888381532217e-06,
      "loss": 1.87218151,
      "memory(GiB)": 55.86,
      "step": 1900,
      "train_speed(iter/s)": 1.656681
    },
    {
      "acc": 0.61975689,
      "epoch": 0.04832572298325723,
      "grad_norm": 7.8125,
      "learning_rate": 4.832572298325723e-06,
      "loss": 1.8513237,
      "memory(GiB)": 55.86,
      "step": 1905,
      "train_speed(iter/s)": 1.657069
    },
    {
      "acc": 0.61313229,
      "epoch": 0.048452562151192285,
      "grad_norm": 7.46875,
      "learning_rate": 4.845256215119229e-06,
      "loss": 1.85294476,
      "memory(GiB)": 55.86,
      "step": 1910,
      "train_speed(iter/s)": 1.65742
    },
    {
      "acc": 0.61485438,
      "epoch": 0.048579401319127344,
      "grad_norm": 5.4375,
      "learning_rate": 4.857940131912735e-06,
      "loss": 1.87055016,
      "memory(GiB)": 55.86,
      "step": 1915,
      "train_speed(iter/s)": 1.65782
    },
    {
      "acc": 0.62736216,
      "epoch": 0.0487062404870624,
      "grad_norm": 6.0625,
      "learning_rate": 4.87062404870624e-06,
      "loss": 1.77576618,
      "memory(GiB)": 55.86,
      "step": 1920,
      "train_speed(iter/s)": 1.658199
    },
    {
      "acc": 0.60139713,
      "epoch": 0.04883307965499746,
      "grad_norm": 6.03125,
      "learning_rate": 4.883307965499746e-06,
      "loss": 1.91723747,
      "memory(GiB)": 55.86,
      "step": 1925,
      "train_speed(iter/s)": 1.658582
    },
    {
      "acc": 0.59416075,
      "epoch": 0.04895991882293252,
      "grad_norm": 6.40625,
      "learning_rate": 4.895991882293252e-06,
      "loss": 2.01217003,
      "memory(GiB)": 55.86,
      "step": 1930,
      "train_speed(iter/s)": 1.658987
    },
    {
      "acc": 0.61094227,
      "epoch": 0.04908675799086758,
      "grad_norm": 6.53125,
      "learning_rate": 4.908675799086758e-06,
      "loss": 1.86329765,
      "memory(GiB)": 55.86,
      "step": 1935,
      "train_speed(iter/s)": 1.659354
    },
    {
      "acc": 0.61396599,
      "epoch": 0.049213597158802636,
      "grad_norm": 6.46875,
      "learning_rate": 4.921359715880264e-06,
      "loss": 1.9199873,
      "memory(GiB)": 55.86,
      "step": 1940,
      "train_speed(iter/s)": 1.65971
    },
    {
      "acc": 0.61482,
      "epoch": 0.049340436326737694,
      "grad_norm": 5.71875,
      "learning_rate": 4.93404363267377e-06,
      "loss": 1.88470364,
      "memory(GiB)": 55.86,
      "step": 1945,
      "train_speed(iter/s)": 1.660082
    },
    {
      "acc": 0.63160276,
      "epoch": 0.04946727549467275,
      "grad_norm": 5.25,
      "learning_rate": 4.946727549467275e-06,
      "loss": 1.85397835,
      "memory(GiB)": 55.86,
      "step": 1950,
      "train_speed(iter/s)": 1.660474
    },
    {
      "acc": 0.61834173,
      "epoch": 0.04959411466260781,
      "grad_norm": 5.75,
      "learning_rate": 4.959411466260781e-06,
      "loss": 1.80206661,
      "memory(GiB)": 55.86,
      "step": 1955,
      "train_speed(iter/s)": 1.66085
    },
    {
      "acc": 0.61714339,
      "epoch": 0.04972095383054287,
      "grad_norm": 7.34375,
      "learning_rate": 4.972095383054287e-06,
      "loss": 1.91153755,
      "memory(GiB)": 55.86,
      "step": 1960,
      "train_speed(iter/s)": 1.661242
    },
    {
      "acc": 0.61616869,
      "epoch": 0.04984779299847793,
      "grad_norm": 6.1875,
      "learning_rate": 4.984779299847793e-06,
      "loss": 1.87368698,
      "memory(GiB)": 55.86,
      "step": 1965,
      "train_speed(iter/s)": 1.661633
    },
    {
      "acc": 0.61846704,
      "epoch": 0.049974632166412986,
      "grad_norm": 8.0625,
      "learning_rate": 4.997463216641299e-06,
      "loss": 1.90532875,
      "memory(GiB)": 55.86,
      "step": 1970,
      "train_speed(iter/s)": 1.662005
    },
    {
      "acc": 0.62661562,
      "epoch": 0.050101471334348044,
      "grad_norm": 6.46875,
      "learning_rate": 5.010147133434805e-06,
      "loss": 1.84291344,
      "memory(GiB)": 55.86,
      "step": 1975,
      "train_speed(iter/s)": 1.66236
    },
    {
      "acc": 0.62849011,
      "epoch": 0.0502283105022831,
      "grad_norm": 8.1875,
      "learning_rate": 5.02283105022831e-06,
      "loss": 1.80382233,
      "memory(GiB)": 55.86,
      "step": 1980,
      "train_speed(iter/s)": 1.662726
    },
    {
      "acc": 0.59887094,
      "epoch": 0.05035514967021816,
      "grad_norm": 7.28125,
      "learning_rate": 5.035514967021817e-06,
      "loss": 1.96077232,
      "memory(GiB)": 55.86,
      "step": 1985,
      "train_speed(iter/s)": 1.663082
    },
    {
      "acc": 0.61298971,
      "epoch": 0.05048198883815322,
      "grad_norm": 6.125,
      "learning_rate": 5.048198883815323e-06,
      "loss": 1.87825432,
      "memory(GiB)": 55.86,
      "step": 1990,
      "train_speed(iter/s)": 1.663439
    },
    {
      "acc": 0.59730964,
      "epoch": 0.05060882800608828,
      "grad_norm": 6.15625,
      "learning_rate": 5.060882800608828e-06,
      "loss": 1.9565403,
      "memory(GiB)": 55.86,
      "step": 1995,
      "train_speed(iter/s)": 1.663826
    },
    {
      "acc": 0.61201725,
      "epoch": 0.050735667174023336,
      "grad_norm": 6.8125,
      "learning_rate": 5.073566717402334e-06,
      "loss": 1.85511951,
      "memory(GiB)": 55.86,
      "step": 2000,
      "train_speed(iter/s)": 1.664195
    },
    {
      "epoch": 0.050735667174023336,
      "eval_acc": 0.601371181793312,
      "eval_loss": 1.8402973413467407,
      "eval_runtime": 58.5184,
      "eval_samples_per_second": 108.855,
      "eval_steps_per_second": 27.222,
      "step": 2000
    },
    {
      "acc": 0.60759439,
      "epoch": 0.050862506341958394,
      "grad_norm": 7.03125,
      "learning_rate": 5.086250634195841e-06,
      "loss": 1.88536301,
      "memory(GiB)": 55.86,
      "step": 2005,
      "train_speed(iter/s)": 1.582812
    },
    {
      "acc": 0.60362034,
      "epoch": 0.05098934550989345,
      "grad_norm": 6.0625,
      "learning_rate": 5.098934550989346e-06,
      "loss": 1.85925827,
      "memory(GiB)": 55.86,
      "step": 2010,
      "train_speed(iter/s)": 1.583373
    },
    {
      "acc": 0.62760158,
      "epoch": 0.05111618467782851,
      "grad_norm": 8.375,
      "learning_rate": 5.111618467782852e-06,
      "loss": 1.84095268,
      "memory(GiB)": 55.86,
      "step": 2015,
      "train_speed(iter/s)": 1.583881
    },
    {
      "acc": 0.62201629,
      "epoch": 0.05124302384576357,
      "grad_norm": 5.53125,
      "learning_rate": 5.124302384576357e-06,
      "loss": 1.81402969,
      "memory(GiB)": 55.86,
      "step": 2020,
      "train_speed(iter/s)": 1.584416
    },
    {
      "acc": 0.62154894,
      "epoch": 0.05136986301369863,
      "grad_norm": 5.59375,
      "learning_rate": 5.136986301369864e-06,
      "loss": 1.83564453,
      "memory(GiB)": 55.86,
      "step": 2025,
      "train_speed(iter/s)": 1.5849
    },
    {
      "acc": 0.62282491,
      "epoch": 0.051496702181633686,
      "grad_norm": 6.71875,
      "learning_rate": 5.149670218163369e-06,
      "loss": 1.83527508,
      "memory(GiB)": 55.86,
      "step": 2030,
      "train_speed(iter/s)": 1.585429
    },
    {
      "acc": 0.62803555,
      "epoch": 0.051623541349568744,
      "grad_norm": 7.875,
      "learning_rate": 5.162354134956875e-06,
      "loss": 1.87986412,
      "memory(GiB)": 55.86,
      "step": 2035,
      "train_speed(iter/s)": 1.585932
    },
    {
      "acc": 0.62820811,
      "epoch": 0.0517503805175038,
      "grad_norm": 6.28125,
      "learning_rate": 5.175038051750381e-06,
      "loss": 1.85619965,
      "memory(GiB)": 55.86,
      "step": 2040,
      "train_speed(iter/s)": 1.586465
    },
    {
      "acc": 0.62108946,
      "epoch": 0.05187721968543886,
      "grad_norm": 10.5,
      "learning_rate": 5.187721968543887e-06,
      "loss": 1.90872402,
      "memory(GiB)": 55.86,
      "step": 2045,
      "train_speed(iter/s)": 1.586983
    },
    {
      "acc": 0.61195669,
      "epoch": 0.05200405885337392,
      "grad_norm": 6.59375,
      "learning_rate": 5.200405885337393e-06,
      "loss": 1.82115593,
      "memory(GiB)": 58.86,
      "step": 2050,
      "train_speed(iter/s)": 1.587455
    },
    {
      "acc": 0.62076006,
      "epoch": 0.05213089802130898,
      "grad_norm": 6.8125,
      "learning_rate": 5.213089802130898e-06,
      "loss": 1.85543823,
      "memory(GiB)": 58.86,
      "step": 2055,
      "train_speed(iter/s)": 1.587975
    },
    {
      "acc": 0.6161839,
      "epoch": 0.052257737189244036,
      "grad_norm": 5.65625,
      "learning_rate": 5.225773718924404e-06,
      "loss": 1.8410347,
      "memory(GiB)": 58.86,
      "step": 2060,
      "train_speed(iter/s)": 1.588462
    },
    {
      "acc": 0.58664618,
      "epoch": 0.052384576357179094,
      "grad_norm": 6.8125,
      "learning_rate": 5.238457635717911e-06,
      "loss": 2.01187592,
      "memory(GiB)": 58.86,
      "step": 2065,
      "train_speed(iter/s)": 1.588928
    },
    {
      "acc": 0.62029772,
      "epoch": 0.05251141552511415,
      "grad_norm": 7.125,
      "learning_rate": 5.251141552511416e-06,
      "loss": 1.8977087,
      "memory(GiB)": 58.86,
      "step": 2070,
      "train_speed(iter/s)": 1.589409
    },
    {
      "acc": 0.6000257,
      "epoch": 0.05263825469304921,
      "grad_norm": 5.5625,
      "learning_rate": 5.263825469304922e-06,
      "loss": 1.95394592,
      "memory(GiB)": 58.86,
      "step": 2075,
      "train_speed(iter/s)": 1.589895
    },
    {
      "acc": 0.59658971,
      "epoch": 0.05276509386098427,
      "grad_norm": 6.28125,
      "learning_rate": 5.276509386098427e-06,
      "loss": 1.94618568,
      "memory(GiB)": 58.86,
      "step": 2080,
      "train_speed(iter/s)": 1.5904
    },
    {
      "acc": 0.61407924,
      "epoch": 0.05289193302891933,
      "grad_norm": 6.75,
      "learning_rate": 5.289193302891934e-06,
      "loss": 1.87246552,
      "memory(GiB)": 58.86,
      "step": 2085,
      "train_speed(iter/s)": 1.590786
    },
    {
      "acc": 0.62276363,
      "epoch": 0.053018772196854386,
      "grad_norm": 5.5625,
      "learning_rate": 5.30187721968544e-06,
      "loss": 1.86005402,
      "memory(GiB)": 58.86,
      "step": 2090,
      "train_speed(iter/s)": 1.591286
    },
    {
      "acc": 0.6083406,
      "epoch": 0.053145611364789444,
      "grad_norm": 5.65625,
      "learning_rate": 5.314561136478945e-06,
      "loss": 1.90465031,
      "memory(GiB)": 58.86,
      "step": 2095,
      "train_speed(iter/s)": 1.59175
    },
    {
      "acc": 0.61667757,
      "epoch": 0.0532724505327245,
      "grad_norm": 6.21875,
      "learning_rate": 5.327245053272451e-06,
      "loss": 1.90533409,
      "memory(GiB)": 58.86,
      "step": 2100,
      "train_speed(iter/s)": 1.592255
    },
    {
      "acc": 0.64405117,
      "epoch": 0.05339928970065956,
      "grad_norm": 6.09375,
      "learning_rate": 5.339928970065957e-06,
      "loss": 1.75282745,
      "memory(GiB)": 58.86,
      "step": 2105,
      "train_speed(iter/s)": 1.592748
    },
    {
      "acc": 0.61783571,
      "epoch": 0.05352612886859462,
      "grad_norm": 7.75,
      "learning_rate": 5.352612886859463e-06,
      "loss": 1.91949978,
      "memory(GiB)": 58.86,
      "step": 2110,
      "train_speed(iter/s)": 1.59325
    },
    {
      "acc": 0.6170114,
      "epoch": 0.05365296803652968,
      "grad_norm": 7.09375,
      "learning_rate": 5.365296803652969e-06,
      "loss": 1.85956173,
      "memory(GiB)": 58.86,
      "step": 2115,
      "train_speed(iter/s)": 1.593743
    },
    {
      "acc": 0.61532836,
      "epoch": 0.053779807204464736,
      "grad_norm": 7.71875,
      "learning_rate": 5.377980720446474e-06,
      "loss": 1.81922073,
      "memory(GiB)": 58.86,
      "step": 2120,
      "train_speed(iter/s)": 1.594196
    },
    {
      "acc": 0.60407591,
      "epoch": 0.053906646372399794,
      "grad_norm": 6.8125,
      "learning_rate": 5.390664637239981e-06,
      "loss": 1.92536125,
      "memory(GiB)": 58.86,
      "step": 2125,
      "train_speed(iter/s)": 1.594668
    },
    {
      "acc": 0.61450729,
      "epoch": 0.05403348554033485,
      "grad_norm": 6.8125,
      "learning_rate": 5.403348554033486e-06,
      "loss": 1.8827652,
      "memory(GiB)": 58.86,
      "step": 2130,
      "train_speed(iter/s)": 1.59517
    },
    {
      "acc": 0.62003036,
      "epoch": 0.05416032470826991,
      "grad_norm": 6.09375,
      "learning_rate": 5.416032470826992e-06,
      "loss": 1.81562634,
      "memory(GiB)": 58.86,
      "step": 2135,
      "train_speed(iter/s)": 1.595655
    },
    {
      "acc": 0.62902613,
      "epoch": 0.05428716387620497,
      "grad_norm": 6.65625,
      "learning_rate": 5.428716387620497e-06,
      "loss": 1.82815914,
      "memory(GiB)": 58.86,
      "step": 2140,
      "train_speed(iter/s)": 1.596142
    },
    {
      "acc": 0.61863089,
      "epoch": 0.05441400304414003,
      "grad_norm": 6.21875,
      "learning_rate": 5.441400304414004e-06,
      "loss": 1.85329399,
      "memory(GiB)": 58.86,
      "step": 2145,
      "train_speed(iter/s)": 1.596633
    },
    {
      "acc": 0.61467247,
      "epoch": 0.054540842212075086,
      "grad_norm": 5.25,
      "learning_rate": 5.45408422120751e-06,
      "loss": 1.88357754,
      "memory(GiB)": 58.86,
      "step": 2150,
      "train_speed(iter/s)": 1.597122
    },
    {
      "acc": 0.61463175,
      "epoch": 0.054667681380010144,
      "grad_norm": 5.90625,
      "learning_rate": 5.466768138001015e-06,
      "loss": 1.94149857,
      "memory(GiB)": 58.86,
      "step": 2155,
      "train_speed(iter/s)": 1.5976
    },
    {
      "acc": 0.62385445,
      "epoch": 0.0547945205479452,
      "grad_norm": 5.28125,
      "learning_rate": 5.479452054794521e-06,
      "loss": 1.93745651,
      "memory(GiB)": 58.86,
      "step": 2160,
      "train_speed(iter/s)": 1.598066
    },
    {
      "acc": 0.60461903,
      "epoch": 0.05492135971588026,
      "grad_norm": 7.28125,
      "learning_rate": 5.492135971588028e-06,
      "loss": 1.86028633,
      "memory(GiB)": 58.86,
      "step": 2165,
      "train_speed(iter/s)": 1.59853
    },
    {
      "acc": 0.592383,
      "epoch": 0.05504819888381532,
      "grad_norm": 5.46875,
      "learning_rate": 5.504819888381533e-06,
      "loss": 1.97254333,
      "memory(GiB)": 58.86,
      "step": 2170,
      "train_speed(iter/s)": 1.598994
    },
    {
      "acc": 0.62739601,
      "epoch": 0.05517503805175038,
      "grad_norm": 5.53125,
      "learning_rate": 5.517503805175039e-06,
      "loss": 1.83993111,
      "memory(GiB)": 58.86,
      "step": 2175,
      "train_speed(iter/s)": 1.599423
    },
    {
      "acc": 0.61081505,
      "epoch": 0.055301877219685436,
      "grad_norm": 6.375,
      "learning_rate": 5.530187721968544e-06,
      "loss": 1.87527924,
      "memory(GiB)": 58.86,
      "step": 2180,
      "train_speed(iter/s)": 1.599862
    },
    {
      "acc": 0.6117691,
      "epoch": 0.055428716387620494,
      "grad_norm": 7.90625,
      "learning_rate": 5.542871638762051e-06,
      "loss": 1.89598579,
      "memory(GiB)": 58.86,
      "step": 2185,
      "train_speed(iter/s)": 1.600323
    },
    {
      "acc": 0.61759844,
      "epoch": 0.05555555555555555,
      "grad_norm": 6.46875,
      "learning_rate": 5.555555555555557e-06,
      "loss": 1.9203064,
      "memory(GiB)": 58.86,
      "step": 2190,
      "train_speed(iter/s)": 1.600771
    },
    {
      "acc": 0.62582135,
      "epoch": 0.05568239472349061,
      "grad_norm": 4.59375,
      "learning_rate": 5.568239472349062e-06,
      "loss": 1.8298502,
      "memory(GiB)": 58.86,
      "step": 2195,
      "train_speed(iter/s)": 1.601227
    },
    {
      "acc": 0.61843352,
      "epoch": 0.05580923389142567,
      "grad_norm": 8.125,
      "learning_rate": 5.580923389142568e-06,
      "loss": 1.88783493,
      "memory(GiB)": 58.86,
      "step": 2200,
      "train_speed(iter/s)": 1.601684
    },
    {
      "acc": 0.61050396,
      "epoch": 0.05593607305936073,
      "grad_norm": 6.59375,
      "learning_rate": 5.593607305936074e-06,
      "loss": 1.8683567,
      "memory(GiB)": 58.86,
      "step": 2205,
      "train_speed(iter/s)": 1.602111
    },
    {
      "acc": 0.61746693,
      "epoch": 0.056062912227295786,
      "grad_norm": 6.3125,
      "learning_rate": 5.60629122272958e-06,
      "loss": 1.84496689,
      "memory(GiB)": 58.86,
      "step": 2210,
      "train_speed(iter/s)": 1.602564
    },
    {
      "acc": 0.61757865,
      "epoch": 0.056189751395230844,
      "grad_norm": 6.71875,
      "learning_rate": 5.618975139523085e-06,
      "loss": 1.88137474,
      "memory(GiB)": 58.86,
      "step": 2215,
      "train_speed(iter/s)": 1.603012
    },
    {
      "acc": 0.62015538,
      "epoch": 0.0563165905631659,
      "grad_norm": 6.1875,
      "learning_rate": 5.631659056316591e-06,
      "loss": 1.81494751,
      "memory(GiB)": 58.86,
      "step": 2220,
      "train_speed(iter/s)": 1.603463
    },
    {
      "acc": 0.62243133,
      "epoch": 0.05644342973110096,
      "grad_norm": 6.5,
      "learning_rate": 5.644342973110098e-06,
      "loss": 1.85718117,
      "memory(GiB)": 58.86,
      "step": 2225,
      "train_speed(iter/s)": 1.60388
    },
    {
      "acc": 0.62325392,
      "epoch": 0.05657026889903602,
      "grad_norm": 7.3125,
      "learning_rate": 5.657026889903603e-06,
      "loss": 1.82916603,
      "memory(GiB)": 58.86,
      "step": 2230,
      "train_speed(iter/s)": 1.604302
    },
    {
      "acc": 0.60833759,
      "epoch": 0.05669710806697108,
      "grad_norm": 5.375,
      "learning_rate": 5.669710806697109e-06,
      "loss": 1.93776398,
      "memory(GiB)": 58.86,
      "step": 2235,
      "train_speed(iter/s)": 1.604749
    },
    {
      "acc": 0.61601906,
      "epoch": 0.056823947234906136,
      "grad_norm": 5.5625,
      "learning_rate": 5.682394723490614e-06,
      "loss": 1.86586227,
      "memory(GiB)": 58.86,
      "step": 2240,
      "train_speed(iter/s)": 1.605201
    },
    {
      "acc": 0.60844631,
      "epoch": 0.056950786402841194,
      "grad_norm": 5.8125,
      "learning_rate": 5.695078640284121e-06,
      "loss": 1.89643288,
      "memory(GiB)": 58.86,
      "step": 2245,
      "train_speed(iter/s)": 1.605632
    },
    {
      "acc": 0.61506748,
      "epoch": 0.05707762557077625,
      "grad_norm": 6.21875,
      "learning_rate": 5.7077625570776266e-06,
      "loss": 1.84293365,
      "memory(GiB)": 58.86,
      "step": 2250,
      "train_speed(iter/s)": 1.606108
    },
    {
      "acc": 0.61220007,
      "epoch": 0.05720446473871131,
      "grad_norm": 5.40625,
      "learning_rate": 5.720446473871132e-06,
      "loss": 1.8711916,
      "memory(GiB)": 58.86,
      "step": 2255,
      "train_speed(iter/s)": 1.60655
    },
    {
      "acc": 0.62895617,
      "epoch": 0.05733130390664637,
      "grad_norm": 9.3125,
      "learning_rate": 5.733130390664638e-06,
      "loss": 1.77090511,
      "memory(GiB)": 58.86,
      "step": 2260,
      "train_speed(iter/s)": 1.607017
    },
    {
      "acc": 0.611234,
      "epoch": 0.05745814307458143,
      "grad_norm": 6.59375,
      "learning_rate": 5.7458143074581445e-06,
      "loss": 1.87374725,
      "memory(GiB)": 58.86,
      "step": 2265,
      "train_speed(iter/s)": 1.607444
    },
    {
      "acc": 0.60035524,
      "epoch": 0.057584982242516486,
      "grad_norm": 6.25,
      "learning_rate": 5.75849822425165e-06,
      "loss": 1.92587185,
      "memory(GiB)": 58.86,
      "step": 2270,
      "train_speed(iter/s)": 1.607901
    },
    {
      "acc": 0.60545778,
      "epoch": 0.057711821410451544,
      "grad_norm": 6.625,
      "learning_rate": 5.7711821410451556e-06,
      "loss": 1.93162308,
      "memory(GiB)": 58.86,
      "step": 2275,
      "train_speed(iter/s)": 1.608325
    },
    {
      "acc": 0.61990809,
      "epoch": 0.0578386605783866,
      "grad_norm": 6.0625,
      "learning_rate": 5.783866057838661e-06,
      "loss": 1.83121796,
      "memory(GiB)": 58.86,
      "step": 2280,
      "train_speed(iter/s)": 1.608759
    },
    {
      "acc": 0.59965029,
      "epoch": 0.05796549974632166,
      "grad_norm": 6.375,
      "learning_rate": 5.7965499746321675e-06,
      "loss": 1.96522408,
      "memory(GiB)": 58.86,
      "step": 2285,
      "train_speed(iter/s)": 1.609193
    },
    {
      "acc": 0.61757164,
      "epoch": 0.05809233891425672,
      "grad_norm": 7.0625,
      "learning_rate": 5.809233891425673e-06,
      "loss": 1.86750145,
      "memory(GiB)": 58.86,
      "step": 2290,
      "train_speed(iter/s)": 1.60959
    },
    {
      "acc": 0.61162195,
      "epoch": 0.05821917808219178,
      "grad_norm": 6.25,
      "learning_rate": 5.821917808219179e-06,
      "loss": 1.86003838,
      "memory(GiB)": 58.86,
      "step": 2295,
      "train_speed(iter/s)": 1.61003
    },
    {
      "acc": 0.61908388,
      "epoch": 0.058346017250126836,
      "grad_norm": 7.1875,
      "learning_rate": 5.834601725012684e-06,
      "loss": 1.84045887,
      "memory(GiB)": 58.86,
      "step": 2300,
      "train_speed(iter/s)": 1.610449
    },
    {
      "acc": 0.61458035,
      "epoch": 0.058472856418061894,
      "grad_norm": 7.65625,
      "learning_rate": 5.8472856418061905e-06,
      "loss": 1.88287506,
      "memory(GiB)": 58.86,
      "step": 2305,
      "train_speed(iter/s)": 1.610903
    },
    {
      "acc": 0.61573048,
      "epoch": 0.05859969558599695,
      "grad_norm": 6.1875,
      "learning_rate": 5.8599695585996965e-06,
      "loss": 1.84793434,
      "memory(GiB)": 58.86,
      "step": 2310,
      "train_speed(iter/s)": 1.61134
    },
    {
      "acc": 0.61905918,
      "epoch": 0.05872653475393201,
      "grad_norm": 5.28125,
      "learning_rate": 5.872653475393202e-06,
      "loss": 1.80390949,
      "memory(GiB)": 58.86,
      "step": 2315,
      "train_speed(iter/s)": 1.611776
    },
    {
      "acc": 0.62466311,
      "epoch": 0.05885337392186707,
      "grad_norm": 6.0,
      "learning_rate": 5.8853373921867076e-06,
      "loss": 1.91816978,
      "memory(GiB)": 58.86,
      "step": 2320,
      "train_speed(iter/s)": 1.612206
    },
    {
      "acc": 0.63113375,
      "epoch": 0.05898021308980213,
      "grad_norm": 6.3125,
      "learning_rate": 5.898021308980214e-06,
      "loss": 1.79769382,
      "memory(GiB)": 58.86,
      "step": 2325,
      "train_speed(iter/s)": 1.612635
    },
    {
      "acc": 0.60280986,
      "epoch": 0.059107052257737186,
      "grad_norm": 6.34375,
      "learning_rate": 5.9107052257737195e-06,
      "loss": 1.98647652,
      "memory(GiB)": 58.86,
      "step": 2330,
      "train_speed(iter/s)": 1.61304
    },
    {
      "acc": 0.60971193,
      "epoch": 0.059233891425672244,
      "grad_norm": 5.46875,
      "learning_rate": 5.9233891425672255e-06,
      "loss": 1.92672958,
      "memory(GiB)": 58.86,
      "step": 2335,
      "train_speed(iter/s)": 1.613439
    },
    {
      "acc": 0.62279806,
      "epoch": 0.0593607305936073,
      "grad_norm": 5.6875,
      "learning_rate": 5.936073059360731e-06,
      "loss": 1.81108055,
      "memory(GiB)": 58.86,
      "step": 2340,
      "train_speed(iter/s)": 1.613866
    },
    {
      "acc": 0.61838765,
      "epoch": 0.05948756976154236,
      "grad_norm": 6.15625,
      "learning_rate": 5.948756976154237e-06,
      "loss": 1.8841713,
      "memory(GiB)": 58.86,
      "step": 2345,
      "train_speed(iter/s)": 1.614266
    },
    {
      "acc": 0.62070599,
      "epoch": 0.05961440892947742,
      "grad_norm": 6.34375,
      "learning_rate": 5.961440892947743e-06,
      "loss": 1.87567253,
      "memory(GiB)": 58.86,
      "step": 2350,
      "train_speed(iter/s)": 1.614679
    },
    {
      "acc": 0.63126111,
      "epoch": 0.05974124809741248,
      "grad_norm": 7.25,
      "learning_rate": 5.9741248097412485e-06,
      "loss": 1.7787096,
      "memory(GiB)": 58.86,
      "step": 2355,
      "train_speed(iter/s)": 1.615087
    },
    {
      "acc": 0.61884856,
      "epoch": 0.059868087265347536,
      "grad_norm": 5.75,
      "learning_rate": 5.9868087265347545e-06,
      "loss": 1.83313141,
      "memory(GiB)": 58.86,
      "step": 2360,
      "train_speed(iter/s)": 1.615467
    },
    {
      "acc": 0.61990161,
      "epoch": 0.059994926433282594,
      "grad_norm": 5.96875,
      "learning_rate": 5.99949264332826e-06,
      "loss": 1.80100517,
      "memory(GiB)": 58.86,
      "step": 2365,
      "train_speed(iter/s)": 1.615866
    },
    {
      "acc": 0.61794252,
      "epoch": 0.06012176560121765,
      "grad_norm": 5.28125,
      "learning_rate": 6.012176560121766e-06,
      "loss": 1.89589882,
      "memory(GiB)": 58.86,
      "step": 2370,
      "train_speed(iter/s)": 1.616262
    },
    {
      "acc": 0.62806449,
      "epoch": 0.06024860476915271,
      "grad_norm": 8.4375,
      "learning_rate": 6.0248604769152715e-06,
      "loss": 1.83353176,
      "memory(GiB)": 58.86,
      "step": 2375,
      "train_speed(iter/s)": 1.616668
    },
    {
      "acc": 0.62483759,
      "epoch": 0.06037544393708777,
      "grad_norm": 7.375,
      "learning_rate": 6.0375443937087775e-06,
      "loss": 1.77357826,
      "memory(GiB)": 58.86,
      "step": 2380,
      "train_speed(iter/s)": 1.617065
    },
    {
      "acc": 0.61796074,
      "epoch": 0.06050228310502283,
      "grad_norm": 5.78125,
      "learning_rate": 6.050228310502284e-06,
      "loss": 1.94466896,
      "memory(GiB)": 58.86,
      "step": 2385,
      "train_speed(iter/s)": 1.617489
    },
    {
      "acc": 0.62225666,
      "epoch": 0.060629122272957886,
      "grad_norm": 5.3125,
      "learning_rate": 6.062912227295789e-06,
      "loss": 1.85230179,
      "memory(GiB)": 58.86,
      "step": 2390,
      "train_speed(iter/s)": 1.617878
    },
    {
      "acc": 0.61804338,
      "epoch": 0.060755961440892944,
      "grad_norm": 6.59375,
      "learning_rate": 6.075596144089295e-06,
      "loss": 1.77972374,
      "memory(GiB)": 58.86,
      "step": 2395,
      "train_speed(iter/s)": 1.618289
    },
    {
      "acc": 0.61175203,
      "epoch": 0.060882800608828,
      "grad_norm": 7.28125,
      "learning_rate": 6.0882800608828005e-06,
      "loss": 1.92963619,
      "memory(GiB)": 58.86,
      "step": 2400,
      "train_speed(iter/s)": 1.618705
    },
    {
      "acc": 0.60799284,
      "epoch": 0.06100963977676306,
      "grad_norm": 5.34375,
      "learning_rate": 6.100963977676307e-06,
      "loss": 1.9112608,
      "memory(GiB)": 58.86,
      "step": 2405,
      "train_speed(iter/s)": 1.619081
    },
    {
      "acc": 0.60597339,
      "epoch": 0.06113647894469812,
      "grad_norm": 6.03125,
      "learning_rate": 6.113647894469813e-06,
      "loss": 1.90033302,
      "memory(GiB)": 58.86,
      "step": 2410,
      "train_speed(iter/s)": 1.619485
    },
    {
      "acc": 0.63157434,
      "epoch": 0.06126331811263318,
      "grad_norm": 5.6875,
      "learning_rate": 6.126331811263318e-06,
      "loss": 1.7876379,
      "memory(GiB)": 58.86,
      "step": 2415,
      "train_speed(iter/s)": 1.619862
    },
    {
      "acc": 0.63476391,
      "epoch": 0.061390157280568236,
      "grad_norm": 5.9375,
      "learning_rate": 6.139015728056824e-06,
      "loss": 1.80197964,
      "memory(GiB)": 58.86,
      "step": 2420,
      "train_speed(iter/s)": 1.620255
    },
    {
      "acc": 0.61584435,
      "epoch": 0.061516996448503294,
      "grad_norm": 5.96875,
      "learning_rate": 6.151699644850331e-06,
      "loss": 1.8763895,
      "memory(GiB)": 58.86,
      "step": 2425,
      "train_speed(iter/s)": 1.620642
    },
    {
      "acc": 0.63030863,
      "epoch": 0.06164383561643835,
      "grad_norm": 5.09375,
      "learning_rate": 6.164383561643836e-06,
      "loss": 1.77596703,
      "memory(GiB)": 58.86,
      "step": 2430,
      "train_speed(iter/s)": 1.621042
    },
    {
      "acc": 0.60907698,
      "epoch": 0.06177067478437341,
      "grad_norm": 5.9375,
      "learning_rate": 6.177067478437342e-06,
      "loss": 1.89386005,
      "memory(GiB)": 58.86,
      "step": 2435,
      "train_speed(iter/s)": 1.621415
    },
    {
      "acc": 0.61367273,
      "epoch": 0.06189751395230847,
      "grad_norm": 5.78125,
      "learning_rate": 6.189751395230847e-06,
      "loss": 1.88750191,
      "memory(GiB)": 58.86,
      "step": 2440,
      "train_speed(iter/s)": 1.621802
    },
    {
      "acc": 0.60630226,
      "epoch": 0.06202435312024353,
      "grad_norm": 10.125,
      "learning_rate": 6.202435312024354e-06,
      "loss": 1.90389862,
      "memory(GiB)": 58.86,
      "step": 2445,
      "train_speed(iter/s)": 1.622167
    },
    {
      "acc": 0.63596616,
      "epoch": 0.062151192288178586,
      "grad_norm": 6.875,
      "learning_rate": 6.215119228817859e-06,
      "loss": 1.74295464,
      "memory(GiB)": 58.86,
      "step": 2450,
      "train_speed(iter/s)": 1.622545
    },
    {
      "acc": 0.60849113,
      "epoch": 0.062278031456113644,
      "grad_norm": 5.5625,
      "learning_rate": 6.227803145611365e-06,
      "loss": 1.89324112,
      "memory(GiB)": 58.86,
      "step": 2455,
      "train_speed(iter/s)": 1.622946
    },
    {
      "acc": 0.60429263,
      "epoch": 0.0624048706240487,
      "grad_norm": 6.65625,
      "learning_rate": 6.24048706240487e-06,
      "loss": 1.89511909,
      "memory(GiB)": 58.86,
      "step": 2460,
      "train_speed(iter/s)": 1.623309
    },
    {
      "acc": 0.62859902,
      "epoch": 0.06253170979198376,
      "grad_norm": 8.1875,
      "learning_rate": 6.253170979198377e-06,
      "loss": 1.83797894,
      "memory(GiB)": 58.86,
      "step": 2465,
      "train_speed(iter/s)": 1.623669
    },
    {
      "acc": 0.61564684,
      "epoch": 0.06265854895991882,
      "grad_norm": 6.71875,
      "learning_rate": 6.265854895991883e-06,
      "loss": 1.86227913,
      "memory(GiB)": 58.86,
      "step": 2470,
      "train_speed(iter/s)": 1.624028
    },
    {
      "acc": 0.62765665,
      "epoch": 0.06278538812785388,
      "grad_norm": 6.5625,
      "learning_rate": 6.278538812785388e-06,
      "loss": 1.87822113,
      "memory(GiB)": 58.86,
      "step": 2475,
      "train_speed(iter/s)": 1.624398
    },
    {
      "acc": 0.62202787,
      "epoch": 0.06291222729578894,
      "grad_norm": 6.78125,
      "learning_rate": 6.291222729578894e-06,
      "loss": 1.84174767,
      "memory(GiB)": 58.86,
      "step": 2480,
      "train_speed(iter/s)": 1.624764
    },
    {
      "acc": 0.61823549,
      "epoch": 0.063039066463724,
      "grad_norm": 5.34375,
      "learning_rate": 6.303906646372401e-06,
      "loss": 1.92524567,
      "memory(GiB)": 58.86,
      "step": 2485,
      "train_speed(iter/s)": 1.625121
    },
    {
      "acc": 0.62868605,
      "epoch": 0.06316590563165905,
      "grad_norm": 7.75,
      "learning_rate": 6.316590563165906e-06,
      "loss": 1.82348976,
      "memory(GiB)": 58.86,
      "step": 2490,
      "train_speed(iter/s)": 1.625492
    },
    {
      "acc": 0.61429405,
      "epoch": 0.06329274479959411,
      "grad_norm": 6.625,
      "learning_rate": 6.329274479959412e-06,
      "loss": 1.84833794,
      "memory(GiB)": 58.86,
      "step": 2495,
      "train_speed(iter/s)": 1.625888
    },
    {
      "acc": 0.6139091,
      "epoch": 0.06341958396752917,
      "grad_norm": 5.84375,
      "learning_rate": 6.341958396752917e-06,
      "loss": 1.91333733,
      "memory(GiB)": 58.86,
      "step": 2500,
      "train_speed(iter/s)": 1.626232
    },
    {
      "acc": 0.62977324,
      "epoch": 0.06354642313546423,
      "grad_norm": 7.0,
      "learning_rate": 6.354642313546424e-06,
      "loss": 1.86739655,
      "memory(GiB)": 58.86,
      "step": 2505,
      "train_speed(iter/s)": 1.626627
    },
    {
      "acc": 0.61065321,
      "epoch": 0.06367326230339929,
      "grad_norm": 5.28125,
      "learning_rate": 6.36732623033993e-06,
      "loss": 1.8769371,
      "memory(GiB)": 58.86,
      "step": 2510,
      "train_speed(iter/s)": 1.626989
    },
    {
      "acc": 0.62435513,
      "epoch": 0.06380010147133434,
      "grad_norm": 5.25,
      "learning_rate": 6.380010147133435e-06,
      "loss": 1.76357212,
      "memory(GiB)": 58.86,
      "step": 2515,
      "train_speed(iter/s)": 1.627336
    },
    {
      "acc": 0.62186022,
      "epoch": 0.0639269406392694,
      "grad_norm": 5.8125,
      "learning_rate": 6.392694063926941e-06,
      "loss": 1.87364273,
      "memory(GiB)": 58.86,
      "step": 2520,
      "train_speed(iter/s)": 1.627699
    },
    {
      "acc": 0.60377526,
      "epoch": 0.06405377980720446,
      "grad_norm": 5.8125,
      "learning_rate": 6.405377980720447e-06,
      "loss": 1.89314537,
      "memory(GiB)": 58.86,
      "step": 2525,
      "train_speed(iter/s)": 1.628049
    },
    {
      "acc": 0.63855057,
      "epoch": 0.06418061897513952,
      "grad_norm": 7.75,
      "learning_rate": 6.418061897513953e-06,
      "loss": 1.75219975,
      "memory(GiB)": 58.86,
      "step": 2530,
      "train_speed(iter/s)": 1.628442
    },
    {
      "acc": 0.61520262,
      "epoch": 0.06430745814307458,
      "grad_norm": 7.90625,
      "learning_rate": 6.430745814307458e-06,
      "loss": 1.82468681,
      "memory(GiB)": 58.86,
      "step": 2535,
      "train_speed(iter/s)": 1.628795
    },
    {
      "acc": 0.6168025,
      "epoch": 0.06443429731100964,
      "grad_norm": 5.6875,
      "learning_rate": 6.443429731100964e-06,
      "loss": 1.86709976,
      "memory(GiB)": 58.86,
      "step": 2540,
      "train_speed(iter/s)": 1.629145
    },
    {
      "acc": 0.64176402,
      "epoch": 0.0645611364789447,
      "grad_norm": 6.15625,
      "learning_rate": 6.456113647894471e-06,
      "loss": 1.78402672,
      "memory(GiB)": 58.86,
      "step": 2545,
      "train_speed(iter/s)": 1.629506
    },
    {
      "acc": 0.62245908,
      "epoch": 0.06468797564687975,
      "grad_norm": 4.71875,
      "learning_rate": 6.468797564687976e-06,
      "loss": 1.80975246,
      "memory(GiB)": 58.86,
      "step": 2550,
      "train_speed(iter/s)": 1.629843
    },
    {
      "acc": 0.64164104,
      "epoch": 0.06481481481481481,
      "grad_norm": 6.78125,
      "learning_rate": 6.481481481481482e-06,
      "loss": 1.72489376,
      "memory(GiB)": 58.86,
      "step": 2555,
      "train_speed(iter/s)": 1.6302
    },
    {
      "acc": 0.62354574,
      "epoch": 0.06494165398274987,
      "grad_norm": 5.46875,
      "learning_rate": 6.494165398274987e-06,
      "loss": 1.83460884,
      "memory(GiB)": 58.86,
      "step": 2560,
      "train_speed(iter/s)": 1.63058
    },
    {
      "acc": 0.60753074,
      "epoch": 0.06506849315068493,
      "grad_norm": 5.96875,
      "learning_rate": 6.506849315068494e-06,
      "loss": 1.86012058,
      "memory(GiB)": 58.86,
      "step": 2565,
      "train_speed(iter/s)": 1.630945
    },
    {
      "acc": 0.61494865,
      "epoch": 0.06519533231861999,
      "grad_norm": 5.84375,
      "learning_rate": 6.519533231862e-06,
      "loss": 1.8817318,
      "memory(GiB)": 58.86,
      "step": 2570,
      "train_speed(iter/s)": 1.631325
    },
    {
      "acc": 0.62781019,
      "epoch": 0.06532217148655504,
      "grad_norm": 6.3125,
      "learning_rate": 6.532217148655505e-06,
      "loss": 1.79104652,
      "memory(GiB)": 58.86,
      "step": 2575,
      "train_speed(iter/s)": 1.631683
    },
    {
      "acc": 0.61440473,
      "epoch": 0.0654490106544901,
      "grad_norm": 5.34375,
      "learning_rate": 6.544901065449011e-06,
      "loss": 1.99972744,
      "memory(GiB)": 58.86,
      "step": 2580,
      "train_speed(iter/s)": 1.632039
    },
    {
      "acc": 0.60830774,
      "epoch": 0.06557584982242516,
      "grad_norm": 10.5,
      "learning_rate": 6.557584982242518e-06,
      "loss": 1.91051674,
      "memory(GiB)": 58.86,
      "step": 2585,
      "train_speed(iter/s)": 1.632391
    },
    {
      "acc": 0.620649,
      "epoch": 0.06570268899036022,
      "grad_norm": 5.78125,
      "learning_rate": 6.570268899036023e-06,
      "loss": 1.86778793,
      "memory(GiB)": 58.86,
      "step": 2590,
      "train_speed(iter/s)": 1.63273
    },
    {
      "acc": 0.61801691,
      "epoch": 0.06582952815829528,
      "grad_norm": 5.90625,
      "learning_rate": 6.582952815829529e-06,
      "loss": 1.8691227,
      "memory(GiB)": 58.86,
      "step": 2595,
      "train_speed(iter/s)": 1.633066
    },
    {
      "acc": 0.62306376,
      "epoch": 0.06595636732623034,
      "grad_norm": 6.65625,
      "learning_rate": 6.595636732623034e-06,
      "loss": 1.92729607,
      "memory(GiB)": 58.86,
      "step": 2600,
      "train_speed(iter/s)": 1.633407
    },
    {
      "acc": 0.62467918,
      "epoch": 0.0660832064941654,
      "grad_norm": 7.84375,
      "learning_rate": 6.608320649416541e-06,
      "loss": 1.80098896,
      "memory(GiB)": 58.86,
      "step": 2605,
      "train_speed(iter/s)": 1.633762
    },
    {
      "acc": 0.61991296,
      "epoch": 0.06621004566210045,
      "grad_norm": 6.9375,
      "learning_rate": 6.621004566210046e-06,
      "loss": 1.8717968,
      "memory(GiB)": 58.86,
      "step": 2610,
      "train_speed(iter/s)": 1.634118
    },
    {
      "acc": 0.62227192,
      "epoch": 0.06633688483003551,
      "grad_norm": 8.4375,
      "learning_rate": 6.633688483003552e-06,
      "loss": 1.82289581,
      "memory(GiB)": 58.86,
      "step": 2615,
      "train_speed(iter/s)": 1.634445
    },
    {
      "acc": 0.61607914,
      "epoch": 0.06646372399797057,
      "grad_norm": 8.875,
      "learning_rate": 6.646372399797057e-06,
      "loss": 1.83439407,
      "memory(GiB)": 58.86,
      "step": 2620,
      "train_speed(iter/s)": 1.634792
    },
    {
      "acc": 0.62091675,
      "epoch": 0.06659056316590563,
      "grad_norm": 7.25,
      "learning_rate": 6.659056316590564e-06,
      "loss": 1.84438725,
      "memory(GiB)": 58.86,
      "step": 2625,
      "train_speed(iter/s)": 1.635129
    },
    {
      "acc": 0.61466908,
      "epoch": 0.06671740233384069,
      "grad_norm": 7.9375,
      "learning_rate": 6.67174023338407e-06,
      "loss": 1.85825348,
      "memory(GiB)": 58.86,
      "step": 2630,
      "train_speed(iter/s)": 1.635488
    },
    {
      "acc": 0.62345629,
      "epoch": 0.06684424150177574,
      "grad_norm": 7.84375,
      "learning_rate": 6.684424150177575e-06,
      "loss": 1.85583801,
      "memory(GiB)": 58.86,
      "step": 2635,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.62290425,
      "epoch": 0.0669710806697108,
      "grad_norm": 7.0,
      "learning_rate": 6.697108066971081e-06,
      "loss": 1.89310856,
      "memory(GiB)": 58.86,
      "step": 2640,
      "train_speed(iter/s)": 1.636168
    },
    {
      "acc": 0.62458086,
      "epoch": 0.06709791983764586,
      "grad_norm": 7.0,
      "learning_rate": 6.709791983764588e-06,
      "loss": 1.79221916,
      "memory(GiB)": 58.86,
      "step": 2645,
      "train_speed(iter/s)": 1.636509
    },
    {
      "acc": 0.62424927,
      "epoch": 0.06722475900558092,
      "grad_norm": 7.375,
      "learning_rate": 6.722475900558093e-06,
      "loss": 1.82292728,
      "memory(GiB)": 58.86,
      "step": 2650,
      "train_speed(iter/s)": 1.636837
    },
    {
      "acc": 0.63885889,
      "epoch": 0.06735159817351598,
      "grad_norm": 10.8125,
      "learning_rate": 6.735159817351599e-06,
      "loss": 1.7846859,
      "memory(GiB)": 58.86,
      "step": 2655,
      "train_speed(iter/s)": 1.637161
    },
    {
      "acc": 0.63184495,
      "epoch": 0.06747843734145104,
      "grad_norm": 7.4375,
      "learning_rate": 6.747843734145104e-06,
      "loss": 1.83132782,
      "memory(GiB)": 58.86,
      "step": 2660,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.62372584,
      "epoch": 0.0676052765093861,
      "grad_norm": 6.1875,
      "learning_rate": 6.760527650938611e-06,
      "loss": 1.79030361,
      "memory(GiB)": 58.86,
      "step": 2665,
      "train_speed(iter/s)": 1.637871
    },
    {
      "acc": 0.61351156,
      "epoch": 0.06773211567732115,
      "grad_norm": 6.125,
      "learning_rate": 6.773211567732117e-06,
      "loss": 1.86900635,
      "memory(GiB)": 58.86,
      "step": 2670,
      "train_speed(iter/s)": 1.638227
    },
    {
      "acc": 0.6281414,
      "epoch": 0.06785895484525621,
      "grad_norm": 5.4375,
      "learning_rate": 6.785895484525622e-06,
      "loss": 1.90075188,
      "memory(GiB)": 58.86,
      "step": 2675,
      "train_speed(iter/s)": 1.638557
    },
    {
      "acc": 0.6361618,
      "epoch": 0.06798579401319127,
      "grad_norm": 6.3125,
      "learning_rate": 6.798579401319128e-06,
      "loss": 1.78285904,
      "memory(GiB)": 58.86,
      "step": 2680,
      "train_speed(iter/s)": 1.638903
    },
    {
      "acc": 0.62580462,
      "epoch": 0.06811263318112633,
      "grad_norm": 6.25,
      "learning_rate": 6.811263318112634e-06,
      "loss": 1.87772331,
      "memory(GiB)": 58.86,
      "step": 2685,
      "train_speed(iter/s)": 1.639252
    },
    {
      "acc": 0.6119997,
      "epoch": 0.06823947234906139,
      "grad_norm": 6.5,
      "learning_rate": 6.82394723490614e-06,
      "loss": 1.84669895,
      "memory(GiB)": 58.86,
      "step": 2690,
      "train_speed(iter/s)": 1.639594
    },
    {
      "acc": 0.62462873,
      "epoch": 0.06836631151699644,
      "grad_norm": 5.46875,
      "learning_rate": 6.836631151699645e-06,
      "loss": 1.86291733,
      "memory(GiB)": 58.86,
      "step": 2695,
      "train_speed(iter/s)": 1.639927
    },
    {
      "acc": 0.60608716,
      "epoch": 0.0684931506849315,
      "grad_norm": 7.0625,
      "learning_rate": 6.849315068493151e-06,
      "loss": 1.9127264,
      "memory(GiB)": 58.86,
      "step": 2700,
      "train_speed(iter/s)": 1.640256
    },
    {
      "acc": 0.61509128,
      "epoch": 0.06861998985286656,
      "grad_norm": 6.59375,
      "learning_rate": 6.861998985286658e-06,
      "loss": 1.89054337,
      "memory(GiB)": 58.86,
      "step": 2705,
      "train_speed(iter/s)": 1.640593
    },
    {
      "acc": 0.61531215,
      "epoch": 0.06874682902080162,
      "grad_norm": 6.8125,
      "learning_rate": 6.874682902080163e-06,
      "loss": 1.94848938,
      "memory(GiB)": 58.86,
      "step": 2710,
      "train_speed(iter/s)": 1.640941
    },
    {
      "acc": 0.62668724,
      "epoch": 0.06887366818873668,
      "grad_norm": 6.03125,
      "learning_rate": 6.887366818873669e-06,
      "loss": 1.79132462,
      "memory(GiB)": 58.86,
      "step": 2715,
      "train_speed(iter/s)": 1.641258
    },
    {
      "acc": 0.62493963,
      "epoch": 0.06900050735667174,
      "grad_norm": 6.96875,
      "learning_rate": 6.900050735667174e-06,
      "loss": 1.8403141,
      "memory(GiB)": 58.86,
      "step": 2720,
      "train_speed(iter/s)": 1.6416
    },
    {
      "acc": 0.62243023,
      "epoch": 0.0691273465246068,
      "grad_norm": 5.8125,
      "learning_rate": 6.912734652460681e-06,
      "loss": 1.82509041,
      "memory(GiB)": 58.86,
      "step": 2725,
      "train_speed(iter/s)": 1.641932
    },
    {
      "acc": 0.61616879,
      "epoch": 0.06925418569254185,
      "grad_norm": 9.875,
      "learning_rate": 6.925418569254187e-06,
      "loss": 1.81707802,
      "memory(GiB)": 58.86,
      "step": 2730,
      "train_speed(iter/s)": 1.64227
    },
    {
      "acc": 0.61606154,
      "epoch": 0.06938102486047691,
      "grad_norm": 6.21875,
      "learning_rate": 6.938102486047692e-06,
      "loss": 1.80503426,
      "memory(GiB)": 58.86,
      "step": 2735,
      "train_speed(iter/s)": 1.642583
    },
    {
      "acc": 0.61662331,
      "epoch": 0.06950786402841197,
      "grad_norm": 5.71875,
      "learning_rate": 6.950786402841198e-06,
      "loss": 1.82287674,
      "memory(GiB)": 58.86,
      "step": 2740,
      "train_speed(iter/s)": 1.642902
    },
    {
      "acc": 0.63214316,
      "epoch": 0.06963470319634703,
      "grad_norm": 5.9375,
      "learning_rate": 6.9634703196347046e-06,
      "loss": 1.77682457,
      "memory(GiB)": 58.86,
      "step": 2745,
      "train_speed(iter/s)": 1.643234
    },
    {
      "acc": 0.62459421,
      "epoch": 0.06976154236428209,
      "grad_norm": 5.78125,
      "learning_rate": 6.97615423642821e-06,
      "loss": 1.86816711,
      "memory(GiB)": 58.86,
      "step": 2750,
      "train_speed(iter/s)": 1.643574
    },
    {
      "acc": 0.61551743,
      "epoch": 0.06988838153221714,
      "grad_norm": 6.09375,
      "learning_rate": 6.988838153221716e-06,
      "loss": 1.86809044,
      "memory(GiB)": 58.86,
      "step": 2755,
      "train_speed(iter/s)": 1.643898
    },
    {
      "acc": 0.61875954,
      "epoch": 0.0700152207001522,
      "grad_norm": 5.59375,
      "learning_rate": 7.001522070015221e-06,
      "loss": 1.8576107,
      "memory(GiB)": 58.86,
      "step": 2760,
      "train_speed(iter/s)": 1.644181
    },
    {
      "acc": 0.63215113,
      "epoch": 0.07014205986808726,
      "grad_norm": 6.375,
      "learning_rate": 7.014205986808728e-06,
      "loss": 1.89916,
      "memory(GiB)": 58.86,
      "step": 2765,
      "train_speed(iter/s)": 1.644487
    },
    {
      "acc": 0.62541895,
      "epoch": 0.07026889903602232,
      "grad_norm": 6.40625,
      "learning_rate": 7.026889903602233e-06,
      "loss": 1.84470425,
      "memory(GiB)": 58.86,
      "step": 2770,
      "train_speed(iter/s)": 1.6448
    },
    {
      "acc": 0.62770796,
      "epoch": 0.07039573820395738,
      "grad_norm": 7.84375,
      "learning_rate": 7.039573820395739e-06,
      "loss": 1.83992119,
      "memory(GiB)": 58.86,
      "step": 2775,
      "train_speed(iter/s)": 1.645114
    },
    {
      "acc": 0.6036365,
      "epoch": 0.07052257737189244,
      "grad_norm": 6.1875,
      "learning_rate": 7.052257737189245e-06,
      "loss": 1.89441395,
      "memory(GiB)": 58.86,
      "step": 2780,
      "train_speed(iter/s)": 1.64544
    },
    {
      "acc": 0.62922888,
      "epoch": 0.0706494165398275,
      "grad_norm": 6.59375,
      "learning_rate": 7.064941653982751e-06,
      "loss": 1.83304195,
      "memory(GiB)": 58.86,
      "step": 2785,
      "train_speed(iter/s)": 1.645744
    },
    {
      "acc": 0.614466,
      "epoch": 0.07077625570776255,
      "grad_norm": 6.28125,
      "learning_rate": 7.077625570776257e-06,
      "loss": 1.87623177,
      "memory(GiB)": 58.86,
      "step": 2790,
      "train_speed(iter/s)": 1.646046
    },
    {
      "acc": 0.6314908,
      "epoch": 0.07090309487569761,
      "grad_norm": 6.90625,
      "learning_rate": 7.090309487569762e-06,
      "loss": 1.75695553,
      "memory(GiB)": 58.86,
      "step": 2795,
      "train_speed(iter/s)": 1.646371
    },
    {
      "acc": 0.64543848,
      "epoch": 0.07102993404363267,
      "grad_norm": 6.6875,
      "learning_rate": 7.102993404363268e-06,
      "loss": 1.75454407,
      "memory(GiB)": 58.86,
      "step": 2800,
      "train_speed(iter/s)": 1.646667
    },
    {
      "acc": 0.63171625,
      "epoch": 0.07115677321156773,
      "grad_norm": 6.5,
      "learning_rate": 7.1156773211567745e-06,
      "loss": 1.83182487,
      "memory(GiB)": 58.86,
      "step": 2805,
      "train_speed(iter/s)": 1.646961
    },
    {
      "acc": 0.63060598,
      "epoch": 0.07128361237950279,
      "grad_norm": 7.25,
      "learning_rate": 7.12836123795028e-06,
      "loss": 1.85082245,
      "memory(GiB)": 58.86,
      "step": 2810,
      "train_speed(iter/s)": 1.647279
    },
    {
      "acc": 0.60732064,
      "epoch": 0.07141045154743784,
      "grad_norm": 5.28125,
      "learning_rate": 7.1410451547437856e-06,
      "loss": 1.85291862,
      "memory(GiB)": 58.86,
      "step": 2815,
      "train_speed(iter/s)": 1.647595
    },
    {
      "acc": 0.61993437,
      "epoch": 0.0715372907153729,
      "grad_norm": 6.65625,
      "learning_rate": 7.153729071537291e-06,
      "loss": 1.76142082,
      "memory(GiB)": 58.86,
      "step": 2820,
      "train_speed(iter/s)": 1.647889
    },
    {
      "acc": 0.63077378,
      "epoch": 0.07166412988330796,
      "grad_norm": 5.875,
      "learning_rate": 7.1664129883307975e-06,
      "loss": 1.80787868,
      "memory(GiB)": 58.86,
      "step": 2825,
      "train_speed(iter/s)": 1.648201
    },
    {
      "acc": 0.64034786,
      "epoch": 0.07179096905124302,
      "grad_norm": 5.75,
      "learning_rate": 7.1790969051243035e-06,
      "loss": 1.76725464,
      "memory(GiB)": 58.86,
      "step": 2830,
      "train_speed(iter/s)": 1.648504
    },
    {
      "acc": 0.60083098,
      "epoch": 0.07191780821917808,
      "grad_norm": 8.0,
      "learning_rate": 7.191780821917809e-06,
      "loss": 1.93156967,
      "memory(GiB)": 58.86,
      "step": 2835,
      "train_speed(iter/s)": 1.648821
    },
    {
      "acc": 0.61838884,
      "epoch": 0.07204464738711314,
      "grad_norm": 6.625,
      "learning_rate": 7.2044647387113146e-06,
      "loss": 1.87825241,
      "memory(GiB)": 58.86,
      "step": 2840,
      "train_speed(iter/s)": 1.649119
    },
    {
      "acc": 0.62993474,
      "epoch": 0.0721714865550482,
      "grad_norm": 5.375,
      "learning_rate": 7.2171486555048205e-06,
      "loss": 1.78725872,
      "memory(GiB)": 58.86,
      "step": 2845,
      "train_speed(iter/s)": 1.649423
    },
    {
      "acc": 0.61560841,
      "epoch": 0.07229832572298325,
      "grad_norm": 6.0625,
      "learning_rate": 7.2298325722983265e-06,
      "loss": 1.83525448,
      "memory(GiB)": 58.86,
      "step": 2850,
      "train_speed(iter/s)": 1.649738
    },
    {
      "acc": 0.64150791,
      "epoch": 0.07242516489091831,
      "grad_norm": 6.4375,
      "learning_rate": 7.242516489091832e-06,
      "loss": 1.83049316,
      "memory(GiB)": 58.86,
      "step": 2855,
      "train_speed(iter/s)": 1.650053
    },
    {
      "acc": 0.62604313,
      "epoch": 0.07255200405885337,
      "grad_norm": 6.25,
      "learning_rate": 7.2552004058853376e-06,
      "loss": 1.77583008,
      "memory(GiB)": 58.86,
      "step": 2860,
      "train_speed(iter/s)": 1.650342
    },
    {
      "acc": 0.62669916,
      "epoch": 0.07267884322678843,
      "grad_norm": 6.03125,
      "learning_rate": 7.267884322678844e-06,
      "loss": 1.89665585,
      "memory(GiB)": 58.86,
      "step": 2865,
      "train_speed(iter/s)": 1.650624
    },
    {
      "acc": 0.62018375,
      "epoch": 0.07280568239472349,
      "grad_norm": 7.21875,
      "learning_rate": 7.2805682394723495e-06,
      "loss": 1.8666666,
      "memory(GiB)": 58.86,
      "step": 2870,
      "train_speed(iter/s)": 1.650898
    },
    {
      "acc": 0.61261616,
      "epoch": 0.07293252156265854,
      "grad_norm": 6.1875,
      "learning_rate": 7.2932521562658555e-06,
      "loss": 1.85211277,
      "memory(GiB)": 58.86,
      "step": 2875,
      "train_speed(iter/s)": 1.651206
    },
    {
      "acc": 0.6122263,
      "epoch": 0.0730593607305936,
      "grad_norm": 5.40625,
      "learning_rate": 7.305936073059361e-06,
      "loss": 1.83267956,
      "memory(GiB)": 58.86,
      "step": 2880,
      "train_speed(iter/s)": 1.651484
    },
    {
      "acc": 0.63585782,
      "epoch": 0.07318619989852866,
      "grad_norm": 5.15625,
      "learning_rate": 7.318619989852867e-06,
      "loss": 1.83922138,
      "memory(GiB)": 58.86,
      "step": 2885,
      "train_speed(iter/s)": 1.651781
    },
    {
      "acc": 0.61210012,
      "epoch": 0.07331303906646372,
      "grad_norm": 7.375,
      "learning_rate": 7.331303906646373e-06,
      "loss": 1.82524376,
      "memory(GiB)": 58.86,
      "step": 2890,
      "train_speed(iter/s)": 1.652078
    },
    {
      "acc": 0.63380165,
      "epoch": 0.07343987823439878,
      "grad_norm": 5.6875,
      "learning_rate": 7.3439878234398785e-06,
      "loss": 1.80623512,
      "memory(GiB)": 58.86,
      "step": 2895,
      "train_speed(iter/s)": 1.652358
    },
    {
      "acc": 0.61072969,
      "epoch": 0.07356671740233384,
      "grad_norm": 6.21875,
      "learning_rate": 7.3566717402333845e-06,
      "loss": 1.95809822,
      "memory(GiB)": 58.86,
      "step": 2900,
      "train_speed(iter/s)": 1.652628
    },
    {
      "acc": 0.62094526,
      "epoch": 0.0736935565702689,
      "grad_norm": 6.34375,
      "learning_rate": 7.369355657026891e-06,
      "loss": 1.83135757,
      "memory(GiB)": 58.86,
      "step": 2905,
      "train_speed(iter/s)": 1.652915
    },
    {
      "acc": 0.59886999,
      "epoch": 0.07382039573820395,
      "grad_norm": 4.9375,
      "learning_rate": 7.382039573820396e-06,
      "loss": 1.9506773,
      "memory(GiB)": 58.86,
      "step": 2910,
      "train_speed(iter/s)": 1.653203
    },
    {
      "acc": 0.62148046,
      "epoch": 0.07394723490613901,
      "grad_norm": 7.59375,
      "learning_rate": 7.394723490613902e-06,
      "loss": 1.86128883,
      "memory(GiB)": 58.86,
      "step": 2915,
      "train_speed(iter/s)": 1.653505
    },
    {
      "acc": 0.60896997,
      "epoch": 0.07407407407407407,
      "grad_norm": 5.9375,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 1.9052927,
      "memory(GiB)": 58.86,
      "step": 2920,
      "train_speed(iter/s)": 1.653763
    },
    {
      "acc": 0.61220808,
      "epoch": 0.07420091324200913,
      "grad_norm": 6.0625,
      "learning_rate": 7.420091324200914e-06,
      "loss": 1.87214355,
      "memory(GiB)": 58.86,
      "step": 2925,
      "train_speed(iter/s)": 1.654055
    },
    {
      "acc": 0.61939154,
      "epoch": 0.07432775240994419,
      "grad_norm": 7.5,
      "learning_rate": 7.432775240994419e-06,
      "loss": 1.87435989,
      "memory(GiB)": 58.86,
      "step": 2930,
      "train_speed(iter/s)": 1.654343
    },
    {
      "acc": 0.63307371,
      "epoch": 0.07445459157787924,
      "grad_norm": 6.625,
      "learning_rate": 7.445459157787925e-06,
      "loss": 1.74041977,
      "memory(GiB)": 58.86,
      "step": 2935,
      "train_speed(iter/s)": 1.65463
    },
    {
      "acc": 0.631357,
      "epoch": 0.0745814307458143,
      "grad_norm": 6.3125,
      "learning_rate": 7.458143074581431e-06,
      "loss": 1.80306778,
      "memory(GiB)": 58.86,
      "step": 2940,
      "train_speed(iter/s)": 1.654906
    },
    {
      "acc": 0.64976625,
      "epoch": 0.07470826991374936,
      "grad_norm": 7.5625,
      "learning_rate": 7.470826991374937e-06,
      "loss": 1.70913353,
      "memory(GiB)": 58.86,
      "step": 2945,
      "train_speed(iter/s)": 1.655179
    },
    {
      "acc": 0.60800514,
      "epoch": 0.07483510908168442,
      "grad_norm": 8.4375,
      "learning_rate": 7.483510908168443e-06,
      "loss": 1.92534523,
      "memory(GiB)": 58.86,
      "step": 2950,
      "train_speed(iter/s)": 1.655468
    },
    {
      "acc": 0.63667126,
      "epoch": 0.07496194824961948,
      "grad_norm": 5.59375,
      "learning_rate": 7.496194824961948e-06,
      "loss": 1.7739357,
      "memory(GiB)": 58.86,
      "step": 2955,
      "train_speed(iter/s)": 1.655722
    },
    {
      "acc": 0.6314599,
      "epoch": 0.07508878741755454,
      "grad_norm": 5.5625,
      "learning_rate": 7.508878741755454e-06,
      "loss": 1.76697655,
      "memory(GiB)": 58.86,
      "step": 2960,
      "train_speed(iter/s)": 1.656003
    },
    {
      "acc": 0.62168779,
      "epoch": 0.0752156265854896,
      "grad_norm": 6.25,
      "learning_rate": 7.521562658548961e-06,
      "loss": 1.80277596,
      "memory(GiB)": 58.86,
      "step": 2965,
      "train_speed(iter/s)": 1.656278
    },
    {
      "acc": 0.63421741,
      "epoch": 0.07534246575342465,
      "grad_norm": 6.3125,
      "learning_rate": 7.534246575342466e-06,
      "loss": 1.79069939,
      "memory(GiB)": 58.86,
      "step": 2970,
      "train_speed(iter/s)": 1.656559
    },
    {
      "acc": 0.63341808,
      "epoch": 0.07546930492135971,
      "grad_norm": 5.46875,
      "learning_rate": 7.546930492135972e-06,
      "loss": 1.768433,
      "memory(GiB)": 58.86,
      "step": 2975,
      "train_speed(iter/s)": 1.656832
    },
    {
      "acc": 0.62834415,
      "epoch": 0.07559614408929477,
      "grad_norm": 8.0625,
      "learning_rate": 7.559614408929477e-06,
      "loss": 1.76345177,
      "memory(GiB)": 58.86,
      "step": 2980,
      "train_speed(iter/s)": 1.657111
    },
    {
      "acc": 0.62464895,
      "epoch": 0.07572298325722983,
      "grad_norm": 6.3125,
      "learning_rate": 7.572298325722984e-06,
      "loss": 1.82218533,
      "memory(GiB)": 58.86,
      "step": 2985,
      "train_speed(iter/s)": 1.657393
    },
    {
      "acc": 0.63267317,
      "epoch": 0.07584982242516489,
      "grad_norm": 6.15625,
      "learning_rate": 7.58498224251649e-06,
      "loss": 1.80164795,
      "memory(GiB)": 58.86,
      "step": 2990,
      "train_speed(iter/s)": 1.657635
    },
    {
      "acc": 0.61835299,
      "epoch": 0.07597666159309995,
      "grad_norm": 5.59375,
      "learning_rate": 7.597666159309995e-06,
      "loss": 1.84410496,
      "memory(GiB)": 58.86,
      "step": 2995,
      "train_speed(iter/s)": 1.657911
    },
    {
      "acc": 0.62701006,
      "epoch": 0.076103500761035,
      "grad_norm": 6.34375,
      "learning_rate": 7.610350076103501e-06,
      "loss": 1.81502705,
      "memory(GiB)": 58.86,
      "step": 3000,
      "train_speed(iter/s)": 1.658185
    },
    {
      "epoch": 0.076103500761035,
      "eval_acc": 0.611511160952017,
      "eval_loss": 1.7780622243881226,
      "eval_runtime": 58.2332,
      "eval_samples_per_second": 109.388,
      "eval_steps_per_second": 27.356,
      "step": 3000
    },
    {
      "acc": 0.63783178,
      "epoch": 0.07623033992897006,
      "grad_norm": 4.90625,
      "learning_rate": 7.623033992897007e-06,
      "loss": 1.73620682,
      "memory(GiB)": 58.86,
      "step": 3005,
      "train_speed(iter/s)": 1.603207
    },
    {
      "acc": 0.61852002,
      "epoch": 0.07635717909690512,
      "grad_norm": 6.71875,
      "learning_rate": 7.635717909690512e-06,
      "loss": 1.84506187,
      "memory(GiB)": 58.86,
      "step": 3010,
      "train_speed(iter/s)": 1.603545
    },
    {
      "acc": 0.62819405,
      "epoch": 0.07648401826484018,
      "grad_norm": 6.71875,
      "learning_rate": 7.648401826484018e-06,
      "loss": 1.79380951,
      "memory(GiB)": 58.86,
      "step": 3015,
      "train_speed(iter/s)": 1.603885
    },
    {
      "acc": 0.63974152,
      "epoch": 0.07661085743277524,
      "grad_norm": 6.0,
      "learning_rate": 7.661085743277524e-06,
      "loss": 1.77939548,
      "memory(GiB)": 58.86,
      "step": 3020,
      "train_speed(iter/s)": 1.604229
    },
    {
      "acc": 0.63172846,
      "epoch": 0.0767376966007103,
      "grad_norm": 5.96875,
      "learning_rate": 7.67376966007103e-06,
      "loss": 1.81084213,
      "memory(GiB)": 58.86,
      "step": 3025,
      "train_speed(iter/s)": 1.604564
    },
    {
      "acc": 0.62624235,
      "epoch": 0.07686453576864535,
      "grad_norm": 5.5625,
      "learning_rate": 7.686453576864536e-06,
      "loss": 1.76887817,
      "memory(GiB)": 58.86,
      "step": 3030,
      "train_speed(iter/s)": 1.604882
    },
    {
      "acc": 0.61102958,
      "epoch": 0.07699137493658041,
      "grad_norm": 6.3125,
      "learning_rate": 7.699137493658042e-06,
      "loss": 1.83594704,
      "memory(GiB)": 58.86,
      "step": 3035,
      "train_speed(iter/s)": 1.605209
    },
    {
      "acc": 0.63113852,
      "epoch": 0.07711821410451547,
      "grad_norm": 6.1875,
      "learning_rate": 7.711821410451548e-06,
      "loss": 1.7567131,
      "memory(GiB)": 58.86,
      "step": 3040,
      "train_speed(iter/s)": 1.605541
    },
    {
      "acc": 0.6191988,
      "epoch": 0.07724505327245053,
      "grad_norm": 7.0,
      "learning_rate": 7.724505327245054e-06,
      "loss": 1.80660229,
      "memory(GiB)": 58.86,
      "step": 3045,
      "train_speed(iter/s)": 1.605879
    },
    {
      "acc": 0.62958927,
      "epoch": 0.07737189244038559,
      "grad_norm": 7.1875,
      "learning_rate": 7.73718924403856e-06,
      "loss": 1.79680157,
      "memory(GiB)": 58.86,
      "step": 3050,
      "train_speed(iter/s)": 1.606219
    },
    {
      "acc": 0.63014517,
      "epoch": 0.07749873160832065,
      "grad_norm": 5.21875,
      "learning_rate": 7.749873160832066e-06,
      "loss": 1.81305237,
      "memory(GiB)": 58.86,
      "step": 3055,
      "train_speed(iter/s)": 1.606538
    },
    {
      "acc": 0.62149038,
      "epoch": 0.0776255707762557,
      "grad_norm": 5.09375,
      "learning_rate": 7.76255707762557e-06,
      "loss": 1.79317703,
      "memory(GiB)": 58.86,
      "step": 3060,
      "train_speed(iter/s)": 1.606861
    },
    {
      "acc": 0.63585677,
      "epoch": 0.07775240994419076,
      "grad_norm": 6.71875,
      "learning_rate": 7.775240994419078e-06,
      "loss": 1.76710701,
      "memory(GiB)": 58.86,
      "step": 3065,
      "train_speed(iter/s)": 1.607193
    },
    {
      "acc": 0.62253685,
      "epoch": 0.07787924911212582,
      "grad_norm": 6.84375,
      "learning_rate": 7.787924911212584e-06,
      "loss": 1.83646851,
      "memory(GiB)": 58.86,
      "step": 3070,
      "train_speed(iter/s)": 1.607499
    },
    {
      "acc": 0.62994585,
      "epoch": 0.07800608828006088,
      "grad_norm": 7.25,
      "learning_rate": 7.800608828006088e-06,
      "loss": 1.83255272,
      "memory(GiB)": 58.86,
      "step": 3075,
      "train_speed(iter/s)": 1.607822
    },
    {
      "acc": 0.62628317,
      "epoch": 0.07813292744799594,
      "grad_norm": 5.9375,
      "learning_rate": 7.813292744799594e-06,
      "loss": 1.79040089,
      "memory(GiB)": 58.86,
      "step": 3080,
      "train_speed(iter/s)": 1.608151
    },
    {
      "acc": 0.6179585,
      "epoch": 0.078259766615931,
      "grad_norm": 6.15625,
      "learning_rate": 7.8259766615931e-06,
      "loss": 1.77247314,
      "memory(GiB)": 58.86,
      "step": 3085,
      "train_speed(iter/s)": 1.608442
    },
    {
      "acc": 0.62175183,
      "epoch": 0.07838660578386605,
      "grad_norm": 6.625,
      "learning_rate": 7.838660578386606e-06,
      "loss": 1.87037334,
      "memory(GiB)": 58.86,
      "step": 3090,
      "train_speed(iter/s)": 1.608778
    },
    {
      "acc": 0.62725267,
      "epoch": 0.07851344495180111,
      "grad_norm": 5.375,
      "learning_rate": 7.851344495180112e-06,
      "loss": 1.85108929,
      "memory(GiB)": 58.86,
      "step": 3095,
      "train_speed(iter/s)": 1.609085
    },
    {
      "acc": 0.62076268,
      "epoch": 0.07864028411973617,
      "grad_norm": 5.84375,
      "learning_rate": 7.864028411973618e-06,
      "loss": 1.83174133,
      "memory(GiB)": 58.86,
      "step": 3100,
      "train_speed(iter/s)": 1.609377
    },
    {
      "acc": 0.62446389,
      "epoch": 0.07876712328767123,
      "grad_norm": 6.28125,
      "learning_rate": 7.876712328767124e-06,
      "loss": 1.79310532,
      "memory(GiB)": 58.86,
      "step": 3105,
      "train_speed(iter/s)": 1.609705
    },
    {
      "acc": 0.61990042,
      "epoch": 0.07889396245560629,
      "grad_norm": 6.125,
      "learning_rate": 7.88939624556063e-06,
      "loss": 1.84493599,
      "memory(GiB)": 58.86,
      "step": 3110,
      "train_speed(iter/s)": 1.610017
    },
    {
      "acc": 0.63923712,
      "epoch": 0.07902080162354135,
      "grad_norm": 6.375,
      "learning_rate": 7.902080162354136e-06,
      "loss": 1.71876297,
      "memory(GiB)": 58.86,
      "step": 3115,
      "train_speed(iter/s)": 1.610349
    },
    {
      "acc": 0.65725875,
      "epoch": 0.0791476407914764,
      "grad_norm": 6.40625,
      "learning_rate": 7.91476407914764e-06,
      "loss": 1.68763828,
      "memory(GiB)": 58.86,
      "step": 3120,
      "train_speed(iter/s)": 1.610685
    },
    {
      "acc": 0.6308207,
      "epoch": 0.07927447995941146,
      "grad_norm": 7.0,
      "learning_rate": 7.927447995941148e-06,
      "loss": 1.7824398,
      "memory(GiB)": 58.86,
      "step": 3125,
      "train_speed(iter/s)": 1.610988
    },
    {
      "acc": 0.62523913,
      "epoch": 0.07940131912734652,
      "grad_norm": 6.71875,
      "learning_rate": 7.940131912734654e-06,
      "loss": 1.81446304,
      "memory(GiB)": 58.86,
      "step": 3130,
      "train_speed(iter/s)": 1.611317
    },
    {
      "acc": 0.63848977,
      "epoch": 0.07952815829528158,
      "grad_norm": 5.21875,
      "learning_rate": 7.952815829528158e-06,
      "loss": 1.73555603,
      "memory(GiB)": 58.86,
      "step": 3135,
      "train_speed(iter/s)": 1.61164
    },
    {
      "acc": 0.63784361,
      "epoch": 0.07965499746321664,
      "grad_norm": 6.375,
      "learning_rate": 7.965499746321664e-06,
      "loss": 1.82037811,
      "memory(GiB)": 58.86,
      "step": 3140,
      "train_speed(iter/s)": 1.611936
    },
    {
      "acc": 0.61641026,
      "epoch": 0.0797818366311517,
      "grad_norm": 5.71875,
      "learning_rate": 7.978183663115172e-06,
      "loss": 1.91095695,
      "memory(GiB)": 58.86,
      "step": 3145,
      "train_speed(iter/s)": 1.612258
    },
    {
      "acc": 0.6256465,
      "epoch": 0.07990867579908675,
      "grad_norm": 4.71875,
      "learning_rate": 7.990867579908676e-06,
      "loss": 1.82138786,
      "memory(GiB)": 58.86,
      "step": 3150,
      "train_speed(iter/s)": 1.61257
    },
    {
      "acc": 0.61393795,
      "epoch": 0.08003551496702181,
      "grad_norm": 5.625,
      "learning_rate": 8.003551496702182e-06,
      "loss": 1.85157642,
      "memory(GiB)": 58.86,
      "step": 3155,
      "train_speed(iter/s)": 1.612871
    },
    {
      "acc": 0.62110429,
      "epoch": 0.08016235413495687,
      "grad_norm": 5.40625,
      "learning_rate": 8.016235413495688e-06,
      "loss": 1.79541931,
      "memory(GiB)": 58.86,
      "step": 3160,
      "train_speed(iter/s)": 1.613198
    },
    {
      "acc": 0.61994252,
      "epoch": 0.08028919330289193,
      "grad_norm": 7.375,
      "learning_rate": 8.028919330289194e-06,
      "loss": 1.91337032,
      "memory(GiB)": 58.86,
      "step": 3165,
      "train_speed(iter/s)": 1.61349
    },
    {
      "acc": 0.63742933,
      "epoch": 0.08041603247082699,
      "grad_norm": 5.0,
      "learning_rate": 8.0416032470827e-06,
      "loss": 1.78433266,
      "memory(GiB)": 58.86,
      "step": 3170,
      "train_speed(iter/s)": 1.613807
    },
    {
      "acc": 0.61900368,
      "epoch": 0.08054287163876205,
      "grad_norm": 5.5,
      "learning_rate": 8.054287163876206e-06,
      "loss": 1.82017097,
      "memory(GiB)": 58.86,
      "step": 3175,
      "train_speed(iter/s)": 1.61411
    },
    {
      "acc": 0.62221527,
      "epoch": 0.0806697108066971,
      "grad_norm": 5.6875,
      "learning_rate": 8.066971080669712e-06,
      "loss": 1.80164413,
      "memory(GiB)": 58.86,
      "step": 3180,
      "train_speed(iter/s)": 1.614429
    },
    {
      "acc": 0.61236415,
      "epoch": 0.08079654997463216,
      "grad_norm": 7.34375,
      "learning_rate": 8.079654997463218e-06,
      "loss": 1.87927094,
      "memory(GiB)": 58.86,
      "step": 3185,
      "train_speed(iter/s)": 1.614761
    },
    {
      "acc": 0.62563963,
      "epoch": 0.08092338914256722,
      "grad_norm": 6.84375,
      "learning_rate": 8.092338914256724e-06,
      "loss": 1.78968449,
      "memory(GiB)": 58.86,
      "step": 3190,
      "train_speed(iter/s)": 1.615053
    },
    {
      "acc": 0.61603351,
      "epoch": 0.08105022831050228,
      "grad_norm": 6.625,
      "learning_rate": 8.105022831050228e-06,
      "loss": 1.75521469,
      "memory(GiB)": 58.86,
      "step": 3195,
      "train_speed(iter/s)": 1.615353
    },
    {
      "acc": 0.63678412,
      "epoch": 0.08117706747843734,
      "grad_norm": 5.59375,
      "learning_rate": 8.117706747843734e-06,
      "loss": 1.75136757,
      "memory(GiB)": 58.86,
      "step": 3200,
      "train_speed(iter/s)": 1.61565
    },
    {
      "acc": 0.6231986,
      "epoch": 0.0813039066463724,
      "grad_norm": 6.09375,
      "learning_rate": 8.130390664637242e-06,
      "loss": 1.77493153,
      "memory(GiB)": 58.86,
      "step": 3205,
      "train_speed(iter/s)": 1.615954
    },
    {
      "acc": 0.63598914,
      "epoch": 0.08143074581430745,
      "grad_norm": 5.46875,
      "learning_rate": 8.143074581430746e-06,
      "loss": 1.75733223,
      "memory(GiB)": 58.86,
      "step": 3210,
      "train_speed(iter/s)": 1.616264
    },
    {
      "acc": 0.63642306,
      "epoch": 0.08155758498224251,
      "grad_norm": 5.78125,
      "learning_rate": 8.155758498224252e-06,
      "loss": 1.78879604,
      "memory(GiB)": 58.86,
      "step": 3215,
      "train_speed(iter/s)": 1.616574
    },
    {
      "acc": 0.63304596,
      "epoch": 0.08168442415017757,
      "grad_norm": 6.5,
      "learning_rate": 8.168442415017758e-06,
      "loss": 1.7665226,
      "memory(GiB)": 58.86,
      "step": 3220,
      "train_speed(iter/s)": 1.616879
    },
    {
      "acc": 0.62088284,
      "epoch": 0.08181126331811263,
      "grad_norm": 6.375,
      "learning_rate": 8.181126331811264e-06,
      "loss": 1.85610085,
      "memory(GiB)": 58.86,
      "step": 3225,
      "train_speed(iter/s)": 1.617189
    },
    {
      "acc": 0.62163486,
      "epoch": 0.08193810248604769,
      "grad_norm": 5.1875,
      "learning_rate": 8.19381024860477e-06,
      "loss": 1.76511917,
      "memory(GiB)": 58.86,
      "step": 3230,
      "train_speed(iter/s)": 1.617504
    },
    {
      "acc": 0.62488856,
      "epoch": 0.08206494165398275,
      "grad_norm": 5.59375,
      "learning_rate": 8.206494165398276e-06,
      "loss": 1.69906731,
      "memory(GiB)": 58.86,
      "step": 3235,
      "train_speed(iter/s)": 1.617814
    },
    {
      "acc": 0.62290168,
      "epoch": 0.0821917808219178,
      "grad_norm": 6.6875,
      "learning_rate": 8.219178082191782e-06,
      "loss": 1.81975422,
      "memory(GiB)": 58.86,
      "step": 3240,
      "train_speed(iter/s)": 1.61812
    },
    {
      "acc": 0.62643986,
      "epoch": 0.08231861998985286,
      "grad_norm": 5.71875,
      "learning_rate": 8.231861998985288e-06,
      "loss": 1.77693825,
      "memory(GiB)": 58.86,
      "step": 3245,
      "train_speed(iter/s)": 1.618406
    },
    {
      "acc": 0.63714232,
      "epoch": 0.08244545915778792,
      "grad_norm": 5.75,
      "learning_rate": 8.244545915778794e-06,
      "loss": 1.71134491,
      "memory(GiB)": 58.86,
      "step": 3250,
      "train_speed(iter/s)": 1.618702
    },
    {
      "acc": 0.61583309,
      "epoch": 0.08257229832572298,
      "grad_norm": 4.90625,
      "learning_rate": 8.2572298325723e-06,
      "loss": 1.83567562,
      "memory(GiB)": 58.86,
      "step": 3255,
      "train_speed(iter/s)": 1.61898
    },
    {
      "acc": 0.6344367,
      "epoch": 0.08269913749365804,
      "grad_norm": 5.59375,
      "learning_rate": 8.269913749365804e-06,
      "loss": 1.73675213,
      "memory(GiB)": 58.86,
      "step": 3260,
      "train_speed(iter/s)": 1.61927
    },
    {
      "acc": 0.61676769,
      "epoch": 0.0828259766615931,
      "grad_norm": 8.3125,
      "learning_rate": 8.282597666159312e-06,
      "loss": 1.76578865,
      "memory(GiB)": 58.86,
      "step": 3265,
      "train_speed(iter/s)": 1.619492
    },
    {
      "acc": 0.6272892,
      "epoch": 0.08295281582952815,
      "grad_norm": 6.25,
      "learning_rate": 8.295281582952816e-06,
      "loss": 1.82483997,
      "memory(GiB)": 58.86,
      "step": 3270,
      "train_speed(iter/s)": 1.619789
    },
    {
      "acc": 0.6161901,
      "epoch": 0.08307965499746321,
      "grad_norm": 6.53125,
      "learning_rate": 8.307965499746322e-06,
      "loss": 1.76700649,
      "memory(GiB)": 58.86,
      "step": 3275,
      "train_speed(iter/s)": 1.620067
    },
    {
      "acc": 0.610918,
      "epoch": 0.08320649416539827,
      "grad_norm": 5.125,
      "learning_rate": 8.320649416539828e-06,
      "loss": 1.89990501,
      "memory(GiB)": 58.86,
      "step": 3280,
      "train_speed(iter/s)": 1.620366
    },
    {
      "acc": 0.6342813,
      "epoch": 0.08333333333333333,
      "grad_norm": 5.46875,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.7141571,
      "memory(GiB)": 58.86,
      "step": 3285,
      "train_speed(iter/s)": 1.620671
    },
    {
      "acc": 0.63332539,
      "epoch": 0.08346017250126839,
      "grad_norm": 7.21875,
      "learning_rate": 8.34601725012684e-06,
      "loss": 1.6897047,
      "memory(GiB)": 58.86,
      "step": 3290,
      "train_speed(iter/s)": 1.620973
    },
    {
      "acc": 0.6279213,
      "epoch": 0.08358701166920345,
      "grad_norm": 8.5625,
      "learning_rate": 8.358701166920346e-06,
      "loss": 1.83149261,
      "memory(GiB)": 58.86,
      "step": 3295,
      "train_speed(iter/s)": 1.621263
    },
    {
      "acc": 0.60932512,
      "epoch": 0.0837138508371385,
      "grad_norm": 5.875,
      "learning_rate": 8.371385083713852e-06,
      "loss": 1.89715729,
      "memory(GiB)": 58.86,
      "step": 3300,
      "train_speed(iter/s)": 1.621549
    },
    {
      "acc": 0.61546173,
      "epoch": 0.08384069000507356,
      "grad_norm": 6.71875,
      "learning_rate": 8.384069000507358e-06,
      "loss": 1.84220676,
      "memory(GiB)": 58.86,
      "step": 3305,
      "train_speed(iter/s)": 1.621842
    },
    {
      "acc": 0.61634483,
      "epoch": 0.08396752917300862,
      "grad_norm": 5.15625,
      "learning_rate": 8.396752917300864e-06,
      "loss": 1.82034645,
      "memory(GiB)": 58.86,
      "step": 3310,
      "train_speed(iter/s)": 1.622119
    },
    {
      "acc": 0.61833067,
      "epoch": 0.08409436834094368,
      "grad_norm": 5.71875,
      "learning_rate": 8.40943683409437e-06,
      "loss": 1.78616543,
      "memory(GiB)": 58.86,
      "step": 3315,
      "train_speed(iter/s)": 1.622398
    },
    {
      "acc": 0.62271051,
      "epoch": 0.08422120750887874,
      "grad_norm": 6.46875,
      "learning_rate": 8.422120750887874e-06,
      "loss": 1.83358517,
      "memory(GiB)": 58.86,
      "step": 3320,
      "train_speed(iter/s)": 1.622639
    },
    {
      "acc": 0.62470541,
      "epoch": 0.0843480466768138,
      "grad_norm": 5.59375,
      "learning_rate": 8.434804667681381e-06,
      "loss": 1.74936008,
      "memory(GiB)": 58.86,
      "step": 3325,
      "train_speed(iter/s)": 1.62293
    },
    {
      "acc": 0.62520065,
      "epoch": 0.08447488584474885,
      "grad_norm": 6.28125,
      "learning_rate": 8.447488584474887e-06,
      "loss": 1.81033707,
      "memory(GiB)": 58.86,
      "step": 3330,
      "train_speed(iter/s)": 1.623192
    },
    {
      "acc": 0.6237464,
      "epoch": 0.08460172501268391,
      "grad_norm": 7.375,
      "learning_rate": 8.460172501268392e-06,
      "loss": 1.81594048,
      "memory(GiB)": 58.86,
      "step": 3335,
      "train_speed(iter/s)": 1.62346
    },
    {
      "acc": 0.62431679,
      "epoch": 0.08472856418061897,
      "grad_norm": 5.15625,
      "learning_rate": 8.472856418061898e-06,
      "loss": 1.77585602,
      "memory(GiB)": 58.86,
      "step": 3340,
      "train_speed(iter/s)": 1.623721
    },
    {
      "acc": 0.64794264,
      "epoch": 0.08485540334855403,
      "grad_norm": 5.71875,
      "learning_rate": 8.485540334855404e-06,
      "loss": 1.729105,
      "memory(GiB)": 58.86,
      "step": 3345,
      "train_speed(iter/s)": 1.624004
    },
    {
      "acc": 0.64635601,
      "epoch": 0.08498224251648909,
      "grad_norm": 4.4375,
      "learning_rate": 8.49822425164891e-06,
      "loss": 1.70486851,
      "memory(GiB)": 58.86,
      "step": 3350,
      "train_speed(iter/s)": 1.62426
    },
    {
      "acc": 0.62374363,
      "epoch": 0.08510908168442415,
      "grad_norm": 5.15625,
      "learning_rate": 8.510908168442416e-06,
      "loss": 1.81973915,
      "memory(GiB)": 58.86,
      "step": 3355,
      "train_speed(iter/s)": 1.624535
    },
    {
      "acc": 0.60954399,
      "epoch": 0.0852359208523592,
      "grad_norm": 6.15625,
      "learning_rate": 8.523592085235922e-06,
      "loss": 1.87591019,
      "memory(GiB)": 58.86,
      "step": 3360,
      "train_speed(iter/s)": 1.62481
    },
    {
      "acc": 0.61560974,
      "epoch": 0.08536276002029426,
      "grad_norm": 5.28125,
      "learning_rate": 8.536276002029428e-06,
      "loss": 1.78173676,
      "memory(GiB)": 58.86,
      "step": 3365,
      "train_speed(iter/s)": 1.625082
    },
    {
      "acc": 0.62815838,
      "epoch": 0.08548959918822932,
      "grad_norm": 7.25,
      "learning_rate": 8.548959918822933e-06,
      "loss": 1.86471519,
      "memory(GiB)": 58.86,
      "step": 3370,
      "train_speed(iter/s)": 1.625369
    },
    {
      "acc": 0.61993814,
      "epoch": 0.08561643835616438,
      "grad_norm": 5.59375,
      "learning_rate": 8.56164383561644e-06,
      "loss": 1.8225563,
      "memory(GiB)": 58.86,
      "step": 3375,
      "train_speed(iter/s)": 1.625629
    },
    {
      "acc": 0.63935537,
      "epoch": 0.08574327752409944,
      "grad_norm": 7.1875,
      "learning_rate": 8.574327752409944e-06,
      "loss": 1.7402483,
      "memory(GiB)": 58.86,
      "step": 3380,
      "train_speed(iter/s)": 1.625892
    },
    {
      "acc": 0.61767402,
      "epoch": 0.0858701166920345,
      "grad_norm": 6.28125,
      "learning_rate": 8.587011669203451e-06,
      "loss": 1.80435104,
      "memory(GiB)": 58.86,
      "step": 3385,
      "train_speed(iter/s)": 1.62616
    },
    {
      "acc": 0.64225984,
      "epoch": 0.08599695585996955,
      "grad_norm": 6.59375,
      "learning_rate": 8.599695585996957e-06,
      "loss": 1.71389217,
      "memory(GiB)": 58.86,
      "step": 3390,
      "train_speed(iter/s)": 1.626428
    },
    {
      "acc": 0.62173142,
      "epoch": 0.08612379502790461,
      "grad_norm": 6.28125,
      "learning_rate": 8.612379502790462e-06,
      "loss": 1.85017319,
      "memory(GiB)": 58.86,
      "step": 3395,
      "train_speed(iter/s)": 1.626673
    },
    {
      "acc": 0.63053288,
      "epoch": 0.08625063419583967,
      "grad_norm": 6.0,
      "learning_rate": 8.625063419583968e-06,
      "loss": 1.80282078,
      "memory(GiB)": 58.86,
      "step": 3400,
      "train_speed(iter/s)": 1.626941
    },
    {
      "acc": 0.63656359,
      "epoch": 0.08637747336377473,
      "grad_norm": 6.15625,
      "learning_rate": 8.637747336377475e-06,
      "loss": 1.81014576,
      "memory(GiB)": 58.86,
      "step": 3405,
      "train_speed(iter/s)": 1.627219
    },
    {
      "acc": 0.63236113,
      "epoch": 0.08650431253170979,
      "grad_norm": 6.625,
      "learning_rate": 8.65043125317098e-06,
      "loss": 1.8170887,
      "memory(GiB)": 58.86,
      "step": 3410,
      "train_speed(iter/s)": 1.627505
    },
    {
      "acc": 0.6143199,
      "epoch": 0.08663115169964485,
      "grad_norm": 6.59375,
      "learning_rate": 8.663115169964485e-06,
      "loss": 1.77595348,
      "memory(GiB)": 58.86,
      "step": 3415,
      "train_speed(iter/s)": 1.627779
    },
    {
      "acc": 0.60883551,
      "epoch": 0.0867579908675799,
      "grad_norm": 7.96875,
      "learning_rate": 8.675799086757991e-06,
      "loss": 1.87337475,
      "memory(GiB)": 58.86,
      "step": 3420,
      "train_speed(iter/s)": 1.628045
    },
    {
      "acc": 0.62656479,
      "epoch": 0.08688483003551496,
      "grad_norm": 6.46875,
      "learning_rate": 8.688483003551497e-06,
      "loss": 1.85432644,
      "memory(GiB)": 58.86,
      "step": 3425,
      "train_speed(iter/s)": 1.628306
    },
    {
      "acc": 0.63665709,
      "epoch": 0.08701166920345002,
      "grad_norm": 5.53125,
      "learning_rate": 8.701166920345003e-06,
      "loss": 1.76383171,
      "memory(GiB)": 58.86,
      "step": 3430,
      "train_speed(iter/s)": 1.62857
    },
    {
      "acc": 0.62604742,
      "epoch": 0.08713850837138508,
      "grad_norm": 5.71875,
      "learning_rate": 8.71385083713851e-06,
      "loss": 1.82312603,
      "memory(GiB)": 67.97,
      "step": 3435,
      "train_speed(iter/s)": 1.628831
    },
    {
      "acc": 0.6141346,
      "epoch": 0.08726534753932014,
      "grad_norm": 6.375,
      "learning_rate": 8.726534753932014e-06,
      "loss": 1.90559444,
      "memory(GiB)": 67.97,
      "step": 3440,
      "train_speed(iter/s)": 1.629102
    },
    {
      "acc": 0.62872844,
      "epoch": 0.0873921867072552,
      "grad_norm": 6.6875,
      "learning_rate": 8.739218670725521e-06,
      "loss": 1.76434479,
      "memory(GiB)": 67.97,
      "step": 3445,
      "train_speed(iter/s)": 1.629372
    },
    {
      "acc": 0.6223835,
      "epoch": 0.08751902587519025,
      "grad_norm": 5.9375,
      "learning_rate": 8.751902587519027e-06,
      "loss": 1.7607605,
      "memory(GiB)": 67.97,
      "step": 3450,
      "train_speed(iter/s)": 1.629637
    },
    {
      "acc": 0.62013903,
      "epoch": 0.08764586504312531,
      "grad_norm": 6.21875,
      "learning_rate": 8.764586504312532e-06,
      "loss": 1.87243843,
      "memory(GiB)": 67.97,
      "step": 3455,
      "train_speed(iter/s)": 1.629888
    },
    {
      "acc": 0.64138584,
      "epoch": 0.08777270421106037,
      "grad_norm": 5.71875,
      "learning_rate": 8.777270421106037e-06,
      "loss": 1.82841682,
      "memory(GiB)": 67.97,
      "step": 3460,
      "train_speed(iter/s)": 1.630133
    },
    {
      "acc": 0.63943453,
      "epoch": 0.08789954337899543,
      "grad_norm": 5.53125,
      "learning_rate": 8.789954337899545e-06,
      "loss": 1.76984653,
      "memory(GiB)": 67.97,
      "step": 3465,
      "train_speed(iter/s)": 1.630394
    },
    {
      "acc": 0.62812405,
      "epoch": 0.08802638254693049,
      "grad_norm": 6.3125,
      "learning_rate": 8.80263825469305e-06,
      "loss": 1.78545189,
      "memory(GiB)": 67.97,
      "step": 3470,
      "train_speed(iter/s)": 1.630663
    },
    {
      "acc": 0.64027443,
      "epoch": 0.08815322171486555,
      "grad_norm": 5.46875,
      "learning_rate": 8.815322171486555e-06,
      "loss": 1.7096199,
      "memory(GiB)": 67.97,
      "step": 3475,
      "train_speed(iter/s)": 1.630922
    },
    {
      "acc": 0.63905883,
      "epoch": 0.0882800608828006,
      "grad_norm": 6.09375,
      "learning_rate": 8.828006088280061e-06,
      "loss": 1.79863663,
      "memory(GiB)": 67.97,
      "step": 3480,
      "train_speed(iter/s)": 1.631198
    },
    {
      "acc": 0.62990875,
      "epoch": 0.08840690005073566,
      "grad_norm": 5.625,
      "learning_rate": 8.840690005073567e-06,
      "loss": 1.77944374,
      "memory(GiB)": 67.97,
      "step": 3485,
      "train_speed(iter/s)": 1.631441
    },
    {
      "acc": 0.61861358,
      "epoch": 0.08853373921867072,
      "grad_norm": 5.90625,
      "learning_rate": 8.853373921867073e-06,
      "loss": 1.75925274,
      "memory(GiB)": 67.97,
      "step": 3490,
      "train_speed(iter/s)": 1.631688
    },
    {
      "acc": 0.63133311,
      "epoch": 0.08866057838660578,
      "grad_norm": 5.4375,
      "learning_rate": 8.86605783866058e-06,
      "loss": 1.81406441,
      "memory(GiB)": 67.97,
      "step": 3495,
      "train_speed(iter/s)": 1.631948
    },
    {
      "acc": 0.63282757,
      "epoch": 0.08878741755454084,
      "grad_norm": 5.0625,
      "learning_rate": 8.878741755454085e-06,
      "loss": 1.73094482,
      "memory(GiB)": 67.97,
      "step": 3500,
      "train_speed(iter/s)": 1.632196
    },
    {
      "acc": 0.62975836,
      "epoch": 0.0889142567224759,
      "grad_norm": 4.875,
      "learning_rate": 8.891425672247591e-06,
      "loss": 1.73807812,
      "memory(GiB)": 67.97,
      "step": 3505,
      "train_speed(iter/s)": 1.632456
    },
    {
      "acc": 0.62790809,
      "epoch": 0.08904109589041095,
      "grad_norm": 5.5,
      "learning_rate": 8.904109589041097e-06,
      "loss": 1.79924641,
      "memory(GiB)": 67.97,
      "step": 3510,
      "train_speed(iter/s)": 1.632705
    },
    {
      "acc": 0.61818233,
      "epoch": 0.08916793505834601,
      "grad_norm": 8.5,
      "learning_rate": 8.916793505834601e-06,
      "loss": 1.83528461,
      "memory(GiB)": 67.97,
      "step": 3515,
      "train_speed(iter/s)": 1.632974
    },
    {
      "acc": 0.63320665,
      "epoch": 0.08929477422628107,
      "grad_norm": 5.1875,
      "learning_rate": 8.929477422628107e-06,
      "loss": 1.80401611,
      "memory(GiB)": 67.97,
      "step": 3520,
      "train_speed(iter/s)": 1.633232
    },
    {
      "acc": 0.63903503,
      "epoch": 0.08942161339421613,
      "grad_norm": 7.0,
      "learning_rate": 8.942161339421615e-06,
      "loss": 1.77746773,
      "memory(GiB)": 67.97,
      "step": 3525,
      "train_speed(iter/s)": 1.633483
    },
    {
      "acc": 0.63534675,
      "epoch": 0.08954845256215119,
      "grad_norm": 7.71875,
      "learning_rate": 8.95484525621512e-06,
      "loss": 1.77584343,
      "memory(GiB)": 67.97,
      "step": 3530,
      "train_speed(iter/s)": 1.633718
    },
    {
      "acc": 0.61750698,
      "epoch": 0.08967529173008625,
      "grad_norm": 5.71875,
      "learning_rate": 8.967529173008625e-06,
      "loss": 1.8113636,
      "memory(GiB)": 67.97,
      "step": 3535,
      "train_speed(iter/s)": 1.633956
    },
    {
      "acc": 0.62888799,
      "epoch": 0.0898021308980213,
      "grad_norm": 5.0625,
      "learning_rate": 8.980213089802131e-06,
      "loss": 1.72378292,
      "memory(GiB)": 67.97,
      "step": 3540,
      "train_speed(iter/s)": 1.634199
    },
    {
      "acc": 0.63384509,
      "epoch": 0.08992897006595636,
      "grad_norm": 5.65625,
      "learning_rate": 8.992897006595637e-06,
      "loss": 1.70481033,
      "memory(GiB)": 67.97,
      "step": 3545,
      "train_speed(iter/s)": 1.634448
    },
    {
      "acc": 0.63124871,
      "epoch": 0.09005580923389142,
      "grad_norm": 5.96875,
      "learning_rate": 9.005580923389143e-06,
      "loss": 1.74125862,
      "memory(GiB)": 67.97,
      "step": 3550,
      "train_speed(iter/s)": 1.634689
    },
    {
      "acc": 0.61392713,
      "epoch": 0.09018264840182648,
      "grad_norm": 5.46875,
      "learning_rate": 9.01826484018265e-06,
      "loss": 1.80518646,
      "memory(GiB)": 67.97,
      "step": 3555,
      "train_speed(iter/s)": 1.63494
    },
    {
      "acc": 0.61655688,
      "epoch": 0.09030948756976154,
      "grad_norm": 8.0,
      "learning_rate": 9.030948756976155e-06,
      "loss": 1.87416306,
      "memory(GiB)": 67.97,
      "step": 3560,
      "train_speed(iter/s)": 1.635193
    },
    {
      "acc": 0.62979341,
      "epoch": 0.0904363267376966,
      "grad_norm": 5.75,
      "learning_rate": 9.043632673769661e-06,
      "loss": 1.79071503,
      "memory(GiB)": 67.97,
      "step": 3565,
      "train_speed(iter/s)": 1.635452
    },
    {
      "acc": 0.62730169,
      "epoch": 0.09056316590563165,
      "grad_norm": 5.9375,
      "learning_rate": 9.056316590563167e-06,
      "loss": 1.73049717,
      "memory(GiB)": 67.97,
      "step": 3570,
      "train_speed(iter/s)": 1.63571
    },
    {
      "acc": 0.62824545,
      "epoch": 0.09069000507356671,
      "grad_norm": 5.09375,
      "learning_rate": 9.069000507356673e-06,
      "loss": 1.73457069,
      "memory(GiB)": 67.97,
      "step": 3575,
      "train_speed(iter/s)": 1.635959
    },
    {
      "acc": 0.63004065,
      "epoch": 0.09081684424150177,
      "grad_norm": 5.71875,
      "learning_rate": 9.081684424150177e-06,
      "loss": 1.76967869,
      "memory(GiB)": 67.97,
      "step": 3580,
      "train_speed(iter/s)": 1.636201
    },
    {
      "acc": 0.63578882,
      "epoch": 0.09094368340943683,
      "grad_norm": 5.25,
      "learning_rate": 9.094368340943685e-06,
      "loss": 1.70772362,
      "memory(GiB)": 67.97,
      "step": 3585,
      "train_speed(iter/s)": 1.63646
    },
    {
      "acc": 0.63399067,
      "epoch": 0.09107052257737189,
      "grad_norm": 8.1875,
      "learning_rate": 9.10705225773719e-06,
      "loss": 1.69498749,
      "memory(GiB)": 67.97,
      "step": 3590,
      "train_speed(iter/s)": 1.636722
    },
    {
      "acc": 0.62841349,
      "epoch": 0.09119736174530695,
      "grad_norm": 5.90625,
      "learning_rate": 9.119736174530695e-06,
      "loss": 1.82112465,
      "memory(GiB)": 67.97,
      "step": 3595,
      "train_speed(iter/s)": 1.636979
    },
    {
      "acc": 0.64118824,
      "epoch": 0.091324200913242,
      "grad_norm": 6.09375,
      "learning_rate": 9.132420091324201e-06,
      "loss": 1.7731266,
      "memory(GiB)": 67.97,
      "step": 3600,
      "train_speed(iter/s)": 1.637212
    },
    {
      "acc": 0.64473572,
      "epoch": 0.09145104008117706,
      "grad_norm": 6.4375,
      "learning_rate": 9.145104008117707e-06,
      "loss": 1.71745529,
      "memory(GiB)": 67.97,
      "step": 3605,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.62576127,
      "epoch": 0.09157787924911212,
      "grad_norm": 4.78125,
      "learning_rate": 9.157787924911213e-06,
      "loss": 1.83209114,
      "memory(GiB)": 67.97,
      "step": 3610,
      "train_speed(iter/s)": 1.63769
    },
    {
      "acc": 0.63057089,
      "epoch": 0.09170471841704718,
      "grad_norm": 4.90625,
      "learning_rate": 9.170471841704719e-06,
      "loss": 1.81908989,
      "memory(GiB)": 67.97,
      "step": 3615,
      "train_speed(iter/s)": 1.637931
    },
    {
      "acc": 0.61526289,
      "epoch": 0.09183155758498224,
      "grad_norm": 7.125,
      "learning_rate": 9.183155758498225e-06,
      "loss": 1.81995583,
      "memory(GiB)": 67.97,
      "step": 3620,
      "train_speed(iter/s)": 1.638178
    },
    {
      "acc": 0.62104368,
      "epoch": 0.0919583967529173,
      "grad_norm": 5.375,
      "learning_rate": 9.195839675291731e-06,
      "loss": 1.83020267,
      "memory(GiB)": 67.97,
      "step": 3625,
      "train_speed(iter/s)": 1.638414
    },
    {
      "acc": 0.63465476,
      "epoch": 0.09208523592085235,
      "grad_norm": 5.9375,
      "learning_rate": 9.208523592085237e-06,
      "loss": 1.76404343,
      "memory(GiB)": 67.97,
      "step": 3630,
      "train_speed(iter/s)": 1.638661
    },
    {
      "acc": 0.64479656,
      "epoch": 0.09221207508878741,
      "grad_norm": 6.5625,
      "learning_rate": 9.221207508878743e-06,
      "loss": 1.71783867,
      "memory(GiB)": 67.97,
      "step": 3635,
      "train_speed(iter/s)": 1.638906
    },
    {
      "acc": 0.62756414,
      "epoch": 0.09233891425672247,
      "grad_norm": 7.125,
      "learning_rate": 9.233891425672247e-06,
      "loss": 1.81249371,
      "memory(GiB)": 67.97,
      "step": 3640,
      "train_speed(iter/s)": 1.639138
    },
    {
      "acc": 0.63336759,
      "epoch": 0.09246575342465753,
      "grad_norm": 6.5,
      "learning_rate": 9.246575342465755e-06,
      "loss": 1.78633366,
      "memory(GiB)": 67.97,
      "step": 3645,
      "train_speed(iter/s)": 1.639376
    },
    {
      "acc": 0.6273036,
      "epoch": 0.09259259259259259,
      "grad_norm": 5.8125,
      "learning_rate": 9.25925925925926e-06,
      "loss": 1.81920395,
      "memory(GiB)": 67.97,
      "step": 3650,
      "train_speed(iter/s)": 1.639596
    },
    {
      "acc": 0.62608423,
      "epoch": 0.09271943176052765,
      "grad_norm": 6.5,
      "learning_rate": 9.271943176052765e-06,
      "loss": 1.75691261,
      "memory(GiB)": 67.97,
      "step": 3655,
      "train_speed(iter/s)": 1.639835
    },
    {
      "acc": 0.6471118,
      "epoch": 0.0928462709284627,
      "grad_norm": 6.46875,
      "learning_rate": 9.284627092846271e-06,
      "loss": 1.73089218,
      "memory(GiB)": 67.97,
      "step": 3660,
      "train_speed(iter/s)": 1.640077
    },
    {
      "acc": 0.6465004,
      "epoch": 0.09297311009639776,
      "grad_norm": 6.6875,
      "learning_rate": 9.297311009639777e-06,
      "loss": 1.79161148,
      "memory(GiB)": 67.97,
      "step": 3665,
      "train_speed(iter/s)": 1.640319
    },
    {
      "acc": 0.63040805,
      "epoch": 0.09309994926433282,
      "grad_norm": 6.21875,
      "learning_rate": 9.309994926433283e-06,
      "loss": 1.85037575,
      "memory(GiB)": 67.97,
      "step": 3670,
      "train_speed(iter/s)": 1.640551
    },
    {
      "acc": 0.61671419,
      "epoch": 0.09322678843226788,
      "grad_norm": 6.375,
      "learning_rate": 9.322678843226789e-06,
      "loss": 1.85424042,
      "memory(GiB)": 67.97,
      "step": 3675,
      "train_speed(iter/s)": 1.64081
    },
    {
      "acc": 0.62683759,
      "epoch": 0.09335362760020294,
      "grad_norm": 8.3125,
      "learning_rate": 9.335362760020295e-06,
      "loss": 1.86326561,
      "memory(GiB)": 67.97,
      "step": 3680,
      "train_speed(iter/s)": 1.641043
    },
    {
      "acc": 0.6272747,
      "epoch": 0.093480466768138,
      "grad_norm": 5.96875,
      "learning_rate": 9.348046676813801e-06,
      "loss": 1.75050812,
      "memory(GiB)": 67.97,
      "step": 3685,
      "train_speed(iter/s)": 1.641284
    },
    {
      "acc": 0.6240057,
      "epoch": 0.09360730593607305,
      "grad_norm": 4.625,
      "learning_rate": 9.360730593607307e-06,
      "loss": 1.79427795,
      "memory(GiB)": 67.97,
      "step": 3690,
      "train_speed(iter/s)": 1.641524
    },
    {
      "acc": 0.64045963,
      "epoch": 0.09373414510400811,
      "grad_norm": 6.125,
      "learning_rate": 9.373414510400813e-06,
      "loss": 1.68342381,
      "memory(GiB)": 67.97,
      "step": 3695,
      "train_speed(iter/s)": 1.641765
    },
    {
      "acc": 0.61470656,
      "epoch": 0.09386098427194317,
      "grad_norm": 7.375,
      "learning_rate": 9.386098427194317e-06,
      "loss": 1.89201355,
      "memory(GiB)": 67.97,
      "step": 3700,
      "train_speed(iter/s)": 1.642002
    },
    {
      "acc": 0.61346741,
      "epoch": 0.09398782343987823,
      "grad_norm": 5.625,
      "learning_rate": 9.398782343987825e-06,
      "loss": 1.87396717,
      "memory(GiB)": 67.97,
      "step": 3705,
      "train_speed(iter/s)": 1.642228
    },
    {
      "acc": 0.63009262,
      "epoch": 0.09411466260781329,
      "grad_norm": 6.15625,
      "learning_rate": 9.41146626078133e-06,
      "loss": 1.72186909,
      "memory(GiB)": 67.97,
      "step": 3710,
      "train_speed(iter/s)": 1.642475
    },
    {
      "acc": 0.64259787,
      "epoch": 0.09424150177574835,
      "grad_norm": 7.65625,
      "learning_rate": 9.424150177574835e-06,
      "loss": 1.76672134,
      "memory(GiB)": 67.97,
      "step": 3715,
      "train_speed(iter/s)": 1.642712
    },
    {
      "acc": 0.61804214,
      "epoch": 0.0943683409436834,
      "grad_norm": 7.59375,
      "learning_rate": 9.436834094368341e-06,
      "loss": 1.79548759,
      "memory(GiB)": 67.97,
      "step": 3720,
      "train_speed(iter/s)": 1.642951
    },
    {
      "acc": 0.647159,
      "epoch": 0.09449518011161846,
      "grad_norm": 7.28125,
      "learning_rate": 9.449518011161849e-06,
      "loss": 1.6965374,
      "memory(GiB)": 67.97,
      "step": 3725,
      "train_speed(iter/s)": 1.643183
    },
    {
      "acc": 0.64615951,
      "epoch": 0.09462201927955352,
      "grad_norm": 6.0,
      "learning_rate": 9.462201927955353e-06,
      "loss": 1.69098835,
      "memory(GiB)": 67.97,
      "step": 3730,
      "train_speed(iter/s)": 1.643423
    },
    {
      "acc": 0.62432489,
      "epoch": 0.09474885844748858,
      "grad_norm": 5.34375,
      "learning_rate": 9.474885844748859e-06,
      "loss": 1.80043926,
      "memory(GiB)": 67.97,
      "step": 3735,
      "train_speed(iter/s)": 1.643661
    },
    {
      "acc": 0.6286046,
      "epoch": 0.09487569761542364,
      "grad_norm": 5.53125,
      "learning_rate": 9.487569761542365e-06,
      "loss": 1.75099373,
      "memory(GiB)": 67.97,
      "step": 3740,
      "train_speed(iter/s)": 1.643905
    },
    {
      "acc": 0.6401032,
      "epoch": 0.0950025367833587,
      "grad_norm": 5.40625,
      "learning_rate": 9.50025367833587e-06,
      "loss": 1.75601978,
      "memory(GiB)": 67.97,
      "step": 3745,
      "train_speed(iter/s)": 1.644131
    },
    {
      "acc": 0.622264,
      "epoch": 0.09512937595129375,
      "grad_norm": 5.1875,
      "learning_rate": 9.512937595129377e-06,
      "loss": 1.78772736,
      "memory(GiB)": 67.97,
      "step": 3750,
      "train_speed(iter/s)": 1.64438
    },
    {
      "acc": 0.63415251,
      "epoch": 0.09525621511922881,
      "grad_norm": 7.0,
      "learning_rate": 9.525621511922883e-06,
      "loss": 1.83068581,
      "memory(GiB)": 67.97,
      "step": 3755,
      "train_speed(iter/s)": 1.644605
    },
    {
      "acc": 0.61628761,
      "epoch": 0.09538305428716387,
      "grad_norm": 6.21875,
      "learning_rate": 9.538305428716389e-06,
      "loss": 1.82850704,
      "memory(GiB)": 67.97,
      "step": 3760,
      "train_speed(iter/s)": 1.644837
    },
    {
      "acc": 0.63104143,
      "epoch": 0.09550989345509893,
      "grad_norm": 6.125,
      "learning_rate": 9.550989345509895e-06,
      "loss": 1.81773071,
      "memory(GiB)": 67.97,
      "step": 3765,
      "train_speed(iter/s)": 1.645072
    },
    {
      "acc": 0.62594171,
      "epoch": 0.09563673262303399,
      "grad_norm": 5.9375,
      "learning_rate": 9.5636732623034e-06,
      "loss": 1.80854778,
      "memory(GiB)": 67.97,
      "step": 3770,
      "train_speed(iter/s)": 1.645287
    },
    {
      "acc": 0.63547354,
      "epoch": 0.09576357179096905,
      "grad_norm": 5.6875,
      "learning_rate": 9.576357179096905e-06,
      "loss": 1.73742657,
      "memory(GiB)": 67.97,
      "step": 3775,
      "train_speed(iter/s)": 1.645499
    },
    {
      "acc": 0.6176383,
      "epoch": 0.0958904109589041,
      "grad_norm": 7.84375,
      "learning_rate": 9.589041095890411e-06,
      "loss": 1.84551468,
      "memory(GiB)": 67.97,
      "step": 3780,
      "train_speed(iter/s)": 1.645712
    },
    {
      "acc": 0.6130053,
      "epoch": 0.09601725012683916,
      "grad_norm": 5.71875,
      "learning_rate": 9.601725012683919e-06,
      "loss": 1.84195786,
      "memory(GiB)": 67.97,
      "step": 3785,
      "train_speed(iter/s)": 1.645931
    },
    {
      "acc": 0.63320622,
      "epoch": 0.09614408929477422,
      "grad_norm": 6.0625,
      "learning_rate": 9.614408929477423e-06,
      "loss": 1.7915554,
      "memory(GiB)": 67.97,
      "step": 3790,
      "train_speed(iter/s)": 1.646163
    },
    {
      "acc": 0.61723895,
      "epoch": 0.09627092846270928,
      "grad_norm": 4.8125,
      "learning_rate": 9.627092846270929e-06,
      "loss": 1.82622223,
      "memory(GiB)": 67.97,
      "step": 3795,
      "train_speed(iter/s)": 1.646367
    },
    {
      "acc": 0.61042767,
      "epoch": 0.09639776763064434,
      "grad_norm": 6.0625,
      "learning_rate": 9.639776763064435e-06,
      "loss": 1.83352184,
      "memory(GiB)": 67.97,
      "step": 3800,
      "train_speed(iter/s)": 1.646589
    },
    {
      "acc": 0.63187056,
      "epoch": 0.0965246067985794,
      "grad_norm": 4.625,
      "learning_rate": 9.65246067985794e-06,
      "loss": 1.82001038,
      "memory(GiB)": 67.97,
      "step": 3805,
      "train_speed(iter/s)": 1.646788
    },
    {
      "acc": 0.6299325,
      "epoch": 0.09665144596651445,
      "grad_norm": 5.5,
      "learning_rate": 9.665144596651447e-06,
      "loss": 1.72859917,
      "memory(GiB)": 67.97,
      "step": 3810,
      "train_speed(iter/s)": 1.647014
    },
    {
      "acc": 0.59229765,
      "epoch": 0.09677828513444951,
      "grad_norm": 7.96875,
      "learning_rate": 9.677828513444953e-06,
      "loss": 1.85296059,
      "memory(GiB)": 67.97,
      "step": 3815,
      "train_speed(iter/s)": 1.647241
    },
    {
      "acc": 0.61764145,
      "epoch": 0.09690512430238457,
      "grad_norm": 7.4375,
      "learning_rate": 9.690512430238459e-06,
      "loss": 1.87241898,
      "memory(GiB)": 67.97,
      "step": 3820,
      "train_speed(iter/s)": 1.647446
    },
    {
      "acc": 0.6548358,
      "epoch": 0.09703196347031963,
      "grad_norm": 5.15625,
      "learning_rate": 9.703196347031965e-06,
      "loss": 1.67850609,
      "memory(GiB)": 67.97,
      "step": 3825,
      "train_speed(iter/s)": 1.647647
    },
    {
      "acc": 0.64989834,
      "epoch": 0.09715880263825469,
      "grad_norm": 5.78125,
      "learning_rate": 9.71588026382547e-06,
      "loss": 1.70219784,
      "memory(GiB)": 67.97,
      "step": 3830,
      "train_speed(iter/s)": 1.647848
    },
    {
      "acc": 0.60952578,
      "epoch": 0.09728564180618975,
      "grad_norm": 5.71875,
      "learning_rate": 9.728564180618977e-06,
      "loss": 1.85248871,
      "memory(GiB)": 67.97,
      "step": 3835,
      "train_speed(iter/s)": 1.648052
    },
    {
      "acc": 0.62997289,
      "epoch": 0.0974124809741248,
      "grad_norm": 5.84375,
      "learning_rate": 9.74124809741248e-06,
      "loss": 1.81175308,
      "memory(GiB)": 67.97,
      "step": 3840,
      "train_speed(iter/s)": 1.64827
    },
    {
      "acc": 0.62509851,
      "epoch": 0.09753932014205986,
      "grad_norm": 5.59375,
      "learning_rate": 9.753932014205988e-06,
      "loss": 1.77528763,
      "memory(GiB)": 67.97,
      "step": 3845,
      "train_speed(iter/s)": 1.648473
    },
    {
      "acc": 0.64610667,
      "epoch": 0.09766615930999492,
      "grad_norm": 5.21875,
      "learning_rate": 9.766615930999493e-06,
      "loss": 1.7297596,
      "memory(GiB)": 67.97,
      "step": 3850,
      "train_speed(iter/s)": 1.648688
    },
    {
      "acc": 0.64355602,
      "epoch": 0.09779299847792998,
      "grad_norm": 5.4375,
      "learning_rate": 9.779299847792999e-06,
      "loss": 1.72099247,
      "memory(GiB)": 67.97,
      "step": 3855,
      "train_speed(iter/s)": 1.648881
    },
    {
      "acc": 0.63804913,
      "epoch": 0.09791983764586504,
      "grad_norm": 6.3125,
      "learning_rate": 9.791983764586505e-06,
      "loss": 1.70037785,
      "memory(GiB)": 67.97,
      "step": 3860,
      "train_speed(iter/s)": 1.649086
    },
    {
      "acc": 0.63336425,
      "epoch": 0.0980466768138001,
      "grad_norm": 6.34375,
      "learning_rate": 9.80466768138001e-06,
      "loss": 1.79189491,
      "memory(GiB)": 67.97,
      "step": 3865,
      "train_speed(iter/s)": 1.649293
    },
    {
      "acc": 0.62268028,
      "epoch": 0.09817351598173515,
      "grad_norm": 6.375,
      "learning_rate": 9.817351598173517e-06,
      "loss": 1.83122673,
      "memory(GiB)": 67.97,
      "step": 3870,
      "train_speed(iter/s)": 1.649501
    },
    {
      "acc": 0.62709866,
      "epoch": 0.09830035514967021,
      "grad_norm": 4.9375,
      "learning_rate": 9.830035514967023e-06,
      "loss": 1.82139587,
      "memory(GiB)": 67.97,
      "step": 3875,
      "train_speed(iter/s)": 1.649688
    },
    {
      "acc": 0.61775036,
      "epoch": 0.09842719431760527,
      "grad_norm": 5.59375,
      "learning_rate": 9.842719431760529e-06,
      "loss": 1.87538872,
      "memory(GiB)": 67.97,
      "step": 3880,
      "train_speed(iter/s)": 1.649888
    },
    {
      "acc": 0.63216476,
      "epoch": 0.09855403348554033,
      "grad_norm": 6.65625,
      "learning_rate": 9.855403348554034e-06,
      "loss": 1.7553051,
      "memory(GiB)": 67.97,
      "step": 3885,
      "train_speed(iter/s)": 1.650098
    },
    {
      "acc": 0.60969896,
      "epoch": 0.09868087265347539,
      "grad_norm": 5.96875,
      "learning_rate": 9.86808726534754e-06,
      "loss": 1.83891125,
      "memory(GiB)": 67.97,
      "step": 3890,
      "train_speed(iter/s)": 1.650325
    },
    {
      "acc": 0.62578363,
      "epoch": 0.09880771182141045,
      "grad_norm": 5.21875,
      "learning_rate": 9.880771182141046e-06,
      "loss": 1.85703278,
      "memory(GiB)": 67.97,
      "step": 3895,
      "train_speed(iter/s)": 1.650527
    },
    {
      "acc": 0.63930302,
      "epoch": 0.0989345509893455,
      "grad_norm": 5.09375,
      "learning_rate": 9.89345509893455e-06,
      "loss": 1.75358734,
      "memory(GiB)": 67.97,
      "step": 3900,
      "train_speed(iter/s)": 1.650726
    },
    {
      "acc": 0.63745022,
      "epoch": 0.09906139015728056,
      "grad_norm": 5.90625,
      "learning_rate": 9.906139015728058e-06,
      "loss": 1.76766968,
      "memory(GiB)": 67.97,
      "step": 3905,
      "train_speed(iter/s)": 1.650928
    },
    {
      "acc": 0.6174964,
      "epoch": 0.09918822932521562,
      "grad_norm": 5.0625,
      "learning_rate": 9.918822932521563e-06,
      "loss": 1.81346703,
      "memory(GiB)": 67.97,
      "step": 3910,
      "train_speed(iter/s)": 1.651113
    },
    {
      "acc": 0.63096094,
      "epoch": 0.09931506849315068,
      "grad_norm": 5.1875,
      "learning_rate": 9.931506849315069e-06,
      "loss": 1.79535713,
      "memory(GiB)": 67.97,
      "step": 3915,
      "train_speed(iter/s)": 1.651319
    },
    {
      "acc": 0.62452917,
      "epoch": 0.09944190766108574,
      "grad_norm": 5.28125,
      "learning_rate": 9.944190766108575e-06,
      "loss": 1.83531189,
      "memory(GiB)": 67.97,
      "step": 3920,
      "train_speed(iter/s)": 1.651524
    },
    {
      "acc": 0.63546438,
      "epoch": 0.0995687468290208,
      "grad_norm": 6.09375,
      "learning_rate": 9.95687468290208e-06,
      "loss": 1.67707253,
      "memory(GiB)": 67.97,
      "step": 3925,
      "train_speed(iter/s)": 1.651732
    },
    {
      "acc": 0.63164568,
      "epoch": 0.09969558599695585,
      "grad_norm": 6.5,
      "learning_rate": 9.969558599695586e-06,
      "loss": 1.77427979,
      "memory(GiB)": 67.97,
      "step": 3930,
      "train_speed(iter/s)": 1.651923
    },
    {
      "acc": 0.62451582,
      "epoch": 0.09982242516489091,
      "grad_norm": 6.09375,
      "learning_rate": 9.982242516489092e-06,
      "loss": 1.81063709,
      "memory(GiB)": 67.97,
      "step": 3935,
      "train_speed(iter/s)": 1.652129
    },
    {
      "acc": 0.64227715,
      "epoch": 0.09994926433282597,
      "grad_norm": 5.625,
      "learning_rate": 9.994926433282598e-06,
      "loss": 1.71588745,
      "memory(GiB)": 67.97,
      "step": 3940,
      "train_speed(iter/s)": 1.652325
    },
    {
      "acc": 0.65402422,
      "epoch": 0.10007610350076103,
      "grad_norm": 5.09375,
      "learning_rate": 9.999999960413982e-06,
      "loss": 1.69254265,
      "memory(GiB)": 67.97,
      "step": 3945,
      "train_speed(iter/s)": 1.652535
    },
    {
      "acc": 0.62749376,
      "epoch": 0.10020294266869609,
      "grad_norm": 9.375,
      "learning_rate": 9.99999971849943e-06,
      "loss": 1.82273827,
      "memory(GiB)": 67.97,
      "step": 3950,
      "train_speed(iter/s)": 1.652739
    },
    {
      "acc": 0.63605247,
      "epoch": 0.10032978183663115,
      "grad_norm": 5.65625,
      "learning_rate": 9.999999256662563e-06,
      "loss": 1.81794357,
      "memory(GiB)": 67.97,
      "step": 3955,
      "train_speed(iter/s)": 1.652957
    },
    {
      "acc": 0.64998913,
      "epoch": 0.1004566210045662,
      "grad_norm": 6.40625,
      "learning_rate": 9.999998574903408e-06,
      "loss": 1.7013237,
      "memory(GiB)": 67.97,
      "step": 3960,
      "train_speed(iter/s)": 1.653166
    },
    {
      "acc": 0.63812647,
      "epoch": 0.10058346017250126,
      "grad_norm": 6.40625,
      "learning_rate": 9.99999767322199e-06,
      "loss": 1.7743,
      "memory(GiB)": 67.97,
      "step": 3965,
      "train_speed(iter/s)": 1.653357
    },
    {
      "acc": 0.62548313,
      "epoch": 0.10071029934043632,
      "grad_norm": 5.09375,
      "learning_rate": 9.999996551618353e-06,
      "loss": 1.72225342,
      "memory(GiB)": 67.97,
      "step": 3970,
      "train_speed(iter/s)": 1.653561
    },
    {
      "acc": 0.61485071,
      "epoch": 0.10083713850837138,
      "grad_norm": 6.125,
      "learning_rate": 9.999995210092545e-06,
      "loss": 1.7970808,
      "memory(GiB)": 67.97,
      "step": 3975,
      "train_speed(iter/s)": 1.653754
    },
    {
      "acc": 0.62164564,
      "epoch": 0.10096397767630644,
      "grad_norm": 6.8125,
      "learning_rate": 9.999993648644622e-06,
      "loss": 1.75651016,
      "memory(GiB)": 67.97,
      "step": 3980,
      "train_speed(iter/s)": 1.653948
    },
    {
      "acc": 0.63222899,
      "epoch": 0.1010908168442415,
      "grad_norm": 6.25,
      "learning_rate": 9.999991867274656e-06,
      "loss": 1.78450928,
      "memory(GiB)": 67.97,
      "step": 3985,
      "train_speed(iter/s)": 1.65414
    },
    {
      "acc": 0.62905321,
      "epoch": 0.10121765601217655,
      "grad_norm": 6.4375,
      "learning_rate": 9.999989865982725e-06,
      "loss": 1.71516094,
      "memory(GiB)": 67.97,
      "step": 3990,
      "train_speed(iter/s)": 1.654343
    },
    {
      "acc": 0.62270198,
      "epoch": 0.10134449518011161,
      "grad_norm": 5.25,
      "learning_rate": 9.999987644768917e-06,
      "loss": 1.72050247,
      "memory(GiB)": 67.97,
      "step": 3995,
      "train_speed(iter/s)": 1.654539
    },
    {
      "acc": 0.63560915,
      "epoch": 0.10147133434804667,
      "grad_norm": 6.3125,
      "learning_rate": 9.999985203633327e-06,
      "loss": 1.78750134,
      "memory(GiB)": 67.97,
      "step": 4000,
      "train_speed(iter/s)": 1.654737
    },
    {
      "epoch": 0.10147133434804667,
      "eval_acc": 0.621197142361916,
      "eval_loss": 1.7191544771194458,
      "eval_runtime": 58.6125,
      "eval_samples_per_second": 108.68,
      "eval_steps_per_second": 27.178,
      "step": 4000
    },
    {
      "acc": 0.62880044,
      "epoch": 0.10159817351598173,
      "grad_norm": 6.9375,
      "learning_rate": 9.999982542576065e-06,
      "loss": 1.74402771,
      "memory(GiB)": 67.97,
      "step": 4005,
      "train_speed(iter/s)": 1.613029
    },
    {
      "acc": 0.63192768,
      "epoch": 0.10172501268391679,
      "grad_norm": 7.03125,
      "learning_rate": 9.999979661597247e-06,
      "loss": 1.79344215,
      "memory(GiB)": 67.97,
      "step": 4010,
      "train_speed(iter/s)": 1.613274
    },
    {
      "acc": 0.63948364,
      "epoch": 0.10185185185185185,
      "grad_norm": 6.125,
      "learning_rate": 9.999976560697002e-06,
      "loss": 1.7201767,
      "memory(GiB)": 67.97,
      "step": 4015,
      "train_speed(iter/s)": 1.613509
    },
    {
      "acc": 0.64680686,
      "epoch": 0.1019786910197869,
      "grad_norm": 6.875,
      "learning_rate": 9.999973239875462e-06,
      "loss": 1.73223801,
      "memory(GiB)": 67.97,
      "step": 4020,
      "train_speed(iter/s)": 1.613747
    },
    {
      "acc": 0.63398533,
      "epoch": 0.10210553018772196,
      "grad_norm": 6.625,
      "learning_rate": 9.999969699132776e-06,
      "loss": 1.7380722,
      "memory(GiB)": 67.97,
      "step": 4025,
      "train_speed(iter/s)": 1.613985
    },
    {
      "acc": 0.61618924,
      "epoch": 0.10223236935565702,
      "grad_norm": 4.8125,
      "learning_rate": 9.999965938469102e-06,
      "loss": 1.82005501,
      "memory(GiB)": 67.97,
      "step": 4030,
      "train_speed(iter/s)": 1.614221
    },
    {
      "acc": 0.62525425,
      "epoch": 0.10235920852359208,
      "grad_norm": 6.03125,
      "learning_rate": 9.9999619578846e-06,
      "loss": 1.78214111,
      "memory(GiB)": 67.97,
      "step": 4035,
      "train_speed(iter/s)": 1.614463
    },
    {
      "acc": 0.63352356,
      "epoch": 0.10248604769152714,
      "grad_norm": 6.71875,
      "learning_rate": 9.999957757379451e-06,
      "loss": 1.75636902,
      "memory(GiB)": 67.97,
      "step": 4040,
      "train_speed(iter/s)": 1.614706
    },
    {
      "acc": 0.64604177,
      "epoch": 0.1026128868594622,
      "grad_norm": 8.0,
      "learning_rate": 9.999953336953834e-06,
      "loss": 1.66816387,
      "memory(GiB)": 67.97,
      "step": 4045,
      "train_speed(iter/s)": 1.61493
    },
    {
      "acc": 0.62357917,
      "epoch": 0.10273972602739725,
      "grad_norm": 5.4375,
      "learning_rate": 9.999948696607946e-06,
      "loss": 1.77020607,
      "memory(GiB)": 67.97,
      "step": 4050,
      "train_speed(iter/s)": 1.615162
    },
    {
      "acc": 0.63200035,
      "epoch": 0.10286656519533231,
      "grad_norm": 5.4375,
      "learning_rate": 9.999943836341992e-06,
      "loss": 1.76096439,
      "memory(GiB)": 67.97,
      "step": 4055,
      "train_speed(iter/s)": 1.61539
    },
    {
      "acc": 0.65547247,
      "epoch": 0.10299340436326737,
      "grad_norm": 5.65625,
      "learning_rate": 9.999938756156185e-06,
      "loss": 1.64716663,
      "memory(GiB)": 67.97,
      "step": 4060,
      "train_speed(iter/s)": 1.615613
    },
    {
      "acc": 0.62961836,
      "epoch": 0.10312024353120243,
      "grad_norm": 5.78125,
      "learning_rate": 9.999933456050747e-06,
      "loss": 1.80200157,
      "memory(GiB)": 67.97,
      "step": 4065,
      "train_speed(iter/s)": 1.615838
    },
    {
      "acc": 0.62936592,
      "epoch": 0.10324708269913749,
      "grad_norm": 5.5,
      "learning_rate": 9.999927936025914e-06,
      "loss": 1.81737175,
      "memory(GiB)": 67.97,
      "step": 4070,
      "train_speed(iter/s)": 1.616064
    },
    {
      "acc": 0.63939648,
      "epoch": 0.10337392186707255,
      "grad_norm": 6.4375,
      "learning_rate": 9.999922196081928e-06,
      "loss": 1.72342014,
      "memory(GiB)": 67.97,
      "step": 4075,
      "train_speed(iter/s)": 1.616312
    },
    {
      "acc": 0.62949147,
      "epoch": 0.1035007610350076,
      "grad_norm": 5.9375,
      "learning_rate": 9.99991623621904e-06,
      "loss": 1.75348682,
      "memory(GiB)": 67.97,
      "step": 4080,
      "train_speed(iter/s)": 1.616523
    },
    {
      "acc": 0.63353796,
      "epoch": 0.10362760020294266,
      "grad_norm": 6.3125,
      "learning_rate": 9.999910056437512e-06,
      "loss": 1.76169186,
      "memory(GiB)": 67.97,
      "step": 4085,
      "train_speed(iter/s)": 1.616759
    },
    {
      "acc": 0.63861713,
      "epoch": 0.10375443937087772,
      "grad_norm": 4.875,
      "learning_rate": 9.999903656737618e-06,
      "loss": 1.71907082,
      "memory(GiB)": 67.97,
      "step": 4090,
      "train_speed(iter/s)": 1.61699
    },
    {
      "acc": 0.62957411,
      "epoch": 0.10388127853881278,
      "grad_norm": 6.28125,
      "learning_rate": 9.999897037119637e-06,
      "loss": 1.73045578,
      "memory(GiB)": 67.97,
      "step": 4095,
      "train_speed(iter/s)": 1.617219
    },
    {
      "acc": 0.62866564,
      "epoch": 0.10400811770674784,
      "grad_norm": 6.40625,
      "learning_rate": 9.999890197583862e-06,
      "loss": 1.77654228,
      "memory(GiB)": 67.97,
      "step": 4100,
      "train_speed(iter/s)": 1.617445
    },
    {
      "acc": 0.6204422,
      "epoch": 0.1041349568746829,
      "grad_norm": 7.0625,
      "learning_rate": 9.999883138130593e-06,
      "loss": 1.80975266,
      "memory(GiB)": 67.97,
      "step": 4105,
      "train_speed(iter/s)": 1.617658
    },
    {
      "acc": 0.64044752,
      "epoch": 0.10426179604261795,
      "grad_norm": 5.375,
      "learning_rate": 9.999875858760143e-06,
      "loss": 1.74375191,
      "memory(GiB)": 67.97,
      "step": 4110,
      "train_speed(iter/s)": 1.617877
    },
    {
      "acc": 0.628263,
      "epoch": 0.10438863521055301,
      "grad_norm": 5.84375,
      "learning_rate": 9.999868359472826e-06,
      "loss": 1.83561497,
      "memory(GiB)": 67.97,
      "step": 4115,
      "train_speed(iter/s)": 1.618091
    },
    {
      "acc": 0.63449526,
      "epoch": 0.10451547437848807,
      "grad_norm": 5.40625,
      "learning_rate": 9.999860640268977e-06,
      "loss": 1.77878284,
      "memory(GiB)": 67.97,
      "step": 4120,
      "train_speed(iter/s)": 1.61832
    },
    {
      "acc": 0.63982677,
      "epoch": 0.10464231354642313,
      "grad_norm": 5.90625,
      "learning_rate": 9.999852701148935e-06,
      "loss": 1.74853573,
      "memory(GiB)": 67.97,
      "step": 4125,
      "train_speed(iter/s)": 1.618538
    },
    {
      "acc": 0.65120993,
      "epoch": 0.10476915271435819,
      "grad_norm": 7.3125,
      "learning_rate": 9.999844542113049e-06,
      "loss": 1.75328979,
      "memory(GiB)": 67.97,
      "step": 4130,
      "train_speed(iter/s)": 1.61877
    },
    {
      "acc": 0.61673384,
      "epoch": 0.10489599188229325,
      "grad_norm": 5.40625,
      "learning_rate": 9.999836163161675e-06,
      "loss": 1.83802223,
      "memory(GiB)": 67.97,
      "step": 4135,
      "train_speed(iter/s)": 1.618999
    },
    {
      "acc": 0.66415944,
      "epoch": 0.1050228310502283,
      "grad_norm": 6.25,
      "learning_rate": 9.999827564295187e-06,
      "loss": 1.63727341,
      "memory(GiB)": 67.97,
      "step": 4140,
      "train_speed(iter/s)": 1.619197
    },
    {
      "acc": 0.64094048,
      "epoch": 0.10514967021816336,
      "grad_norm": 5.28125,
      "learning_rate": 9.999818745513958e-06,
      "loss": 1.73015862,
      "memory(GiB)": 67.97,
      "step": 4145,
      "train_speed(iter/s)": 1.619426
    },
    {
      "acc": 0.63930955,
      "epoch": 0.10527650938609842,
      "grad_norm": 5.6875,
      "learning_rate": 9.99980970681838e-06,
      "loss": 1.75726871,
      "memory(GiB)": 67.97,
      "step": 4150,
      "train_speed(iter/s)": 1.619641
    },
    {
      "acc": 0.64495535,
      "epoch": 0.10540334855403348,
      "grad_norm": 6.4375,
      "learning_rate": 9.999800448208846e-06,
      "loss": 1.69719219,
      "memory(GiB)": 67.97,
      "step": 4155,
      "train_speed(iter/s)": 1.619867
    },
    {
      "acc": 0.62620168,
      "epoch": 0.10553018772196854,
      "grad_norm": 5.1875,
      "learning_rate": 9.999790969685767e-06,
      "loss": 1.74291191,
      "memory(GiB)": 67.97,
      "step": 4160,
      "train_speed(iter/s)": 1.620084
    },
    {
      "acc": 0.6214191,
      "epoch": 0.1056570268899036,
      "grad_norm": 7.40625,
      "learning_rate": 9.999781271249559e-06,
      "loss": 1.76673584,
      "memory(GiB)": 67.97,
      "step": 4165,
      "train_speed(iter/s)": 1.620306
    },
    {
      "acc": 0.64081421,
      "epoch": 0.10578386605783865,
      "grad_norm": 5.21875,
      "learning_rate": 9.999771352900647e-06,
      "loss": 1.71336899,
      "memory(GiB)": 67.97,
      "step": 4170,
      "train_speed(iter/s)": 1.62052
    },
    {
      "acc": 0.64104505,
      "epoch": 0.10591070522577371,
      "grad_norm": 5.6875,
      "learning_rate": 9.999761214639469e-06,
      "loss": 1.63464375,
      "memory(GiB)": 67.97,
      "step": 4175,
      "train_speed(iter/s)": 1.620736
    },
    {
      "acc": 0.64937425,
      "epoch": 0.10603754439370877,
      "grad_norm": 5.59375,
      "learning_rate": 9.999750856466472e-06,
      "loss": 1.63867226,
      "memory(GiB)": 67.97,
      "step": 4180,
      "train_speed(iter/s)": 1.620903
    },
    {
      "acc": 0.63505259,
      "epoch": 0.10616438356164383,
      "grad_norm": 5.375,
      "learning_rate": 9.99974027838211e-06,
      "loss": 1.68519821,
      "memory(GiB)": 67.97,
      "step": 4185,
      "train_speed(iter/s)": 1.621141
    },
    {
      "acc": 0.64382148,
      "epoch": 0.10629122272957889,
      "grad_norm": 4.4375,
      "learning_rate": 9.999729480386846e-06,
      "loss": 1.69050255,
      "memory(GiB)": 67.97,
      "step": 4190,
      "train_speed(iter/s)": 1.62136
    },
    {
      "acc": 0.6431694,
      "epoch": 0.10641806189751395,
      "grad_norm": 5.59375,
      "learning_rate": 9.999718462481157e-06,
      "loss": 1.69474392,
      "memory(GiB)": 67.97,
      "step": 4195,
      "train_speed(iter/s)": 1.621585
    },
    {
      "acc": 0.64862137,
      "epoch": 0.106544901065449,
      "grad_norm": 5.15625,
      "learning_rate": 9.99970722466553e-06,
      "loss": 1.7278923,
      "memory(GiB)": 67.97,
      "step": 4200,
      "train_speed(iter/s)": 1.621792
    },
    {
      "acc": 0.64202552,
      "epoch": 0.10667174023338406,
      "grad_norm": 7.5,
      "learning_rate": 9.999695766940458e-06,
      "loss": 1.70533943,
      "memory(GiB)": 67.97,
      "step": 4205,
      "train_speed(iter/s)": 1.622012
    },
    {
      "acc": 0.62963924,
      "epoch": 0.10679857940131912,
      "grad_norm": 5.25,
      "learning_rate": 9.999684089306442e-06,
      "loss": 1.75145111,
      "memory(GiB)": 67.97,
      "step": 4210,
      "train_speed(iter/s)": 1.62223
    },
    {
      "acc": 0.63862095,
      "epoch": 0.10692541856925418,
      "grad_norm": 5.03125,
      "learning_rate": 9.999672191763999e-06,
      "loss": 1.71522713,
      "memory(GiB)": 67.97,
      "step": 4215,
      "train_speed(iter/s)": 1.62243
    },
    {
      "acc": 0.64715567,
      "epoch": 0.10705225773718924,
      "grad_norm": 6.71875,
      "learning_rate": 9.99966007431365e-06,
      "loss": 1.62397194,
      "memory(GiB)": 67.97,
      "step": 4220,
      "train_speed(iter/s)": 1.622625
    },
    {
      "acc": 0.63859258,
      "epoch": 0.1071790969051243,
      "grad_norm": 4.5625,
      "learning_rate": 9.99964773695593e-06,
      "loss": 1.73200054,
      "memory(GiB)": 67.97,
      "step": 4225,
      "train_speed(iter/s)": 1.622831
    },
    {
      "acc": 0.64712224,
      "epoch": 0.10730593607305935,
      "grad_norm": 6.125,
      "learning_rate": 9.999635179691381e-06,
      "loss": 1.77615089,
      "memory(GiB)": 67.97,
      "step": 4230,
      "train_speed(iter/s)": 1.623055
    },
    {
      "acc": 0.63496761,
      "epoch": 0.10743277524099441,
      "grad_norm": 5.40625,
      "learning_rate": 9.999622402520553e-06,
      "loss": 1.77117615,
      "memory(GiB)": 67.97,
      "step": 4235,
      "train_speed(iter/s)": 1.623269
    },
    {
      "acc": 0.6219996,
      "epoch": 0.10755961440892947,
      "grad_norm": 6.71875,
      "learning_rate": 9.999609405444012e-06,
      "loss": 1.77838955,
      "memory(GiB)": 67.97,
      "step": 4240,
      "train_speed(iter/s)": 1.623476
    },
    {
      "acc": 0.63712187,
      "epoch": 0.10768645357686453,
      "grad_norm": 5.84375,
      "learning_rate": 9.999596188462328e-06,
      "loss": 1.73780289,
      "memory(GiB)": 67.97,
      "step": 4245,
      "train_speed(iter/s)": 1.623676
    },
    {
      "acc": 0.64100833,
      "epoch": 0.10781329274479959,
      "grad_norm": 4.65625,
      "learning_rate": 9.99958275157608e-06,
      "loss": 1.71376457,
      "memory(GiB)": 67.97,
      "step": 4250,
      "train_speed(iter/s)": 1.623885
    },
    {
      "acc": 0.64295502,
      "epoch": 0.10794013191273465,
      "grad_norm": 5.96875,
      "learning_rate": 9.999569094785862e-06,
      "loss": 1.78116894,
      "memory(GiB)": 67.97,
      "step": 4255,
      "train_speed(iter/s)": 1.6241
    },
    {
      "acc": 0.64984322,
      "epoch": 0.1080669710806697,
      "grad_norm": 6.40625,
      "learning_rate": 9.999555218092273e-06,
      "loss": 1.7100069,
      "memory(GiB)": 67.97,
      "step": 4260,
      "train_speed(iter/s)": 1.624316
    },
    {
      "acc": 0.64320669,
      "epoch": 0.10819381024860476,
      "grad_norm": 5.6875,
      "learning_rate": 9.999541121495926e-06,
      "loss": 1.73749485,
      "memory(GiB)": 67.97,
      "step": 4265,
      "train_speed(iter/s)": 1.624532
    },
    {
      "acc": 0.6280302,
      "epoch": 0.10832064941653982,
      "grad_norm": 7.09375,
      "learning_rate": 9.999526804997439e-06,
      "loss": 1.77813206,
      "memory(GiB)": 67.97,
      "step": 4270,
      "train_speed(iter/s)": 1.624748
    },
    {
      "acc": 0.62221069,
      "epoch": 0.10844748858447488,
      "grad_norm": 5.4375,
      "learning_rate": 9.99951226859744e-06,
      "loss": 1.83957424,
      "memory(GiB)": 67.97,
      "step": 4275,
      "train_speed(iter/s)": 1.624963
    },
    {
      "acc": 0.64271383,
      "epoch": 0.10857432775240994,
      "grad_norm": 6.375,
      "learning_rate": 9.999497512296572e-06,
      "loss": 1.73349743,
      "memory(GiB)": 67.97,
      "step": 4280,
      "train_speed(iter/s)": 1.625174
    },
    {
      "acc": 0.6472898,
      "epoch": 0.108701166920345,
      "grad_norm": 10.25,
      "learning_rate": 9.999482536095483e-06,
      "loss": 1.73808861,
      "memory(GiB)": 67.97,
      "step": 4285,
      "train_speed(iter/s)": 1.625395
    },
    {
      "acc": 0.63170528,
      "epoch": 0.10882800608828005,
      "grad_norm": 4.78125,
      "learning_rate": 9.999467339994827e-06,
      "loss": 1.73651104,
      "memory(GiB)": 67.97,
      "step": 4290,
      "train_speed(iter/s)": 1.625598
    },
    {
      "acc": 0.64741445,
      "epoch": 0.10895484525621511,
      "grad_norm": 7.03125,
      "learning_rate": 9.99945192399528e-06,
      "loss": 1.71542015,
      "memory(GiB)": 67.97,
      "step": 4295,
      "train_speed(iter/s)": 1.625806
    },
    {
      "acc": 0.62867537,
      "epoch": 0.10908168442415017,
      "grad_norm": 6.09375,
      "learning_rate": 9.999436288097515e-06,
      "loss": 1.73983574,
      "memory(GiB)": 67.97,
      "step": 4300,
      "train_speed(iter/s)": 1.626007
    },
    {
      "acc": 0.63284001,
      "epoch": 0.10920852359208523,
      "grad_norm": 6.71875,
      "learning_rate": 9.99942043230222e-06,
      "loss": 1.77272644,
      "memory(GiB)": 67.97,
      "step": 4305,
      "train_speed(iter/s)": 1.626206
    },
    {
      "acc": 0.63630443,
      "epoch": 0.10933536276002029,
      "grad_norm": 6.90625,
      "learning_rate": 9.999404356610095e-06,
      "loss": 1.6563343,
      "memory(GiB)": 67.97,
      "step": 4310,
      "train_speed(iter/s)": 1.626398
    },
    {
      "acc": 0.6436305,
      "epoch": 0.10946220192795535,
      "grad_norm": 6.65625,
      "learning_rate": 9.999388061021846e-06,
      "loss": 1.74741096,
      "memory(GiB)": 67.97,
      "step": 4315,
      "train_speed(iter/s)": 1.626618
    },
    {
      "acc": 0.62731266,
      "epoch": 0.1095890410958904,
      "grad_norm": 5.65625,
      "learning_rate": 9.99937154553819e-06,
      "loss": 1.77855453,
      "memory(GiB)": 67.97,
      "step": 4320,
      "train_speed(iter/s)": 1.62683
    },
    {
      "acc": 0.61831484,
      "epoch": 0.10971588026382546,
      "grad_norm": 5.96875,
      "learning_rate": 9.999354810159852e-06,
      "loss": 1.74897957,
      "memory(GiB)": 67.97,
      "step": 4325,
      "train_speed(iter/s)": 1.627037
    },
    {
      "acc": 0.6361763,
      "epoch": 0.10984271943176052,
      "grad_norm": 5.5,
      "learning_rate": 9.999337854887567e-06,
      "loss": 1.7201046,
      "memory(GiB)": 67.97,
      "step": 4330,
      "train_speed(iter/s)": 1.627238
    },
    {
      "acc": 0.64260607,
      "epoch": 0.10996955859969558,
      "grad_norm": 6.0,
      "learning_rate": 9.999320679722086e-06,
      "loss": 1.7507761,
      "memory(GiB)": 67.97,
      "step": 4335,
      "train_speed(iter/s)": 1.627454
    },
    {
      "acc": 0.6355649,
      "epoch": 0.11009639776763064,
      "grad_norm": 5.625,
      "learning_rate": 9.999303284664159e-06,
      "loss": 1.71949749,
      "memory(GiB)": 67.97,
      "step": 4340,
      "train_speed(iter/s)": 1.627654
    },
    {
      "acc": 0.65132828,
      "epoch": 0.1102232369355657,
      "grad_norm": 6.0,
      "learning_rate": 9.999285669714555e-06,
      "loss": 1.68354759,
      "memory(GiB)": 67.97,
      "step": 4345,
      "train_speed(iter/s)": 1.627861
    },
    {
      "acc": 0.6522799,
      "epoch": 0.11035007610350075,
      "grad_norm": 5.5,
      "learning_rate": 9.999267834874044e-06,
      "loss": 1.61790657,
      "memory(GiB)": 67.97,
      "step": 4350,
      "train_speed(iter/s)": 1.628069
    },
    {
      "acc": 0.6344573,
      "epoch": 0.11047691527143581,
      "grad_norm": 6.875,
      "learning_rate": 9.999249780143416e-06,
      "loss": 1.69082794,
      "memory(GiB)": 67.97,
      "step": 4355,
      "train_speed(iter/s)": 1.628275
    },
    {
      "acc": 0.63763828,
      "epoch": 0.11060375443937087,
      "grad_norm": 5.15625,
      "learning_rate": 9.999231505523463e-06,
      "loss": 1.74555588,
      "memory(GiB)": 67.97,
      "step": 4360,
      "train_speed(iter/s)": 1.628477
    },
    {
      "acc": 0.61145363,
      "epoch": 0.11073059360730593,
      "grad_norm": 5.59375,
      "learning_rate": 9.999213011014987e-06,
      "loss": 1.79144897,
      "memory(GiB)": 67.97,
      "step": 4365,
      "train_speed(iter/s)": 1.628679
    },
    {
      "acc": 0.63161697,
      "epoch": 0.11085743277524099,
      "grad_norm": 5.125,
      "learning_rate": 9.999194296618805e-06,
      "loss": 1.75264168,
      "memory(GiB)": 67.97,
      "step": 4370,
      "train_speed(iter/s)": 1.628875
    },
    {
      "acc": 0.64262152,
      "epoch": 0.11098427194317605,
      "grad_norm": 11.0,
      "learning_rate": 9.999175362335735e-06,
      "loss": 1.7129652,
      "memory(GiB)": 67.97,
      "step": 4375,
      "train_speed(iter/s)": 1.62909
    },
    {
      "acc": 0.64140253,
      "epoch": 0.1111111111111111,
      "grad_norm": 5.9375,
      "learning_rate": 9.999156208166614e-06,
      "loss": 1.73836594,
      "memory(GiB)": 67.97,
      "step": 4380,
      "train_speed(iter/s)": 1.629305
    },
    {
      "acc": 0.63255754,
      "epoch": 0.11123795027904616,
      "grad_norm": 5.78125,
      "learning_rate": 9.999136834112284e-06,
      "loss": 1.71961479,
      "memory(GiB)": 67.97,
      "step": 4385,
      "train_speed(iter/s)": 1.629511
    },
    {
      "acc": 0.63558664,
      "epoch": 0.11136478944698122,
      "grad_norm": 4.8125,
      "learning_rate": 9.999117240173597e-06,
      "loss": 1.76141224,
      "memory(GiB)": 67.97,
      "step": 4390,
      "train_speed(iter/s)": 1.629715
    },
    {
      "acc": 0.62749786,
      "epoch": 0.11149162861491628,
      "grad_norm": 6.09375,
      "learning_rate": 9.999097426351412e-06,
      "loss": 1.80290337,
      "memory(GiB)": 67.97,
      "step": 4395,
      "train_speed(iter/s)": 1.629938
    },
    {
      "acc": 0.63315077,
      "epoch": 0.11161846778285134,
      "grad_norm": 5.84375,
      "learning_rate": 9.999077392646606e-06,
      "loss": 1.72891464,
      "memory(GiB)": 67.97,
      "step": 4400,
      "train_speed(iter/s)": 1.630144
    },
    {
      "acc": 0.64550276,
      "epoch": 0.1117453069507864,
      "grad_norm": 6.375,
      "learning_rate": 9.999057139060055e-06,
      "loss": 1.73157539,
      "memory(GiB)": 67.97,
      "step": 4405,
      "train_speed(iter/s)": 1.630344
    },
    {
      "acc": 0.64680681,
      "epoch": 0.11187214611872145,
      "grad_norm": 5.34375,
      "learning_rate": 9.999036665592653e-06,
      "loss": 1.69201775,
      "memory(GiB)": 67.97,
      "step": 4410,
      "train_speed(iter/s)": 1.630547
    },
    {
      "acc": 0.6476182,
      "epoch": 0.11199898528665651,
      "grad_norm": 5.28125,
      "learning_rate": 9.999015972245298e-06,
      "loss": 1.72968769,
      "memory(GiB)": 67.97,
      "step": 4415,
      "train_speed(iter/s)": 1.630743
    },
    {
      "acc": 0.63964167,
      "epoch": 0.11212582445459157,
      "grad_norm": 6.46875,
      "learning_rate": 9.998995059018901e-06,
      "loss": 1.72246742,
      "memory(GiB)": 67.97,
      "step": 4420,
      "train_speed(iter/s)": 1.630941
    },
    {
      "acc": 0.62957039,
      "epoch": 0.11225266362252663,
      "grad_norm": 6.0,
      "learning_rate": 9.998973925914384e-06,
      "loss": 1.69691658,
      "memory(GiB)": 67.97,
      "step": 4425,
      "train_speed(iter/s)": 1.631131
    },
    {
      "acc": 0.63208961,
      "epoch": 0.11237950279046169,
      "grad_norm": 5.34375,
      "learning_rate": 9.998952572932675e-06,
      "loss": 1.75270672,
      "memory(GiB)": 67.97,
      "step": 4430,
      "train_speed(iter/s)": 1.631324
    },
    {
      "acc": 0.61630292,
      "epoch": 0.11250634195839675,
      "grad_norm": 5.96875,
      "learning_rate": 9.998931000074712e-06,
      "loss": 1.83897057,
      "memory(GiB)": 67.97,
      "step": 4435,
      "train_speed(iter/s)": 1.631537
    },
    {
      "acc": 0.63067393,
      "epoch": 0.1126331811263318,
      "grad_norm": 5.75,
      "learning_rate": 9.998909207341446e-06,
      "loss": 1.72502518,
      "memory(GiB)": 67.97,
      "step": 4440,
      "train_speed(iter/s)": 1.631739
    },
    {
      "acc": 0.62769232,
      "epoch": 0.11276002029426686,
      "grad_norm": 5.5625,
      "learning_rate": 9.998887194733833e-06,
      "loss": 1.81189041,
      "memory(GiB)": 67.97,
      "step": 4445,
      "train_speed(iter/s)": 1.631941
    },
    {
      "acc": 0.64530735,
      "epoch": 0.11288685946220192,
      "grad_norm": 5.8125,
      "learning_rate": 9.998864962252843e-06,
      "loss": 1.66652012,
      "memory(GiB)": 67.97,
      "step": 4450,
      "train_speed(iter/s)": 1.632139
    },
    {
      "acc": 0.62774019,
      "epoch": 0.11301369863013698,
      "grad_norm": 5.40625,
      "learning_rate": 9.998842509899456e-06,
      "loss": 1.78872833,
      "memory(GiB)": 67.97,
      "step": 4455,
      "train_speed(iter/s)": 1.632349
    },
    {
      "acc": 0.63577385,
      "epoch": 0.11314053779807204,
      "grad_norm": 8.625,
      "learning_rate": 9.998819837674655e-06,
      "loss": 1.7395277,
      "memory(GiB)": 67.97,
      "step": 4460,
      "train_speed(iter/s)": 1.632552
    },
    {
      "acc": 0.64892306,
      "epoch": 0.1132673769660071,
      "grad_norm": 9.0,
      "learning_rate": 9.99879694557944e-06,
      "loss": 1.70531273,
      "memory(GiB)": 67.97,
      "step": 4465,
      "train_speed(iter/s)": 1.632742
    },
    {
      "acc": 0.6506196,
      "epoch": 0.11339421613394216,
      "grad_norm": 5.53125,
      "learning_rate": 9.998773833614816e-06,
      "loss": 1.72897644,
      "memory(GiB)": 67.97,
      "step": 4470,
      "train_speed(iter/s)": 1.632947
    },
    {
      "acc": 0.65738592,
      "epoch": 0.11352105530187721,
      "grad_norm": 6.09375,
      "learning_rate": 9.998750501781803e-06,
      "loss": 1.62091389,
      "memory(GiB)": 67.97,
      "step": 4475,
      "train_speed(iter/s)": 1.633152
    },
    {
      "acc": 0.64763403,
      "epoch": 0.11364789446981227,
      "grad_norm": 6.4375,
      "learning_rate": 9.998726950081425e-06,
      "loss": 1.73866997,
      "memory(GiB)": 67.97,
      "step": 4480,
      "train_speed(iter/s)": 1.633341
    },
    {
      "acc": 0.63893633,
      "epoch": 0.11377473363774733,
      "grad_norm": 6.9375,
      "learning_rate": 9.998703178514717e-06,
      "loss": 1.7975563,
      "memory(GiB)": 67.97,
      "step": 4485,
      "train_speed(iter/s)": 1.63354
    },
    {
      "acc": 0.63117728,
      "epoch": 0.11390157280568239,
      "grad_norm": 6.5625,
      "learning_rate": 9.998679187082724e-06,
      "loss": 1.70226288,
      "memory(GiB)": 67.97,
      "step": 4490,
      "train_speed(iter/s)": 1.633732
    },
    {
      "acc": 0.62577353,
      "epoch": 0.11402841197361745,
      "grad_norm": 5.5625,
      "learning_rate": 9.998654975786506e-06,
      "loss": 1.71917152,
      "memory(GiB)": 67.97,
      "step": 4495,
      "train_speed(iter/s)": 1.633929
    },
    {
      "acc": 0.63941765,
      "epoch": 0.1141552511415525,
      "grad_norm": 6.34375,
      "learning_rate": 9.998630544627123e-06,
      "loss": 1.68745232,
      "memory(GiB)": 67.97,
      "step": 4500,
      "train_speed(iter/s)": 1.634134
    },
    {
      "acc": 0.64434137,
      "epoch": 0.11428209030948756,
      "grad_norm": 4.90625,
      "learning_rate": 9.998605893605653e-06,
      "loss": 1.68942833,
      "memory(GiB)": 67.97,
      "step": 4505,
      "train_speed(iter/s)": 1.634309
    },
    {
      "acc": 0.64555216,
      "epoch": 0.11440892947742262,
      "grad_norm": 5.375,
      "learning_rate": 9.998581022723178e-06,
      "loss": 1.75617676,
      "memory(GiB)": 67.97,
      "step": 4510,
      "train_speed(iter/s)": 1.634497
    },
    {
      "acc": 0.61802735,
      "epoch": 0.11453576864535768,
      "grad_norm": 5.5,
      "learning_rate": 9.998555931980792e-06,
      "loss": 1.77623138,
      "memory(GiB)": 67.97,
      "step": 4515,
      "train_speed(iter/s)": 1.634674
    },
    {
      "acc": 0.62451324,
      "epoch": 0.11466260781329274,
      "grad_norm": 5.125,
      "learning_rate": 9.998530621379599e-06,
      "loss": 1.76006927,
      "memory(GiB)": 67.97,
      "step": 4520,
      "train_speed(iter/s)": 1.634847
    },
    {
      "acc": 0.64568024,
      "epoch": 0.1147894469812278,
      "grad_norm": 5.21875,
      "learning_rate": 9.998505090920713e-06,
      "loss": 1.71749382,
      "memory(GiB)": 67.97,
      "step": 4525,
      "train_speed(iter/s)": 1.635037
    },
    {
      "acc": 0.64712114,
      "epoch": 0.11491628614916286,
      "grad_norm": 6.125,
      "learning_rate": 9.998479340605257e-06,
      "loss": 1.70768147,
      "memory(GiB)": 67.97,
      "step": 4530,
      "train_speed(iter/s)": 1.635245
    },
    {
      "acc": 0.63834581,
      "epoch": 0.11504312531709791,
      "grad_norm": 5.0625,
      "learning_rate": 9.99845337043436e-06,
      "loss": 1.71314068,
      "memory(GiB)": 67.97,
      "step": 4535,
      "train_speed(iter/s)": 1.635435
    },
    {
      "acc": 0.61966105,
      "epoch": 0.11516996448503297,
      "grad_norm": 6.5625,
      "learning_rate": 9.998427180409171e-06,
      "loss": 1.83528786,
      "memory(GiB)": 67.97,
      "step": 4540,
      "train_speed(iter/s)": 1.635619
    },
    {
      "acc": 0.63189793,
      "epoch": 0.11529680365296803,
      "grad_norm": 8.0625,
      "learning_rate": 9.998400770530836e-06,
      "loss": 1.74503021,
      "memory(GiB)": 67.97,
      "step": 4545,
      "train_speed(iter/s)": 1.635808
    },
    {
      "acc": 0.63483181,
      "epoch": 0.11542364282090309,
      "grad_norm": 5.40625,
      "learning_rate": 9.99837414080052e-06,
      "loss": 1.71271553,
      "memory(GiB)": 67.97,
      "step": 4550,
      "train_speed(iter/s)": 1.636013
    },
    {
      "acc": 0.62522945,
      "epoch": 0.11555048198883815,
      "grad_norm": 5.625,
      "learning_rate": 9.998347291219393e-06,
      "loss": 1.75175209,
      "memory(GiB)": 67.97,
      "step": 4555,
      "train_speed(iter/s)": 1.636215
    },
    {
      "acc": 0.64219851,
      "epoch": 0.1156773211567732,
      "grad_norm": 5.34375,
      "learning_rate": 9.998320221788635e-06,
      "loss": 1.71677856,
      "memory(GiB)": 67.97,
      "step": 4560,
      "train_speed(iter/s)": 1.636405
    },
    {
      "acc": 0.61600294,
      "epoch": 0.11580416032470826,
      "grad_norm": 6.78125,
      "learning_rate": 9.998292932509438e-06,
      "loss": 1.89740715,
      "memory(GiB)": 67.97,
      "step": 4565,
      "train_speed(iter/s)": 1.636596
    },
    {
      "acc": 0.63052859,
      "epoch": 0.11593099949264332,
      "grad_norm": 5.34375,
      "learning_rate": 9.998265423383003e-06,
      "loss": 1.78069267,
      "memory(GiB)": 67.97,
      "step": 4570,
      "train_speed(iter/s)": 1.636785
    },
    {
      "acc": 0.63983808,
      "epoch": 0.11605783866057838,
      "grad_norm": 6.875,
      "learning_rate": 9.998237694410537e-06,
      "loss": 1.73593369,
      "memory(GiB)": 67.97,
      "step": 4575,
      "train_speed(iter/s)": 1.636975
    },
    {
      "acc": 0.64412074,
      "epoch": 0.11618467782851344,
      "grad_norm": 6.5,
      "learning_rate": 9.998209745593264e-06,
      "loss": 1.68864708,
      "memory(GiB)": 67.97,
      "step": 4580,
      "train_speed(iter/s)": 1.637169
    },
    {
      "acc": 0.63633366,
      "epoch": 0.1163115169964485,
      "grad_norm": 5.8125,
      "learning_rate": 9.99818157693241e-06,
      "loss": 1.73843613,
      "memory(GiB)": 67.97,
      "step": 4585,
      "train_speed(iter/s)": 1.637363
    },
    {
      "acc": 0.65344076,
      "epoch": 0.11643835616438356,
      "grad_norm": 5.375,
      "learning_rate": 9.998153188429216e-06,
      "loss": 1.60083752,
      "memory(GiB)": 67.97,
      "step": 4590,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.64183235,
      "epoch": 0.11656519533231861,
      "grad_norm": 5.625,
      "learning_rate": 9.99812458008493e-06,
      "loss": 1.6896965,
      "memory(GiB)": 67.97,
      "step": 4595,
      "train_speed(iter/s)": 1.637728
    },
    {
      "acc": 0.63621726,
      "epoch": 0.11669203450025367,
      "grad_norm": 4.8125,
      "learning_rate": 9.998095751900806e-06,
      "loss": 1.76160297,
      "memory(GiB)": 67.97,
      "step": 4600,
      "train_speed(iter/s)": 1.637911
    },
    {
      "acc": 0.62720642,
      "epoch": 0.11681887366818873,
      "grad_norm": 4.46875,
      "learning_rate": 9.99806670387812e-06,
      "loss": 1.74097443,
      "memory(GiB)": 67.97,
      "step": 4605,
      "train_speed(iter/s)": 1.638106
    },
    {
      "acc": 0.63699913,
      "epoch": 0.11694571283612379,
      "grad_norm": 6.1875,
      "learning_rate": 9.998037436018144e-06,
      "loss": 1.74612026,
      "memory(GiB)": 67.97,
      "step": 4610,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.63713827,
      "epoch": 0.11707255200405885,
      "grad_norm": 7.84375,
      "learning_rate": 9.998007948322168e-06,
      "loss": 1.66562538,
      "memory(GiB)": 67.97,
      "step": 4615,
      "train_speed(iter/s)": 1.638474
    },
    {
      "acc": 0.63443341,
      "epoch": 0.1171993911719939,
      "grad_norm": 5.34375,
      "learning_rate": 9.997978240791487e-06,
      "loss": 1.75145226,
      "memory(GiB)": 67.97,
      "step": 4620,
      "train_speed(iter/s)": 1.638672
    },
    {
      "acc": 0.63157201,
      "epoch": 0.11732623033992896,
      "grad_norm": 8.625,
      "learning_rate": 9.99794831342741e-06,
      "loss": 1.75364857,
      "memory(GiB)": 67.97,
      "step": 4625,
      "train_speed(iter/s)": 1.638874
    },
    {
      "acc": 0.63413148,
      "epoch": 0.11745306950786402,
      "grad_norm": 5.375,
      "learning_rate": 9.99791816623125e-06,
      "loss": 1.75627136,
      "memory(GiB)": 67.97,
      "step": 4630,
      "train_speed(iter/s)": 1.63907
    },
    {
      "acc": 0.6432013,
      "epoch": 0.11757990867579908,
      "grad_norm": 4.8125,
      "learning_rate": 9.997887799204335e-06,
      "loss": 1.67022057,
      "memory(GiB)": 67.97,
      "step": 4635,
      "train_speed(iter/s)": 1.639261
    },
    {
      "acc": 0.62581482,
      "epoch": 0.11770674784373414,
      "grad_norm": 5.3125,
      "learning_rate": 9.997857212348e-06,
      "loss": 1.7720356,
      "memory(GiB)": 67.97,
      "step": 4640,
      "train_speed(iter/s)": 1.639439
    },
    {
      "acc": 0.64077492,
      "epoch": 0.1178335870116692,
      "grad_norm": 5.09375,
      "learning_rate": 9.997826405663593e-06,
      "loss": 1.72522316,
      "memory(GiB)": 67.97,
      "step": 4645,
      "train_speed(iter/s)": 1.639626
    },
    {
      "acc": 0.65947065,
      "epoch": 0.11796042617960426,
      "grad_norm": 6.09375,
      "learning_rate": 9.997795379152468e-06,
      "loss": 1.58124809,
      "memory(GiB)": 67.97,
      "step": 4650,
      "train_speed(iter/s)": 1.639802
    },
    {
      "acc": 0.61063266,
      "epoch": 0.11808726534753931,
      "grad_norm": 5.8125,
      "learning_rate": 9.997764132815985e-06,
      "loss": 1.7946537,
      "memory(GiB)": 67.97,
      "step": 4655,
      "train_speed(iter/s)": 1.639985
    },
    {
      "acc": 0.6272243,
      "epoch": 0.11821410451547437,
      "grad_norm": 5.8125,
      "learning_rate": 9.997732666655524e-06,
      "loss": 1.79654083,
      "memory(GiB)": 67.97,
      "step": 4660,
      "train_speed(iter/s)": 1.640177
    },
    {
      "acc": 0.6152998,
      "epoch": 0.11834094368340943,
      "grad_norm": 5.25,
      "learning_rate": 9.997700980672469e-06,
      "loss": 1.77889538,
      "memory(GiB)": 67.97,
      "step": 4665,
      "train_speed(iter/s)": 1.64038
    },
    {
      "acc": 0.64574528,
      "epoch": 0.11846778285134449,
      "grad_norm": 5.71875,
      "learning_rate": 9.997669074868208e-06,
      "loss": 1.67501144,
      "memory(GiB)": 67.97,
      "step": 4670,
      "train_speed(iter/s)": 1.640572
    },
    {
      "acc": 0.64163051,
      "epoch": 0.11859462201927955,
      "grad_norm": 6.0625,
      "learning_rate": 9.997636949244151e-06,
      "loss": 1.72753601,
      "memory(GiB)": 67.97,
      "step": 4675,
      "train_speed(iter/s)": 1.640769
    },
    {
      "acc": 0.63248835,
      "epoch": 0.1187214611872146,
      "grad_norm": 5.4375,
      "learning_rate": 9.997604603801707e-06,
      "loss": 1.76771755,
      "memory(GiB)": 67.97,
      "step": 4680,
      "train_speed(iter/s)": 1.640963
    },
    {
      "acc": 0.63034139,
      "epoch": 0.11884830035514966,
      "grad_norm": 5.875,
      "learning_rate": 9.9975720385423e-06,
      "loss": 1.81456032,
      "memory(GiB)": 67.97,
      "step": 4685,
      "train_speed(iter/s)": 1.641155
    },
    {
      "acc": 0.64110069,
      "epoch": 0.11897513952308472,
      "grad_norm": 6.5,
      "learning_rate": 9.997539253467361e-06,
      "loss": 1.7475235,
      "memory(GiB)": 67.97,
      "step": 4690,
      "train_speed(iter/s)": 1.641342
    },
    {
      "acc": 0.63641825,
      "epoch": 0.11910197869101978,
      "grad_norm": 5.46875,
      "learning_rate": 9.997506248578334e-06,
      "loss": 1.73779526,
      "memory(GiB)": 67.97,
      "step": 4695,
      "train_speed(iter/s)": 1.641525
    },
    {
      "acc": 0.64413662,
      "epoch": 0.11922881785895484,
      "grad_norm": 7.3125,
      "learning_rate": 9.997473023876671e-06,
      "loss": 1.78399849,
      "memory(GiB)": 67.97,
      "step": 4700,
      "train_speed(iter/s)": 1.641712
    },
    {
      "acc": 0.63945947,
      "epoch": 0.1193556570268899,
      "grad_norm": 5.03125,
      "learning_rate": 9.997439579363831e-06,
      "loss": 1.79095154,
      "memory(GiB)": 67.97,
      "step": 4705,
      "train_speed(iter/s)": 1.641897
    },
    {
      "acc": 0.63179083,
      "epoch": 0.11948249619482496,
      "grad_norm": 6.0625,
      "learning_rate": 9.997405915041288e-06,
      "loss": 1.72449188,
      "memory(GiB)": 67.97,
      "step": 4710,
      "train_speed(iter/s)": 1.642066
    },
    {
      "acc": 0.64959311,
      "epoch": 0.11960933536276001,
      "grad_norm": 6.375,
      "learning_rate": 9.99737203091052e-06,
      "loss": 1.68974953,
      "memory(GiB)": 67.97,
      "step": 4715,
      "train_speed(iter/s)": 1.642225
    },
    {
      "acc": 0.65050182,
      "epoch": 0.11973617453069507,
      "grad_norm": 5.75,
      "learning_rate": 9.997337926973018e-06,
      "loss": 1.70047817,
      "memory(GiB)": 67.97,
      "step": 4720,
      "train_speed(iter/s)": 1.642401
    },
    {
      "acc": 0.63357754,
      "epoch": 0.11986301369863013,
      "grad_norm": 5.78125,
      "learning_rate": 9.997303603230282e-06,
      "loss": 1.7486412,
      "memory(GiB)": 67.97,
      "step": 4725,
      "train_speed(iter/s)": 1.642576
    },
    {
      "acc": 0.64942546,
      "epoch": 0.11998985286656519,
      "grad_norm": 6.4375,
      "learning_rate": 9.997269059683822e-06,
      "loss": 1.68743744,
      "memory(GiB)": 67.97,
      "step": 4730,
      "train_speed(iter/s)": 1.642746
    },
    {
      "acc": 0.62710915,
      "epoch": 0.12011669203450025,
      "grad_norm": 6.84375,
      "learning_rate": 9.997234296335159e-06,
      "loss": 1.75961323,
      "memory(GiB)": 67.97,
      "step": 4735,
      "train_speed(iter/s)": 1.642937
    },
    {
      "acc": 0.64256353,
      "epoch": 0.1202435312024353,
      "grad_norm": 5.34375,
      "learning_rate": 9.997199313185821e-06,
      "loss": 1.70316162,
      "memory(GiB)": 67.97,
      "step": 4740,
      "train_speed(iter/s)": 1.643104
    },
    {
      "acc": 0.65233736,
      "epoch": 0.12037037037037036,
      "grad_norm": 5.40625,
      "learning_rate": 9.997164110237345e-06,
      "loss": 1.60657234,
      "memory(GiB)": 67.97,
      "step": 4745,
      "train_speed(iter/s)": 1.643284
    },
    {
      "acc": 0.64171166,
      "epoch": 0.12049720953830542,
      "grad_norm": 6.53125,
      "learning_rate": 9.99712868749128e-06,
      "loss": 1.70478973,
      "memory(GiB)": 67.97,
      "step": 4750,
      "train_speed(iter/s)": 1.643465
    },
    {
      "acc": 0.66235285,
      "epoch": 0.12062404870624048,
      "grad_norm": 6.78125,
      "learning_rate": 9.997093044949186e-06,
      "loss": 1.71249313,
      "memory(GiB)": 67.97,
      "step": 4755,
      "train_speed(iter/s)": 1.643645
    },
    {
      "acc": 0.64120498,
      "epoch": 0.12075088787417554,
      "grad_norm": 6.0,
      "learning_rate": 9.997057182612631e-06,
      "loss": 1.67925282,
      "memory(GiB)": 67.97,
      "step": 4760,
      "train_speed(iter/s)": 1.643808
    },
    {
      "acc": 0.63244638,
      "epoch": 0.1208777270421106,
      "grad_norm": 6.5625,
      "learning_rate": 9.997021100483188e-06,
      "loss": 1.72408867,
      "memory(GiB)": 67.97,
      "step": 4765,
      "train_speed(iter/s)": 1.643983
    },
    {
      "acc": 0.64735184,
      "epoch": 0.12100456621004566,
      "grad_norm": 6.4375,
      "learning_rate": 9.996984798562448e-06,
      "loss": 1.65241432,
      "memory(GiB)": 67.97,
      "step": 4770,
      "train_speed(iter/s)": 1.644146
    },
    {
      "acc": 0.63440666,
      "epoch": 0.12113140537798071,
      "grad_norm": 6.09375,
      "learning_rate": 9.996948276852008e-06,
      "loss": 1.70845795,
      "memory(GiB)": 67.97,
      "step": 4775,
      "train_speed(iter/s)": 1.644322
    },
    {
      "acc": 0.63711462,
      "epoch": 0.12125824454591577,
      "grad_norm": 5.46875,
      "learning_rate": 9.99691153535347e-06,
      "loss": 1.7778965,
      "memory(GiB)": 67.97,
      "step": 4780,
      "train_speed(iter/s)": 1.644505
    },
    {
      "acc": 0.64322777,
      "epoch": 0.12138508371385083,
      "grad_norm": 5.34375,
      "learning_rate": 9.996874574068457e-06,
      "loss": 1.69959793,
      "memory(GiB)": 67.97,
      "step": 4785,
      "train_speed(iter/s)": 1.644671
    },
    {
      "acc": 0.6406477,
      "epoch": 0.12151192288178589,
      "grad_norm": 4.71875,
      "learning_rate": 9.996837392998586e-06,
      "loss": 1.72130203,
      "memory(GiB)": 67.97,
      "step": 4790,
      "train_speed(iter/s)": 1.644846
    },
    {
      "acc": 0.64000039,
      "epoch": 0.12163876204972095,
      "grad_norm": 6.21875,
      "learning_rate": 9.996799992145501e-06,
      "loss": 1.72441673,
      "memory(GiB)": 67.97,
      "step": 4795,
      "train_speed(iter/s)": 1.645014
    },
    {
      "acc": 0.64301157,
      "epoch": 0.121765601217656,
      "grad_norm": 5.40625,
      "learning_rate": 9.996762371510843e-06,
      "loss": 1.76455154,
      "memory(GiB)": 67.97,
      "step": 4800,
      "train_speed(iter/s)": 1.645192
    },
    {
      "acc": 0.64327803,
      "epoch": 0.12189244038559106,
      "grad_norm": 5.8125,
      "learning_rate": 9.996724531096264e-06,
      "loss": 1.72058315,
      "memory(GiB)": 67.97,
      "step": 4805,
      "train_speed(iter/s)": 1.645363
    },
    {
      "acc": 0.64176278,
      "epoch": 0.12201927955352612,
      "grad_norm": 6.34375,
      "learning_rate": 9.996686470903434e-06,
      "loss": 1.73159981,
      "memory(GiB)": 67.97,
      "step": 4810,
      "train_speed(iter/s)": 1.645528
    },
    {
      "acc": 0.64407272,
      "epoch": 0.12214611872146118,
      "grad_norm": 5.4375,
      "learning_rate": 9.996648190934025e-06,
      "loss": 1.6728405,
      "memory(GiB)": 67.97,
      "step": 4815,
      "train_speed(iter/s)": 1.645701
    },
    {
      "acc": 0.64219403,
      "epoch": 0.12227295788939624,
      "grad_norm": 6.1875,
      "learning_rate": 9.996609691189718e-06,
      "loss": 1.6965086,
      "memory(GiB)": 67.97,
      "step": 4820,
      "train_speed(iter/s)": 1.645873
    },
    {
      "acc": 0.64908476,
      "epoch": 0.1223997970573313,
      "grad_norm": 5.34375,
      "learning_rate": 9.996570971672209e-06,
      "loss": 1.68272762,
      "memory(GiB)": 67.97,
      "step": 4825,
      "train_speed(iter/s)": 1.646047
    },
    {
      "acc": 0.62543702,
      "epoch": 0.12252663622526636,
      "grad_norm": 6.0,
      "learning_rate": 9.996532032383202e-06,
      "loss": 1.81016521,
      "memory(GiB)": 67.97,
      "step": 4830,
      "train_speed(iter/s)": 1.646202
    },
    {
      "acc": 0.63629918,
      "epoch": 0.12265347539320141,
      "grad_norm": 7.03125,
      "learning_rate": 9.996492873324406e-06,
      "loss": 1.79810486,
      "memory(GiB)": 67.97,
      "step": 4835,
      "train_speed(iter/s)": 1.646374
    },
    {
      "acc": 0.63945713,
      "epoch": 0.12278031456113647,
      "grad_norm": 5.875,
      "learning_rate": 9.996453494497546e-06,
      "loss": 1.7556982,
      "memory(GiB)": 67.97,
      "step": 4840,
      "train_speed(iter/s)": 1.646542
    },
    {
      "acc": 0.63499227,
      "epoch": 0.12290715372907153,
      "grad_norm": 7.15625,
      "learning_rate": 9.996413895904355e-06,
      "loss": 1.72905922,
      "memory(GiB)": 67.97,
      "step": 4845,
      "train_speed(iter/s)": 1.646724
    },
    {
      "acc": 0.64340296,
      "epoch": 0.12303399289700659,
      "grad_norm": 5.875,
      "learning_rate": 9.996374077546573e-06,
      "loss": 1.67888451,
      "memory(GiB)": 67.97,
      "step": 4850,
      "train_speed(iter/s)": 1.646888
    },
    {
      "acc": 0.63953152,
      "epoch": 0.12316083206494165,
      "grad_norm": 5.53125,
      "learning_rate": 9.996334039425952e-06,
      "loss": 1.71106052,
      "memory(GiB)": 67.97,
      "step": 4855,
      "train_speed(iter/s)": 1.647057
    },
    {
      "acc": 0.65192795,
      "epoch": 0.1232876712328767,
      "grad_norm": 7.15625,
      "learning_rate": 9.996293781544255e-06,
      "loss": 1.65116768,
      "memory(GiB)": 67.97,
      "step": 4860,
      "train_speed(iter/s)": 1.647223
    },
    {
      "acc": 0.63779411,
      "epoch": 0.12341451040081176,
      "grad_norm": 5.21875,
      "learning_rate": 9.996253303903247e-06,
      "loss": 1.6962162,
      "memory(GiB)": 67.97,
      "step": 4865,
      "train_speed(iter/s)": 1.647387
    },
    {
      "acc": 0.63340173,
      "epoch": 0.12354134956874682,
      "grad_norm": 5.21875,
      "learning_rate": 9.996212606504713e-06,
      "loss": 1.68488808,
      "memory(GiB)": 67.97,
      "step": 4870,
      "train_speed(iter/s)": 1.647569
    },
    {
      "acc": 0.65302467,
      "epoch": 0.12366818873668188,
      "grad_norm": 5.71875,
      "learning_rate": 9.996171689350444e-06,
      "loss": 1.72270584,
      "memory(GiB)": 67.97,
      "step": 4875,
      "train_speed(iter/s)": 1.647726
    },
    {
      "acc": 0.65180416,
      "epoch": 0.12379502790461694,
      "grad_norm": 5.1875,
      "learning_rate": 9.996130552442237e-06,
      "loss": 1.64479866,
      "memory(GiB)": 67.97,
      "step": 4880,
      "train_speed(iter/s)": 1.647895
    },
    {
      "acc": 0.63972445,
      "epoch": 0.123921867072552,
      "grad_norm": 6.40625,
      "learning_rate": 9.996089195781902e-06,
      "loss": 1.61258755,
      "memory(GiB)": 67.97,
      "step": 4885,
      "train_speed(iter/s)": 1.64805
    },
    {
      "acc": 0.63272543,
      "epoch": 0.12404870624048706,
      "grad_norm": 5.71875,
      "learning_rate": 9.996047619371256e-06,
      "loss": 1.74801426,
      "memory(GiB)": 67.97,
      "step": 4890,
      "train_speed(iter/s)": 1.648203
    },
    {
      "acc": 0.64775581,
      "epoch": 0.12417554540842211,
      "grad_norm": 7.125,
      "learning_rate": 9.996005823212132e-06,
      "loss": 1.71116333,
      "memory(GiB)": 67.97,
      "step": 4895,
      "train_speed(iter/s)": 1.648357
    },
    {
      "acc": 0.63807197,
      "epoch": 0.12430238457635717,
      "grad_norm": 8.125,
      "learning_rate": 9.995963807306368e-06,
      "loss": 1.69778404,
      "memory(GiB)": 67.97,
      "step": 4900,
      "train_speed(iter/s)": 1.648531
    },
    {
      "acc": 0.64420915,
      "epoch": 0.12442922374429223,
      "grad_norm": 5.28125,
      "learning_rate": 9.995921571655808e-06,
      "loss": 1.70640068,
      "memory(GiB)": 67.97,
      "step": 4905,
      "train_speed(iter/s)": 1.648694
    },
    {
      "acc": 0.62597256,
      "epoch": 0.12455606291222729,
      "grad_norm": 6.9375,
      "learning_rate": 9.995879116262312e-06,
      "loss": 1.75829544,
      "memory(GiB)": 67.97,
      "step": 4910,
      "train_speed(iter/s)": 1.648874
    },
    {
      "acc": 0.65001144,
      "epoch": 0.12468290208016235,
      "grad_norm": 5.28125,
      "learning_rate": 9.995836441127749e-06,
      "loss": 1.62578773,
      "memory(GiB)": 67.97,
      "step": 4915,
      "train_speed(iter/s)": 1.649025
    },
    {
      "acc": 0.64603391,
      "epoch": 0.1248097412480974,
      "grad_norm": 5.40625,
      "learning_rate": 9.995793546253993e-06,
      "loss": 1.63102798,
      "memory(GiB)": 67.97,
      "step": 4920,
      "train_speed(iter/s)": 1.649174
    },
    {
      "acc": 0.64495444,
      "epoch": 0.12493658041603246,
      "grad_norm": 5.8125,
      "learning_rate": 9.995750431642933e-06,
      "loss": 1.71283379,
      "memory(GiB)": 67.97,
      "step": 4925,
      "train_speed(iter/s)": 1.649334
    },
    {
      "acc": 0.63811507,
      "epoch": 0.12506341958396752,
      "grad_norm": 6.84375,
      "learning_rate": 9.995707097296465e-06,
      "loss": 1.72495098,
      "memory(GiB)": 67.97,
      "step": 4930,
      "train_speed(iter/s)": 1.649506
    },
    {
      "acc": 0.63929243,
      "epoch": 0.1251902587519026,
      "grad_norm": 5.625,
      "learning_rate": 9.995663543216493e-06,
      "loss": 1.75177402,
      "memory(GiB)": 67.97,
      "step": 4935,
      "train_speed(iter/s)": 1.649671
    },
    {
      "acc": 0.64940729,
      "epoch": 0.12531709791983764,
      "grad_norm": 5.34375,
      "learning_rate": 9.995619769404936e-06,
      "loss": 1.71212883,
      "memory(GiB)": 67.97,
      "step": 4940,
      "train_speed(iter/s)": 1.649835
    },
    {
      "acc": 0.62320843,
      "epoch": 0.1254439370877727,
      "grad_norm": 7.125,
      "learning_rate": 9.995575775863717e-06,
      "loss": 1.79358368,
      "memory(GiB)": 67.97,
      "step": 4945,
      "train_speed(iter/s)": 1.649988
    },
    {
      "acc": 0.66100016,
      "epoch": 0.12557077625570776,
      "grad_norm": 6.90625,
      "learning_rate": 9.995531562594773e-06,
      "loss": 1.69016685,
      "memory(GiB)": 67.97,
      "step": 4950,
      "train_speed(iter/s)": 1.650137
    },
    {
      "acc": 0.65467863,
      "epoch": 0.12569761542364283,
      "grad_norm": 6.21875,
      "learning_rate": 9.995487129600046e-06,
      "loss": 1.64531898,
      "memory(GiB)": 67.97,
      "step": 4955,
      "train_speed(iter/s)": 1.650292
    },
    {
      "acc": 0.64707317,
      "epoch": 0.12582445459157787,
      "grad_norm": 5.40625,
      "learning_rate": 9.995442476881491e-06,
      "loss": 1.67819366,
      "memory(GiB)": 67.97,
      "step": 4960,
      "train_speed(iter/s)": 1.650443
    },
    {
      "acc": 0.64336152,
      "epoch": 0.12595129375951294,
      "grad_norm": 4.8125,
      "learning_rate": 9.995397604441076e-06,
      "loss": 1.68882675,
      "memory(GiB)": 67.97,
      "step": 4965,
      "train_speed(iter/s)": 1.650604
    },
    {
      "acc": 0.64805965,
      "epoch": 0.126078132927448,
      "grad_norm": 6.09375,
      "learning_rate": 9.995352512280767e-06,
      "loss": 1.73531609,
      "memory(GiB)": 67.97,
      "step": 4970,
      "train_speed(iter/s)": 1.650772
    },
    {
      "acc": 0.64736953,
      "epoch": 0.12620497209538306,
      "grad_norm": 6.5,
      "learning_rate": 9.995307200402555e-06,
      "loss": 1.72024765,
      "memory(GiB)": 67.97,
      "step": 4975,
      "train_speed(iter/s)": 1.650916
    },
    {
      "acc": 0.64368896,
      "epoch": 0.1263318112633181,
      "grad_norm": 6.09375,
      "learning_rate": 9.995261668808429e-06,
      "loss": 1.63274879,
      "memory(GiB)": 67.97,
      "step": 4980,
      "train_speed(iter/s)": 1.651074
    },
    {
      "acc": 0.6580523,
      "epoch": 0.12645865043125318,
      "grad_norm": 5.5,
      "learning_rate": 9.995215917500395e-06,
      "loss": 1.64459648,
      "memory(GiB)": 67.97,
      "step": 4985,
      "train_speed(iter/s)": 1.65123
    },
    {
      "acc": 0.62267323,
      "epoch": 0.12658548959918822,
      "grad_norm": 5.1875,
      "learning_rate": 9.995169946480459e-06,
      "loss": 1.70902634,
      "memory(GiB)": 67.97,
      "step": 4990,
      "train_speed(iter/s)": 1.651386
    },
    {
      "acc": 0.63211341,
      "epoch": 0.1267123287671233,
      "grad_norm": 6.375,
      "learning_rate": 9.99512375575065e-06,
      "loss": 1.69203529,
      "memory(GiB)": 67.97,
      "step": 4995,
      "train_speed(iter/s)": 1.651543
    },
    {
      "acc": 0.6414547,
      "epoch": 0.12683916793505834,
      "grad_norm": 6.28125,
      "learning_rate": 9.995077345312994e-06,
      "loss": 1.70896854,
      "memory(GiB)": 67.97,
      "step": 5000,
      "train_speed(iter/s)": 1.651709
    },
    {
      "epoch": 0.12683916793505834,
      "eval_acc": 0.6274574664252585,
      "eval_loss": 1.6814990043640137,
      "eval_runtime": 58.1476,
      "eval_samples_per_second": 109.549,
      "eval_steps_per_second": 27.396,
      "step": 5000
    },
    {
      "acc": 0.63395348,
      "epoch": 0.1269660071029934,
      "grad_norm": 6.6875,
      "learning_rate": 9.995030715169535e-06,
      "loss": 1.74872665,
      "memory(GiB)": 67.97,
      "step": 5005,
      "train_speed(iter/s)": 1.618533
    },
    {
      "acc": 0.62984314,
      "epoch": 0.12709284627092846,
      "grad_norm": 5.40625,
      "learning_rate": 9.994983865322327e-06,
      "loss": 1.79818287,
      "memory(GiB)": 67.97,
      "step": 5010,
      "train_speed(iter/s)": 1.618712
    },
    {
      "acc": 0.63942041,
      "epoch": 0.12721968543886353,
      "grad_norm": 7.40625,
      "learning_rate": 9.994936795773424e-06,
      "loss": 1.71386108,
      "memory(GiB)": 67.97,
      "step": 5015,
      "train_speed(iter/s)": 1.61889
    },
    {
      "acc": 0.64456043,
      "epoch": 0.12734652460679857,
      "grad_norm": 4.65625,
      "learning_rate": 9.994889506524903e-06,
      "loss": 1.70381527,
      "memory(GiB)": 67.97,
      "step": 5020,
      "train_speed(iter/s)": 1.619065
    },
    {
      "acc": 0.64319324,
      "epoch": 0.12747336377473364,
      "grad_norm": 6.59375,
      "learning_rate": 9.994841997578839e-06,
      "loss": 1.7647501,
      "memory(GiB)": 67.97,
      "step": 5025,
      "train_speed(iter/s)": 1.619243
    },
    {
      "acc": 0.64262609,
      "epoch": 0.1276002029426687,
      "grad_norm": 4.4375,
      "learning_rate": 9.994794268937325e-06,
      "loss": 1.72183475,
      "memory(GiB)": 67.97,
      "step": 5030,
      "train_speed(iter/s)": 1.619405
    },
    {
      "acc": 0.6308147,
      "epoch": 0.12772704211060376,
      "grad_norm": 8.1875,
      "learning_rate": 9.994746320602457e-06,
      "loss": 1.77366085,
      "memory(GiB)": 67.97,
      "step": 5035,
      "train_speed(iter/s)": 1.619597
    },
    {
      "acc": 0.64514465,
      "epoch": 0.1278538812785388,
      "grad_norm": 4.875,
      "learning_rate": 9.994698152576347e-06,
      "loss": 1.73413086,
      "memory(GiB)": 67.97,
      "step": 5040,
      "train_speed(iter/s)": 1.619782
    },
    {
      "acc": 0.64518166,
      "epoch": 0.12798072044647388,
      "grad_norm": 5.875,
      "learning_rate": 9.994649764861114e-06,
      "loss": 1.69440193,
      "memory(GiB)": 67.97,
      "step": 5045,
      "train_speed(iter/s)": 1.619967
    },
    {
      "acc": 0.63295403,
      "epoch": 0.12810755961440892,
      "grad_norm": 5.5625,
      "learning_rate": 9.994601157458882e-06,
      "loss": 1.70097904,
      "memory(GiB)": 67.97,
      "step": 5050,
      "train_speed(iter/s)": 1.620153
    },
    {
      "acc": 0.61968193,
      "epoch": 0.128234398782344,
      "grad_norm": 5.8125,
      "learning_rate": 9.994552330371792e-06,
      "loss": 1.78249187,
      "memory(GiB)": 67.97,
      "step": 5055,
      "train_speed(iter/s)": 1.620332
    },
    {
      "acc": 0.62847509,
      "epoch": 0.12836123795027904,
      "grad_norm": 5.0625,
      "learning_rate": 9.994503283601993e-06,
      "loss": 1.73668785,
      "memory(GiB)": 67.97,
      "step": 5060,
      "train_speed(iter/s)": 1.620509
    },
    {
      "acc": 0.62697902,
      "epoch": 0.1284880771182141,
      "grad_norm": 5.0,
      "learning_rate": 9.99445401715164e-06,
      "loss": 1.75026798,
      "memory(GiB)": 67.97,
      "step": 5065,
      "train_speed(iter/s)": 1.620676
    },
    {
      "acc": 0.64302969,
      "epoch": 0.12861491628614916,
      "grad_norm": 5.21875,
      "learning_rate": 9.994404531022901e-06,
      "loss": 1.6923233,
      "memory(GiB)": 67.97,
      "step": 5070,
      "train_speed(iter/s)": 1.620845
    },
    {
      "acc": 0.64488964,
      "epoch": 0.12874175545408423,
      "grad_norm": 6.78125,
      "learning_rate": 9.994354825217954e-06,
      "loss": 1.66673298,
      "memory(GiB)": 67.97,
      "step": 5075,
      "train_speed(iter/s)": 1.621022
    },
    {
      "acc": 0.63519349,
      "epoch": 0.12886859462201927,
      "grad_norm": 6.6875,
      "learning_rate": 9.99430489973898e-06,
      "loss": 1.73534412,
      "memory(GiB)": 67.97,
      "step": 5080,
      "train_speed(iter/s)": 1.621195
    },
    {
      "acc": 0.62550731,
      "epoch": 0.12899543378995434,
      "grad_norm": 8.25,
      "learning_rate": 9.994254754588182e-06,
      "loss": 1.80104256,
      "memory(GiB)": 67.97,
      "step": 5085,
      "train_speed(iter/s)": 1.621379
    },
    {
      "acc": 0.64148254,
      "epoch": 0.1291222729578894,
      "grad_norm": 5.125,
      "learning_rate": 9.99420438976776e-06,
      "loss": 1.63720379,
      "memory(GiB)": 67.97,
      "step": 5090,
      "train_speed(iter/s)": 1.621502
    },
    {
      "acc": 0.63018656,
      "epoch": 0.12924911212582446,
      "grad_norm": 6.0,
      "learning_rate": 9.994153805279932e-06,
      "loss": 1.72082748,
      "memory(GiB)": 67.97,
      "step": 5095,
      "train_speed(iter/s)": 1.621638
    },
    {
      "acc": 0.64703856,
      "epoch": 0.1293759512937595,
      "grad_norm": 5.875,
      "learning_rate": 9.994103001126923e-06,
      "loss": 1.65867615,
      "memory(GiB)": 67.97,
      "step": 5100,
      "train_speed(iter/s)": 1.621809
    },
    {
      "acc": 0.64282503,
      "epoch": 0.12950279046169458,
      "grad_norm": 6.21875,
      "learning_rate": 9.994051977310966e-06,
      "loss": 1.69839401,
      "memory(GiB)": 67.97,
      "step": 5105,
      "train_speed(iter/s)": 1.621992
    },
    {
      "acc": 0.62093878,
      "epoch": 0.12962962962962962,
      "grad_norm": 5.53125,
      "learning_rate": 9.994000733834307e-06,
      "loss": 1.80523911,
      "memory(GiB)": 67.97,
      "step": 5110,
      "train_speed(iter/s)": 1.622167
    },
    {
      "acc": 0.6438539,
      "epoch": 0.1297564687975647,
      "grad_norm": 4.71875,
      "learning_rate": 9.993949270699197e-06,
      "loss": 1.69229202,
      "memory(GiB)": 67.97,
      "step": 5115,
      "train_speed(iter/s)": 1.622334
    },
    {
      "acc": 0.65608463,
      "epoch": 0.12988330796549974,
      "grad_norm": 5.4375,
      "learning_rate": 9.993897587907904e-06,
      "loss": 1.63800278,
      "memory(GiB)": 67.97,
      "step": 5120,
      "train_speed(iter/s)": 1.622521
    },
    {
      "acc": 0.63888769,
      "epoch": 0.1300101471334348,
      "grad_norm": 6.125,
      "learning_rate": 9.993845685462697e-06,
      "loss": 1.71679993,
      "memory(GiB)": 67.97,
      "step": 5125,
      "train_speed(iter/s)": 1.622697
    },
    {
      "acc": 0.61410284,
      "epoch": 0.13013698630136986,
      "grad_norm": 7.03125,
      "learning_rate": 9.993793563365864e-06,
      "loss": 1.76411209,
      "memory(GiB)": 67.97,
      "step": 5130,
      "train_speed(iter/s)": 1.622872
    },
    {
      "acc": 0.63688469,
      "epoch": 0.13026382546930493,
      "grad_norm": 5.375,
      "learning_rate": 9.993741221619692e-06,
      "loss": 1.69519062,
      "memory(GiB)": 67.97,
      "step": 5135,
      "train_speed(iter/s)": 1.623041
    },
    {
      "acc": 0.64492025,
      "epoch": 0.13039066463723997,
      "grad_norm": 5.46875,
      "learning_rate": 9.993688660226486e-06,
      "loss": 1.71028175,
      "memory(GiB)": 67.97,
      "step": 5140,
      "train_speed(iter/s)": 1.623214
    },
    {
      "acc": 0.65607786,
      "epoch": 0.13051750380517504,
      "grad_norm": 8.75,
      "learning_rate": 9.993635879188557e-06,
      "loss": 1.63757133,
      "memory(GiB)": 67.97,
      "step": 5145,
      "train_speed(iter/s)": 1.623372
    },
    {
      "acc": 0.64062729,
      "epoch": 0.1306443429731101,
      "grad_norm": 6.09375,
      "learning_rate": 9.993582878508229e-06,
      "loss": 1.69953728,
      "memory(GiB)": 67.97,
      "step": 5150,
      "train_speed(iter/s)": 1.623553
    },
    {
      "acc": 0.62717762,
      "epoch": 0.13077118214104516,
      "grad_norm": 5.4375,
      "learning_rate": 9.993529658187829e-06,
      "loss": 1.739147,
      "memory(GiB)": 67.97,
      "step": 5155,
      "train_speed(iter/s)": 1.623731
    },
    {
      "acc": 0.64492464,
      "epoch": 0.1308980213089802,
      "grad_norm": 5.59375,
      "learning_rate": 9.9934762182297e-06,
      "loss": 1.69992256,
      "memory(GiB)": 67.97,
      "step": 5160,
      "train_speed(iter/s)": 1.623903
    },
    {
      "acc": 0.6508604,
      "epoch": 0.13102486047691528,
      "grad_norm": 5.6875,
      "learning_rate": 9.993422558636194e-06,
      "loss": 1.6587532,
      "memory(GiB)": 67.97,
      "step": 5165,
      "train_speed(iter/s)": 1.624087
    },
    {
      "acc": 0.64501171,
      "epoch": 0.13115169964485032,
      "grad_norm": 5.0625,
      "learning_rate": 9.99336867940967e-06,
      "loss": 1.6909111,
      "memory(GiB)": 67.97,
      "step": 5170,
      "train_speed(iter/s)": 1.624255
    },
    {
      "acc": 0.65164576,
      "epoch": 0.1312785388127854,
      "grad_norm": 6.28125,
      "learning_rate": 9.993314580552497e-06,
      "loss": 1.63433113,
      "memory(GiB)": 67.97,
      "step": 5175,
      "train_speed(iter/s)": 1.624426
    },
    {
      "acc": 0.63783703,
      "epoch": 0.13140537798072044,
      "grad_norm": 7.71875,
      "learning_rate": 9.993260262067054e-06,
      "loss": 1.66052856,
      "memory(GiB)": 67.97,
      "step": 5180,
      "train_speed(iter/s)": 1.624587
    },
    {
      "acc": 0.62946987,
      "epoch": 0.1315322171486555,
      "grad_norm": 6.65625,
      "learning_rate": 9.993205723955734e-06,
      "loss": 1.76554985,
      "memory(GiB)": 67.97,
      "step": 5185,
      "train_speed(iter/s)": 1.624754
    },
    {
      "acc": 0.63935471,
      "epoch": 0.13165905631659056,
      "grad_norm": 7.125,
      "learning_rate": 9.993150966220933e-06,
      "loss": 1.65852242,
      "memory(GiB)": 67.97,
      "step": 5190,
      "train_speed(iter/s)": 1.624926
    },
    {
      "acc": 0.63495178,
      "epoch": 0.13178589548452563,
      "grad_norm": 5.5,
      "learning_rate": 9.993095988865057e-06,
      "loss": 1.73110809,
      "memory(GiB)": 67.97,
      "step": 5195,
      "train_speed(iter/s)": 1.625111
    },
    {
      "acc": 0.6334991,
      "epoch": 0.13191273465246067,
      "grad_norm": 4.75,
      "learning_rate": 9.99304079189053e-06,
      "loss": 1.75615196,
      "memory(GiB)": 67.97,
      "step": 5200,
      "train_speed(iter/s)": 1.625279
    },
    {
      "acc": 0.64708085,
      "epoch": 0.13203957382039574,
      "grad_norm": 16.125,
      "learning_rate": 9.992985375299775e-06,
      "loss": 1.64645767,
      "memory(GiB)": 67.97,
      "step": 5205,
      "train_speed(iter/s)": 1.62545
    },
    {
      "acc": 0.62608576,
      "epoch": 0.1321664129883308,
      "grad_norm": 7.71875,
      "learning_rate": 9.992929739095232e-06,
      "loss": 1.72588921,
      "memory(GiB)": 67.97,
      "step": 5210,
      "train_speed(iter/s)": 1.625619
    },
    {
      "acc": 0.62113743,
      "epoch": 0.13229325215626586,
      "grad_norm": 6.5625,
      "learning_rate": 9.992873883279345e-06,
      "loss": 1.79214783,
      "memory(GiB)": 67.97,
      "step": 5215,
      "train_speed(iter/s)": 1.625798
    },
    {
      "acc": 0.62433376,
      "epoch": 0.1324200913242009,
      "grad_norm": 5.125,
      "learning_rate": 9.992817807854575e-06,
      "loss": 1.72761688,
      "memory(GiB)": 67.97,
      "step": 5220,
      "train_speed(iter/s)": 1.625978
    },
    {
      "acc": 0.6437819,
      "epoch": 0.13254693049213598,
      "grad_norm": 5.53125,
      "learning_rate": 9.992761512823386e-06,
      "loss": 1.70482292,
      "memory(GiB)": 67.97,
      "step": 5225,
      "train_speed(iter/s)": 1.626153
    },
    {
      "acc": 0.63523312,
      "epoch": 0.13267376966007102,
      "grad_norm": 6.4375,
      "learning_rate": 9.992704998188255e-06,
      "loss": 1.7106987,
      "memory(GiB)": 67.97,
      "step": 5230,
      "train_speed(iter/s)": 1.626319
    },
    {
      "acc": 0.61373291,
      "epoch": 0.1328006088280061,
      "grad_norm": 6.125,
      "learning_rate": 9.992648263951668e-06,
      "loss": 1.82784958,
      "memory(GiB)": 67.97,
      "step": 5235,
      "train_speed(iter/s)": 1.626497
    },
    {
      "acc": 0.64305134,
      "epoch": 0.13292744799594114,
      "grad_norm": 5.4375,
      "learning_rate": 9.992591310116118e-06,
      "loss": 1.73701458,
      "memory(GiB)": 67.97,
      "step": 5240,
      "train_speed(iter/s)": 1.62667
    },
    {
      "acc": 0.63666525,
      "epoch": 0.1330542871638762,
      "grad_norm": 5.84375,
      "learning_rate": 9.992534136684112e-06,
      "loss": 1.67466583,
      "memory(GiB)": 67.97,
      "step": 5245,
      "train_speed(iter/s)": 1.626844
    },
    {
      "acc": 0.63887062,
      "epoch": 0.13318112633181126,
      "grad_norm": 5.65625,
      "learning_rate": 9.992476743658165e-06,
      "loss": 1.70007858,
      "memory(GiB)": 67.97,
      "step": 5250,
      "train_speed(iter/s)": 1.627002
    },
    {
      "acc": 0.65181856,
      "epoch": 0.13330796549974633,
      "grad_norm": 5.0,
      "learning_rate": 9.992419131040803e-06,
      "loss": 1.62725792,
      "memory(GiB)": 67.97,
      "step": 5255,
      "train_speed(iter/s)": 1.62717
    },
    {
      "acc": 0.65718045,
      "epoch": 0.13343480466768137,
      "grad_norm": 5.96875,
      "learning_rate": 9.992361298834555e-06,
      "loss": 1.6195652,
      "memory(GiB)": 67.97,
      "step": 5260,
      "train_speed(iter/s)": 1.627347
    },
    {
      "acc": 0.63881211,
      "epoch": 0.13356164383561644,
      "grad_norm": 5.5,
      "learning_rate": 9.99230324704197e-06,
      "loss": 1.76107922,
      "memory(GiB)": 67.97,
      "step": 5265,
      "train_speed(iter/s)": 1.627522
    },
    {
      "acc": 0.6444293,
      "epoch": 0.1336884830035515,
      "grad_norm": 6.78125,
      "learning_rate": 9.992244975665598e-06,
      "loss": 1.69464779,
      "memory(GiB)": 67.97,
      "step": 5270,
      "train_speed(iter/s)": 1.627692
    },
    {
      "acc": 0.64365602,
      "epoch": 0.13381532217148656,
      "grad_norm": 4.96875,
      "learning_rate": 9.992186484708003e-06,
      "loss": 1.72151871,
      "memory(GiB)": 67.97,
      "step": 5275,
      "train_speed(iter/s)": 1.62786
    },
    {
      "acc": 0.64648089,
      "epoch": 0.1339421613394216,
      "grad_norm": 5.90625,
      "learning_rate": 9.992127774171759e-06,
      "loss": 1.65160332,
      "memory(GiB)": 67.97,
      "step": 5280,
      "train_speed(iter/s)": 1.628024
    },
    {
      "acc": 0.62512159,
      "epoch": 0.13406900050735668,
      "grad_norm": 5.875,
      "learning_rate": 9.992068844059446e-06,
      "loss": 1.83518524,
      "memory(GiB)": 67.97,
      "step": 5285,
      "train_speed(iter/s)": 1.628187
    },
    {
      "acc": 0.62819347,
      "epoch": 0.13419583967529172,
      "grad_norm": 7.21875,
      "learning_rate": 9.992009694373658e-06,
      "loss": 1.70785141,
      "memory(GiB)": 67.97,
      "step": 5290,
      "train_speed(iter/s)": 1.628364
    },
    {
      "acc": 0.64102268,
      "epoch": 0.1343226788432268,
      "grad_norm": 7.53125,
      "learning_rate": 9.991950325116995e-06,
      "loss": 1.72000961,
      "memory(GiB)": 67.97,
      "step": 5295,
      "train_speed(iter/s)": 1.628533
    },
    {
      "acc": 0.63425627,
      "epoch": 0.13444951801116184,
      "grad_norm": 5.3125,
      "learning_rate": 9.99189073629207e-06,
      "loss": 1.6980072,
      "memory(GiB)": 67.97,
      "step": 5300,
      "train_speed(iter/s)": 1.628696
    },
    {
      "acc": 0.62713289,
      "epoch": 0.1345763571790969,
      "grad_norm": 7.5625,
      "learning_rate": 9.991830927901505e-06,
      "loss": 1.81331329,
      "memory(GiB)": 67.97,
      "step": 5305,
      "train_speed(iter/s)": 1.628864
    },
    {
      "acc": 0.64455872,
      "epoch": 0.13470319634703196,
      "grad_norm": 5.75,
      "learning_rate": 9.991770899947925e-06,
      "loss": 1.64396591,
      "memory(GiB)": 67.97,
      "step": 5310,
      "train_speed(iter/s)": 1.629023
    },
    {
      "acc": 0.64155502,
      "epoch": 0.13483003551496703,
      "grad_norm": 7.5,
      "learning_rate": 9.991710652433977e-06,
      "loss": 1.71283951,
      "memory(GiB)": 67.97,
      "step": 5315,
      "train_speed(iter/s)": 1.629201
    },
    {
      "acc": 0.63127637,
      "epoch": 0.13495687468290207,
      "grad_norm": 5.75,
      "learning_rate": 9.991650185362308e-06,
      "loss": 1.72396545,
      "memory(GiB)": 67.97,
      "step": 5320,
      "train_speed(iter/s)": 1.629372
    },
    {
      "acc": 0.63823328,
      "epoch": 0.13508371385083714,
      "grad_norm": 6.0,
      "learning_rate": 9.991589498735577e-06,
      "loss": 1.67893753,
      "memory(GiB)": 67.97,
      "step": 5325,
      "train_speed(iter/s)": 1.629541
    },
    {
      "acc": 0.64746094,
      "epoch": 0.1352105530187722,
      "grad_norm": 7.375,
      "learning_rate": 9.991528592556454e-06,
      "loss": 1.65668163,
      "memory(GiB)": 67.97,
      "step": 5330,
      "train_speed(iter/s)": 1.629711
    },
    {
      "acc": 0.62300062,
      "epoch": 0.13533739218670726,
      "grad_norm": 6.71875,
      "learning_rate": 9.991467466827618e-06,
      "loss": 1.7772522,
      "memory(GiB)": 67.97,
      "step": 5335,
      "train_speed(iter/s)": 1.629871
    },
    {
      "acc": 0.64545665,
      "epoch": 0.1354642313546423,
      "grad_norm": 8.1875,
      "learning_rate": 9.99140612155176e-06,
      "loss": 1.73008785,
      "memory(GiB)": 67.97,
      "step": 5340,
      "train_speed(iter/s)": 1.630035
    },
    {
      "acc": 0.62435536,
      "epoch": 0.13559107052257738,
      "grad_norm": 6.03125,
      "learning_rate": 9.991344556731572e-06,
      "loss": 1.75409298,
      "memory(GiB)": 67.97,
      "step": 5345,
      "train_speed(iter/s)": 1.630215
    },
    {
      "acc": 0.65214739,
      "epoch": 0.13571790969051242,
      "grad_norm": 5.4375,
      "learning_rate": 9.991282772369766e-06,
      "loss": 1.66706543,
      "memory(GiB)": 67.97,
      "step": 5350,
      "train_speed(iter/s)": 1.630395
    },
    {
      "acc": 0.63869658,
      "epoch": 0.1358447488584475,
      "grad_norm": 6.5625,
      "learning_rate": 9.99122076846906e-06,
      "loss": 1.67649441,
      "memory(GiB)": 67.97,
      "step": 5355,
      "train_speed(iter/s)": 1.630559
    },
    {
      "acc": 0.64688721,
      "epoch": 0.13597158802638254,
      "grad_norm": 5.21875,
      "learning_rate": 9.991158545032181e-06,
      "loss": 1.69243584,
      "memory(GiB)": 67.97,
      "step": 5360,
      "train_speed(iter/s)": 1.630734
    },
    {
      "acc": 0.65743103,
      "epoch": 0.1360984271943176,
      "grad_norm": 5.5,
      "learning_rate": 9.991096102061865e-06,
      "loss": 1.63055229,
      "memory(GiB)": 67.97,
      "step": 5365,
      "train_speed(iter/s)": 1.630901
    },
    {
      "acc": 0.64526978,
      "epoch": 0.13622526636225266,
      "grad_norm": 6.84375,
      "learning_rate": 9.991033439560858e-06,
      "loss": 1.71617832,
      "memory(GiB)": 67.97,
      "step": 5370,
      "train_speed(iter/s)": 1.631076
    },
    {
      "acc": 0.63090262,
      "epoch": 0.13635210553018773,
      "grad_norm": 5.0625,
      "learning_rate": 9.990970557531918e-06,
      "loss": 1.78805809,
      "memory(GiB)": 67.97,
      "step": 5375,
      "train_speed(iter/s)": 1.631237
    },
    {
      "acc": 0.6560832,
      "epoch": 0.13647894469812277,
      "grad_norm": 6.0625,
      "learning_rate": 9.990907455977809e-06,
      "loss": 1.66885719,
      "memory(GiB)": 67.97,
      "step": 5380,
      "train_speed(iter/s)": 1.631399
    },
    {
      "acc": 0.63162436,
      "epoch": 0.13660578386605784,
      "grad_norm": 4.65625,
      "learning_rate": 9.990844134901308e-06,
      "loss": 1.75449448,
      "memory(GiB)": 67.97,
      "step": 5385,
      "train_speed(iter/s)": 1.631556
    },
    {
      "acc": 0.66275859,
      "epoch": 0.1367326230339929,
      "grad_norm": 5.3125,
      "learning_rate": 9.9907805943052e-06,
      "loss": 1.61374607,
      "memory(GiB)": 67.97,
      "step": 5390,
      "train_speed(iter/s)": 1.631707
    },
    {
      "acc": 0.63994455,
      "epoch": 0.13685946220192796,
      "grad_norm": 4.90625,
      "learning_rate": 9.990716834192278e-06,
      "loss": 1.64527721,
      "memory(GiB)": 67.97,
      "step": 5395,
      "train_speed(iter/s)": 1.631855
    },
    {
      "acc": 0.62512465,
      "epoch": 0.136986301369863,
      "grad_norm": 5.71875,
      "learning_rate": 9.990652854565348e-06,
      "loss": 1.7353159,
      "memory(GiB)": 67.97,
      "step": 5400,
      "train_speed(iter/s)": 1.632022
    },
    {
      "acc": 0.6311697,
      "epoch": 0.13711314053779808,
      "grad_norm": 6.21875,
      "learning_rate": 9.990588655427225e-06,
      "loss": 1.78660164,
      "memory(GiB)": 67.97,
      "step": 5405,
      "train_speed(iter/s)": 1.632179
    },
    {
      "acc": 0.64658265,
      "epoch": 0.13723997970573312,
      "grad_norm": 6.53125,
      "learning_rate": 9.99052423678073e-06,
      "loss": 1.69756222,
      "memory(GiB)": 67.97,
      "step": 5410,
      "train_speed(iter/s)": 1.632339
    },
    {
      "acc": 0.64642596,
      "epoch": 0.1373668188736682,
      "grad_norm": 5.75,
      "learning_rate": 9.990459598628697e-06,
      "loss": 1.65353832,
      "memory(GiB)": 67.97,
      "step": 5415,
      "train_speed(iter/s)": 1.632504
    },
    {
      "acc": 0.63903875,
      "epoch": 0.13749365804160324,
      "grad_norm": 4.75,
      "learning_rate": 9.990394740973972e-06,
      "loss": 1.63889236,
      "memory(GiB)": 67.97,
      "step": 5420,
      "train_speed(iter/s)": 1.632666
    },
    {
      "acc": 0.64777212,
      "epoch": 0.1376204972095383,
      "grad_norm": 5.84375,
      "learning_rate": 9.990329663819405e-06,
      "loss": 1.72402916,
      "memory(GiB)": 67.97,
      "step": 5425,
      "train_speed(iter/s)": 1.632839
    },
    {
      "acc": 0.64259725,
      "epoch": 0.13774733637747336,
      "grad_norm": 5.53125,
      "learning_rate": 9.99026436716786e-06,
      "loss": 1.6757103,
      "memory(GiB)": 67.97,
      "step": 5430,
      "train_speed(iter/s)": 1.632997
    },
    {
      "acc": 0.63455553,
      "epoch": 0.13787417554540843,
      "grad_norm": 4.6875,
      "learning_rate": 9.990198851022207e-06,
      "loss": 1.69476357,
      "memory(GiB)": 67.97,
      "step": 5435,
      "train_speed(iter/s)": 1.63316
    },
    {
      "acc": 0.64285312,
      "epoch": 0.13800101471334347,
      "grad_norm": 5.84375,
      "learning_rate": 9.99013311538533e-06,
      "loss": 1.69048004,
      "memory(GiB)": 67.97,
      "step": 5440,
      "train_speed(iter/s)": 1.633312
    },
    {
      "acc": 0.64904666,
      "epoch": 0.13812785388127855,
      "grad_norm": 6.53125,
      "learning_rate": 9.99006716026012e-06,
      "loss": 1.71266918,
      "memory(GiB)": 67.97,
      "step": 5445,
      "train_speed(iter/s)": 1.633475
    },
    {
      "acc": 0.64135752,
      "epoch": 0.1382546930492136,
      "grad_norm": 5.25,
      "learning_rate": 9.990000985649475e-06,
      "loss": 1.73248444,
      "memory(GiB)": 67.97,
      "step": 5450,
      "train_speed(iter/s)": 1.633645
    },
    {
      "acc": 0.63617897,
      "epoch": 0.13838153221714866,
      "grad_norm": 6.71875,
      "learning_rate": 9.989934591556308e-06,
      "loss": 1.69448586,
      "memory(GiB)": 67.97,
      "step": 5455,
      "train_speed(iter/s)": 1.633809
    },
    {
      "acc": 0.62789268,
      "epoch": 0.1385083713850837,
      "grad_norm": 6.65625,
      "learning_rate": 9.98986797798354e-06,
      "loss": 1.75231285,
      "memory(GiB)": 67.97,
      "step": 5460,
      "train_speed(iter/s)": 1.633962
    },
    {
      "acc": 0.64234023,
      "epoch": 0.13863521055301878,
      "grad_norm": 7.0,
      "learning_rate": 9.989801144934102e-06,
      "loss": 1.68317013,
      "memory(GiB)": 67.97,
      "step": 5465,
      "train_speed(iter/s)": 1.634124
    },
    {
      "acc": 0.64935846,
      "epoch": 0.13876204972095382,
      "grad_norm": 6.375,
      "learning_rate": 9.98973409241093e-06,
      "loss": 1.68969078,
      "memory(GiB)": 67.97,
      "step": 5470,
      "train_speed(iter/s)": 1.634283
    },
    {
      "acc": 0.63298841,
      "epoch": 0.1388888888888889,
      "grad_norm": 5.40625,
      "learning_rate": 9.989666820416974e-06,
      "loss": 1.71576881,
      "memory(GiB)": 67.97,
      "step": 5475,
      "train_speed(iter/s)": 1.634457
    },
    {
      "acc": 0.6435339,
      "epoch": 0.13901572805682394,
      "grad_norm": 7.8125,
      "learning_rate": 9.989599328955195e-06,
      "loss": 1.73315163,
      "memory(GiB)": 67.97,
      "step": 5480,
      "train_speed(iter/s)": 1.634623
    },
    {
      "acc": 0.64531841,
      "epoch": 0.139142567224759,
      "grad_norm": 4.75,
      "learning_rate": 9.98953161802856e-06,
      "loss": 1.68681412,
      "memory(GiB)": 67.97,
      "step": 5485,
      "train_speed(iter/s)": 1.634772
    },
    {
      "acc": 0.65644813,
      "epoch": 0.13926940639269406,
      "grad_norm": 6.59375,
      "learning_rate": 9.98946368764005e-06,
      "loss": 1.65063095,
      "memory(GiB)": 67.97,
      "step": 5490,
      "train_speed(iter/s)": 1.634937
    },
    {
      "acc": 0.66878214,
      "epoch": 0.13939624556062913,
      "grad_norm": 7.0625,
      "learning_rate": 9.989395537792647e-06,
      "loss": 1.66400394,
      "memory(GiB)": 67.97,
      "step": 5495,
      "train_speed(iter/s)": 1.635093
    },
    {
      "acc": 0.63075151,
      "epoch": 0.13952308472856417,
      "grad_norm": 6.1875,
      "learning_rate": 9.989327168489356e-06,
      "loss": 1.7121563,
      "memory(GiB)": 67.97,
      "step": 5500,
      "train_speed(iter/s)": 1.635255
    },
    {
      "acc": 0.64085975,
      "epoch": 0.13964992389649925,
      "grad_norm": 5.15625,
      "learning_rate": 9.989258579733179e-06,
      "loss": 1.71124191,
      "memory(GiB)": 67.97,
      "step": 5505,
      "train_speed(iter/s)": 1.635401
    },
    {
      "acc": 0.63293228,
      "epoch": 0.1397767630644343,
      "grad_norm": 5.78125,
      "learning_rate": 9.989189771527133e-06,
      "loss": 1.67303009,
      "memory(GiB)": 67.97,
      "step": 5510,
      "train_speed(iter/s)": 1.63556
    },
    {
      "acc": 0.63710937,
      "epoch": 0.13990360223236936,
      "grad_norm": 5.15625,
      "learning_rate": 9.989120743874248e-06,
      "loss": 1.73125992,
      "memory(GiB)": 67.97,
      "step": 5515,
      "train_speed(iter/s)": 1.635716
    },
    {
      "acc": 0.62786932,
      "epoch": 0.1400304414003044,
      "grad_norm": 8.625,
      "learning_rate": 9.989051496777556e-06,
      "loss": 1.7746994,
      "memory(GiB)": 67.97,
      "step": 5520,
      "train_speed(iter/s)": 1.635886
    },
    {
      "acc": 0.63814631,
      "epoch": 0.14015728056823948,
      "grad_norm": 5.5625,
      "learning_rate": 9.988982030240104e-06,
      "loss": 1.73342056,
      "memory(GiB)": 67.97,
      "step": 5525,
      "train_speed(iter/s)": 1.636032
    },
    {
      "acc": 0.64379349,
      "epoch": 0.14028411973617452,
      "grad_norm": 6.78125,
      "learning_rate": 9.988912344264949e-06,
      "loss": 1.77013035,
      "memory(GiB)": 67.97,
      "step": 5530,
      "train_speed(iter/s)": 1.636192
    },
    {
      "acc": 0.63793926,
      "epoch": 0.1404109589041096,
      "grad_norm": 4.875,
      "learning_rate": 9.988842438855156e-06,
      "loss": 1.63599529,
      "memory(GiB)": 67.97,
      "step": 5535,
      "train_speed(iter/s)": 1.636346
    },
    {
      "acc": 0.63129282,
      "epoch": 0.14053779807204464,
      "grad_norm": 6.4375,
      "learning_rate": 9.988772314013799e-06,
      "loss": 1.70913277,
      "memory(GiB)": 67.97,
      "step": 5540,
      "train_speed(iter/s)": 1.636497
    },
    {
      "acc": 0.63681321,
      "epoch": 0.1406646372399797,
      "grad_norm": 5.96875,
      "learning_rate": 9.988701969743961e-06,
      "loss": 1.70088158,
      "memory(GiB)": 67.97,
      "step": 5545,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.63065233,
      "epoch": 0.14079147640791476,
      "grad_norm": 4.46875,
      "learning_rate": 9.98863140604874e-06,
      "loss": 1.73746414,
      "memory(GiB)": 67.97,
      "step": 5550,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.64270663,
      "epoch": 0.14091831557584983,
      "grad_norm": 5.40625,
      "learning_rate": 9.988560622931233e-06,
      "loss": 1.6762867,
      "memory(GiB)": 67.97,
      "step": 5555,
      "train_speed(iter/s)": 1.636971
    },
    {
      "acc": 0.65530734,
      "epoch": 0.14104515474378487,
      "grad_norm": 5.375,
      "learning_rate": 9.988489620394562e-06,
      "loss": 1.63868313,
      "memory(GiB)": 67.97,
      "step": 5560,
      "train_speed(iter/s)": 1.637134
    },
    {
      "acc": 0.65375457,
      "epoch": 0.14117199391171995,
      "grad_norm": 5.4375,
      "learning_rate": 9.988418398441842e-06,
      "loss": 1.65380325,
      "memory(GiB)": 67.97,
      "step": 5565,
      "train_speed(iter/s)": 1.637292
    },
    {
      "acc": 0.6210989,
      "epoch": 0.141298833079655,
      "grad_norm": 6.1875,
      "learning_rate": 9.98834695707621e-06,
      "loss": 1.80401039,
      "memory(GiB)": 67.97,
      "step": 5570,
      "train_speed(iter/s)": 1.637461
    },
    {
      "acc": 0.62638574,
      "epoch": 0.14142567224759006,
      "grad_norm": 4.96875,
      "learning_rate": 9.98827529630081e-06,
      "loss": 1.72670822,
      "memory(GiB)": 67.97,
      "step": 5575,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.64285045,
      "epoch": 0.1415525114155251,
      "grad_norm": 5.40625,
      "learning_rate": 9.988203416118788e-06,
      "loss": 1.63737946,
      "memory(GiB)": 67.97,
      "step": 5580,
      "train_speed(iter/s)": 1.637773
    },
    {
      "acc": 0.65205193,
      "epoch": 0.14167935058346018,
      "grad_norm": 5.40625,
      "learning_rate": 9.98813131653331e-06,
      "loss": 1.6648632,
      "memory(GiB)": 67.97,
      "step": 5585,
      "train_speed(iter/s)": 1.637925
    },
    {
      "acc": 0.63377504,
      "epoch": 0.14180618975139522,
      "grad_norm": 5.65625,
      "learning_rate": 9.988058997547548e-06,
      "loss": 1.70777187,
      "memory(GiB)": 67.97,
      "step": 5590,
      "train_speed(iter/s)": 1.63807
    },
    {
      "acc": 0.63157835,
      "epoch": 0.1419330289193303,
      "grad_norm": 10.75,
      "learning_rate": 9.987986459164678e-06,
      "loss": 1.80104675,
      "memory(GiB)": 67.97,
      "step": 5595,
      "train_speed(iter/s)": 1.638239
    },
    {
      "acc": 0.6438448,
      "epoch": 0.14205986808726534,
      "grad_norm": 6.3125,
      "learning_rate": 9.987913701387897e-06,
      "loss": 1.69255219,
      "memory(GiB)": 67.97,
      "step": 5600,
      "train_speed(iter/s)": 1.638393
    },
    {
      "acc": 0.63091969,
      "epoch": 0.1421867072552004,
      "grad_norm": 5.15625,
      "learning_rate": 9.9878407242204e-06,
      "loss": 1.75262699,
      "memory(GiB)": 67.97,
      "step": 5605,
      "train_speed(iter/s)": 1.63855
    },
    {
      "acc": 0.63085594,
      "epoch": 0.14231354642313546,
      "grad_norm": 6.3125,
      "learning_rate": 9.9877675276654e-06,
      "loss": 1.6860054,
      "memory(GiB)": 67.97,
      "step": 5610,
      "train_speed(iter/s)": 1.638696
    },
    {
      "acc": 0.64327564,
      "epoch": 0.14244038559107053,
      "grad_norm": 5.09375,
      "learning_rate": 9.987694111726114e-06,
      "loss": 1.64752007,
      "memory(GiB)": 67.97,
      "step": 5615,
      "train_speed(iter/s)": 1.638851
    },
    {
      "acc": 0.64797969,
      "epoch": 0.14256722475900557,
      "grad_norm": 5.15625,
      "learning_rate": 9.987620476405774e-06,
      "loss": 1.66869946,
      "memory(GiB)": 67.97,
      "step": 5620,
      "train_speed(iter/s)": 1.638997
    },
    {
      "acc": 0.65582399,
      "epoch": 0.14269406392694065,
      "grad_norm": 5.40625,
      "learning_rate": 9.987546621707616e-06,
      "loss": 1.67556496,
      "memory(GiB)": 67.97,
      "step": 5625,
      "train_speed(iter/s)": 1.639138
    },
    {
      "acc": 0.63089805,
      "epoch": 0.1428209030948757,
      "grad_norm": 6.0,
      "learning_rate": 9.98747254763489e-06,
      "loss": 1.71389046,
      "memory(GiB)": 67.97,
      "step": 5630,
      "train_speed(iter/s)": 1.639295
    },
    {
      "acc": 0.62215295,
      "epoch": 0.14294774226281076,
      "grad_norm": 5.84375,
      "learning_rate": 9.987398254190855e-06,
      "loss": 1.75646858,
      "memory(GiB)": 67.97,
      "step": 5635,
      "train_speed(iter/s)": 1.639445
    },
    {
      "acc": 0.64388542,
      "epoch": 0.1430745814307458,
      "grad_norm": 5.75,
      "learning_rate": 9.987323741378777e-06,
      "loss": 1.72612591,
      "memory(GiB)": 67.97,
      "step": 5640,
      "train_speed(iter/s)": 1.639604
    },
    {
      "acc": 0.66521358,
      "epoch": 0.14320142059868088,
      "grad_norm": 6.125,
      "learning_rate": 9.987249009201934e-06,
      "loss": 1.62808895,
      "memory(GiB)": 67.97,
      "step": 5645,
      "train_speed(iter/s)": 1.639755
    },
    {
      "acc": 0.64707842,
      "epoch": 0.14332825976661592,
      "grad_norm": 7.59375,
      "learning_rate": 9.987174057663613e-06,
      "loss": 1.65816956,
      "memory(GiB)": 67.97,
      "step": 5650,
      "train_speed(iter/s)": 1.639911
    },
    {
      "acc": 0.63463721,
      "epoch": 0.143455098934551,
      "grad_norm": 6.46875,
      "learning_rate": 9.987098886767111e-06,
      "loss": 1.697579,
      "memory(GiB)": 67.97,
      "step": 5655,
      "train_speed(iter/s)": 1.640061
    },
    {
      "acc": 0.62340398,
      "epoch": 0.14358193810248604,
      "grad_norm": 5.84375,
      "learning_rate": 9.987023496515734e-06,
      "loss": 1.80572548,
      "memory(GiB)": 67.97,
      "step": 5660,
      "train_speed(iter/s)": 1.640203
    },
    {
      "acc": 0.64071021,
      "epoch": 0.1437087772704211,
      "grad_norm": 8.1875,
      "learning_rate": 9.9869478869128e-06,
      "loss": 1.71325188,
      "memory(GiB)": 67.97,
      "step": 5665,
      "train_speed(iter/s)": 1.640358
    },
    {
      "acc": 0.63145103,
      "epoch": 0.14383561643835616,
      "grad_norm": 5.625,
      "learning_rate": 9.98687205796163e-06,
      "loss": 1.76592846,
      "memory(GiB)": 67.97,
      "step": 5670,
      "train_speed(iter/s)": 1.640488
    },
    {
      "acc": 0.64574471,
      "epoch": 0.14396245560629123,
      "grad_norm": 6.25,
      "learning_rate": 9.986796009665562e-06,
      "loss": 1.64609814,
      "memory(GiB)": 67.97,
      "step": 5675,
      "train_speed(iter/s)": 1.640646
    },
    {
      "acc": 0.64121971,
      "epoch": 0.14408929477422627,
      "grad_norm": 4.78125,
      "learning_rate": 9.986719742027944e-06,
      "loss": 1.63788223,
      "memory(GiB)": 67.97,
      "step": 5680,
      "train_speed(iter/s)": 1.640796
    },
    {
      "acc": 0.6402359,
      "epoch": 0.14421613394216135,
      "grad_norm": 7.09375,
      "learning_rate": 9.986643255052125e-06,
      "loss": 1.67876205,
      "memory(GiB)": 67.97,
      "step": 5685,
      "train_speed(iter/s)": 1.640942
    },
    {
      "acc": 0.63779802,
      "epoch": 0.1443429731100964,
      "grad_norm": 5.0,
      "learning_rate": 9.986566548741473e-06,
      "loss": 1.69805088,
      "memory(GiB)": 67.97,
      "step": 5690,
      "train_speed(iter/s)": 1.641099
    },
    {
      "acc": 0.65231657,
      "epoch": 0.14446981227803146,
      "grad_norm": 5.5625,
      "learning_rate": 9.98648962309936e-06,
      "loss": 1.68096237,
      "memory(GiB)": 67.97,
      "step": 5695,
      "train_speed(iter/s)": 1.64124
    },
    {
      "acc": 0.64773283,
      "epoch": 0.1445966514459665,
      "grad_norm": 6.84375,
      "learning_rate": 9.986412478129171e-06,
      "loss": 1.61987782,
      "memory(GiB)": 67.97,
      "step": 5700,
      "train_speed(iter/s)": 1.64139
    },
    {
      "acc": 0.64352512,
      "epoch": 0.14472349061390158,
      "grad_norm": 8.4375,
      "learning_rate": 9.9863351138343e-06,
      "loss": 1.68878593,
      "memory(GiB)": 67.97,
      "step": 5705,
      "train_speed(iter/s)": 1.641544
    },
    {
      "acc": 0.64809437,
      "epoch": 0.14485032978183662,
      "grad_norm": 6.25,
      "learning_rate": 9.986257530218146e-06,
      "loss": 1.6550457,
      "memory(GiB)": 67.97,
      "step": 5710,
      "train_speed(iter/s)": 1.641698
    },
    {
      "acc": 0.6398253,
      "epoch": 0.1449771689497717,
      "grad_norm": 5.25,
      "learning_rate": 9.986179727284124e-06,
      "loss": 1.6624176,
      "memory(GiB)": 67.97,
      "step": 5715,
      "train_speed(iter/s)": 1.641846
    },
    {
      "acc": 0.62591176,
      "epoch": 0.14510400811770674,
      "grad_norm": 7.75,
      "learning_rate": 9.986101705035656e-06,
      "loss": 1.75093708,
      "memory(GiB)": 67.97,
      "step": 5720,
      "train_speed(iter/s)": 1.641979
    },
    {
      "acc": 0.62979212,
      "epoch": 0.1452308472856418,
      "grad_norm": 5.8125,
      "learning_rate": 9.986023463476175e-06,
      "loss": 1.76803741,
      "memory(GiB)": 67.97,
      "step": 5725,
      "train_speed(iter/s)": 1.642124
    },
    {
      "acc": 0.63339853,
      "epoch": 0.14535768645357686,
      "grad_norm": 6.3125,
      "learning_rate": 9.985945002609119e-06,
      "loss": 1.7575386,
      "memory(GiB)": 67.97,
      "step": 5730,
      "train_speed(iter/s)": 1.642276
    },
    {
      "acc": 0.63349552,
      "epoch": 0.14548452562151193,
      "grad_norm": 6.84375,
      "learning_rate": 9.985866322437942e-06,
      "loss": 1.77933197,
      "memory(GiB)": 67.97,
      "step": 5735,
      "train_speed(iter/s)": 1.642426
    },
    {
      "acc": 0.64729791,
      "epoch": 0.14561136478944697,
      "grad_norm": 7.1875,
      "learning_rate": 9.985787422966105e-06,
      "loss": 1.67173462,
      "memory(GiB)": 67.97,
      "step": 5740,
      "train_speed(iter/s)": 1.642566
    },
    {
      "acc": 0.64463506,
      "epoch": 0.14573820395738205,
      "grad_norm": 6.71875,
      "learning_rate": 9.985708304197075e-06,
      "loss": 1.78628616,
      "memory(GiB)": 67.97,
      "step": 5745,
      "train_speed(iter/s)": 1.642714
    },
    {
      "acc": 0.64048452,
      "epoch": 0.1458650431253171,
      "grad_norm": 5.34375,
      "learning_rate": 9.985628966134336e-06,
      "loss": 1.70549278,
      "memory(GiB)": 67.97,
      "step": 5750,
      "train_speed(iter/s)": 1.642866
    },
    {
      "acc": 0.63640237,
      "epoch": 0.14599188229325216,
      "grad_norm": 7.0625,
      "learning_rate": 9.985549408781377e-06,
      "loss": 1.69229431,
      "memory(GiB)": 67.97,
      "step": 5755,
      "train_speed(iter/s)": 1.642996
    },
    {
      "acc": 0.63343587,
      "epoch": 0.1461187214611872,
      "grad_norm": 5.28125,
      "learning_rate": 9.985469632141693e-06,
      "loss": 1.7327282,
      "memory(GiB)": 67.97,
      "step": 5760,
      "train_speed(iter/s)": 1.643145
    },
    {
      "acc": 0.635888,
      "epoch": 0.14624556062912228,
      "grad_norm": 5.15625,
      "learning_rate": 9.985389636218797e-06,
      "loss": 1.69949722,
      "memory(GiB)": 67.97,
      "step": 5765,
      "train_speed(iter/s)": 1.64328
    },
    {
      "acc": 0.6408618,
      "epoch": 0.14637239979705732,
      "grad_norm": 5.8125,
      "learning_rate": 9.985309421016207e-06,
      "loss": 1.70303917,
      "memory(GiB)": 67.97,
      "step": 5770,
      "train_speed(iter/s)": 1.64342
    },
    {
      "acc": 0.65503311,
      "epoch": 0.1464992389649924,
      "grad_norm": 5.6875,
      "learning_rate": 9.985228986537451e-06,
      "loss": 1.67711391,
      "memory(GiB)": 67.97,
      "step": 5775,
      "train_speed(iter/s)": 1.643574
    },
    {
      "acc": 0.64052763,
      "epoch": 0.14662607813292744,
      "grad_norm": 5.71875,
      "learning_rate": 9.985148332786068e-06,
      "loss": 1.6546669,
      "memory(GiB)": 67.97,
      "step": 5780,
      "train_speed(iter/s)": 1.643721
    },
    {
      "acc": 0.63725519,
      "epoch": 0.1467529173008625,
      "grad_norm": 6.0,
      "learning_rate": 9.985067459765603e-06,
      "loss": 1.72509747,
      "memory(GiB)": 67.97,
      "step": 5785,
      "train_speed(iter/s)": 1.643875
    },
    {
      "acc": 0.64202337,
      "epoch": 0.14687975646879756,
      "grad_norm": 5.75,
      "learning_rate": 9.984986367479615e-06,
      "loss": 1.73176651,
      "memory(GiB)": 67.97,
      "step": 5790,
      "train_speed(iter/s)": 1.644012
    },
    {
      "acc": 0.64810824,
      "epoch": 0.14700659563673263,
      "grad_norm": 5.8125,
      "learning_rate": 9.984905055931668e-06,
      "loss": 1.67020073,
      "memory(GiB)": 67.97,
      "step": 5795,
      "train_speed(iter/s)": 1.644161
    },
    {
      "acc": 0.61541805,
      "epoch": 0.14713343480466767,
      "grad_norm": 5.1875,
      "learning_rate": 9.984823525125342e-06,
      "loss": 1.7698822,
      "memory(GiB)": 67.97,
      "step": 5800,
      "train_speed(iter/s)": 1.644316
    },
    {
      "acc": 0.63629761,
      "epoch": 0.14726027397260275,
      "grad_norm": 5.25,
      "learning_rate": 9.984741775064222e-06,
      "loss": 1.72454681,
      "memory(GiB)": 67.97,
      "step": 5805,
      "train_speed(iter/s)": 1.64446
    },
    {
      "acc": 0.62756248,
      "epoch": 0.1473871131405378,
      "grad_norm": 5.03125,
      "learning_rate": 9.984659805751904e-06,
      "loss": 1.77317219,
      "memory(GiB)": 67.97,
      "step": 5810,
      "train_speed(iter/s)": 1.644601
    },
    {
      "acc": 0.65102062,
      "epoch": 0.14751395230847286,
      "grad_norm": 5.25,
      "learning_rate": 9.984577617191993e-06,
      "loss": 1.66598091,
      "memory(GiB)": 67.97,
      "step": 5815,
      "train_speed(iter/s)": 1.644747
    },
    {
      "acc": 0.63081174,
      "epoch": 0.1476407914764079,
      "grad_norm": 6.28125,
      "learning_rate": 9.984495209388102e-06,
      "loss": 1.74582272,
      "memory(GiB)": 67.97,
      "step": 5820,
      "train_speed(iter/s)": 1.644897
    },
    {
      "acc": 0.63116512,
      "epoch": 0.14776763064434298,
      "grad_norm": 6.34375,
      "learning_rate": 9.984412582343859e-06,
      "loss": 1.76524906,
      "memory(GiB)": 67.97,
      "step": 5825,
      "train_speed(iter/s)": 1.645042
    },
    {
      "acc": 0.64548478,
      "epoch": 0.14789446981227802,
      "grad_norm": 6.0,
      "learning_rate": 9.984329736062896e-06,
      "loss": 1.64662361,
      "memory(GiB)": 67.97,
      "step": 5830,
      "train_speed(iter/s)": 1.645193
    },
    {
      "acc": 0.62500458,
      "epoch": 0.1480213089802131,
      "grad_norm": 5.46875,
      "learning_rate": 9.984246670548858e-06,
      "loss": 1.72222424,
      "memory(GiB)": 67.97,
      "step": 5835,
      "train_speed(iter/s)": 1.645333
    },
    {
      "acc": 0.640452,
      "epoch": 0.14814814814814814,
      "grad_norm": 5.15625,
      "learning_rate": 9.984163385805398e-06,
      "loss": 1.74711456,
      "memory(GiB)": 67.97,
      "step": 5840,
      "train_speed(iter/s)": 1.645473
    },
    {
      "acc": 0.63930683,
      "epoch": 0.1482749873160832,
      "grad_norm": 7.28125,
      "learning_rate": 9.984079881836182e-06,
      "loss": 1.71177349,
      "memory(GiB)": 67.97,
      "step": 5845,
      "train_speed(iter/s)": 1.645621
    },
    {
      "acc": 0.63935652,
      "epoch": 0.14840182648401826,
      "grad_norm": 6.9375,
      "learning_rate": 9.983996158644877e-06,
      "loss": 1.6764328,
      "memory(GiB)": 67.97,
      "step": 5850,
      "train_speed(iter/s)": 1.645755
    },
    {
      "acc": 0.64308186,
      "epoch": 0.14852866565195333,
      "grad_norm": 4.5,
      "learning_rate": 9.983912216235172e-06,
      "loss": 1.60972023,
      "memory(GiB)": 67.97,
      "step": 5855,
      "train_speed(iter/s)": 1.645904
    },
    {
      "acc": 0.64627028,
      "epoch": 0.14865550481988837,
      "grad_norm": 8.6875,
      "learning_rate": 9.983828054610754e-06,
      "loss": 1.69884377,
      "memory(GiB)": 67.97,
      "step": 5860,
      "train_speed(iter/s)": 1.646051
    },
    {
      "acc": 0.64437056,
      "epoch": 0.14878234398782345,
      "grad_norm": 5.75,
      "learning_rate": 9.983743673775328e-06,
      "loss": 1.73473854,
      "memory(GiB)": 67.97,
      "step": 5865,
      "train_speed(iter/s)": 1.646185
    },
    {
      "acc": 0.64316578,
      "epoch": 0.1489091831557585,
      "grad_norm": 7.0,
      "learning_rate": 9.983659073732604e-06,
      "loss": 1.67449303,
      "memory(GiB)": 67.97,
      "step": 5870,
      "train_speed(iter/s)": 1.646339
    },
    {
      "acc": 0.64334426,
      "epoch": 0.14903602232369356,
      "grad_norm": 6.78125,
      "learning_rate": 9.983574254486303e-06,
      "loss": 1.70057468,
      "memory(GiB)": 67.97,
      "step": 5875,
      "train_speed(iter/s)": 1.646484
    },
    {
      "acc": 0.6575243,
      "epoch": 0.1491628614916286,
      "grad_norm": 4.8125,
      "learning_rate": 9.983489216040158e-06,
      "loss": 1.67150078,
      "memory(GiB)": 67.97,
      "step": 5880,
      "train_speed(iter/s)": 1.646621
    },
    {
      "acc": 0.64971218,
      "epoch": 0.14928970065956368,
      "grad_norm": 5.40625,
      "learning_rate": 9.983403958397907e-06,
      "loss": 1.69433556,
      "memory(GiB)": 67.97,
      "step": 5885,
      "train_speed(iter/s)": 1.646766
    },
    {
      "acc": 0.63386822,
      "epoch": 0.14941653982749872,
      "grad_norm": 5.65625,
      "learning_rate": 9.9833184815633e-06,
      "loss": 1.86196461,
      "memory(GiB)": 67.97,
      "step": 5890,
      "train_speed(iter/s)": 1.646909
    },
    {
      "acc": 0.63559723,
      "epoch": 0.1495433789954338,
      "grad_norm": 5.1875,
      "learning_rate": 9.983232785540097e-06,
      "loss": 1.79789734,
      "memory(GiB)": 67.97,
      "step": 5895,
      "train_speed(iter/s)": 1.647054
    },
    {
      "acc": 0.63170757,
      "epoch": 0.14967021816336884,
      "grad_norm": 5.9375,
      "learning_rate": 9.983146870332068e-06,
      "loss": 1.69956932,
      "memory(GiB)": 67.97,
      "step": 5900,
      "train_speed(iter/s)": 1.647196
    },
    {
      "acc": 0.63182459,
      "epoch": 0.1497970573313039,
      "grad_norm": 10.8125,
      "learning_rate": 9.98306073594299e-06,
      "loss": 1.73979073,
      "memory(GiB)": 67.97,
      "step": 5905,
      "train_speed(iter/s)": 1.647341
    },
    {
      "acc": 0.65083694,
      "epoch": 0.14992389649923896,
      "grad_norm": 6.84375,
      "learning_rate": 9.982974382376656e-06,
      "loss": 1.67259693,
      "memory(GiB)": 67.97,
      "step": 5910,
      "train_speed(iter/s)": 1.647474
    },
    {
      "acc": 0.65537863,
      "epoch": 0.15005073566717403,
      "grad_norm": 5.9375,
      "learning_rate": 9.98288780963686e-06,
      "loss": 1.65226727,
      "memory(GiB)": 67.97,
      "step": 5915,
      "train_speed(iter/s)": 1.64761
    },
    {
      "acc": 0.63600521,
      "epoch": 0.15017757483510907,
      "grad_norm": 4.6875,
      "learning_rate": 9.98280101772741e-06,
      "loss": 1.71045208,
      "memory(GiB)": 67.97,
      "step": 5920,
      "train_speed(iter/s)": 1.647745
    },
    {
      "acc": 0.63900399,
      "epoch": 0.15030441400304415,
      "grad_norm": 5.25,
      "learning_rate": 9.982714006652126e-06,
      "loss": 1.7458086,
      "memory(GiB)": 67.97,
      "step": 5925,
      "train_speed(iter/s)": 1.647878
    },
    {
      "acc": 0.63540921,
      "epoch": 0.1504312531709792,
      "grad_norm": 5.0,
      "learning_rate": 9.982626776414834e-06,
      "loss": 1.68440323,
      "memory(GiB)": 67.97,
      "step": 5930,
      "train_speed(iter/s)": 1.648008
    },
    {
      "acc": 0.62862368,
      "epoch": 0.15055809233891426,
      "grad_norm": 5.21875,
      "learning_rate": 9.98253932701937e-06,
      "loss": 1.70194435,
      "memory(GiB)": 67.97,
      "step": 5935,
      "train_speed(iter/s)": 1.648144
    },
    {
      "acc": 0.64620218,
      "epoch": 0.1506849315068493,
      "grad_norm": 7.0625,
      "learning_rate": 9.98245165846958e-06,
      "loss": 1.75366306,
      "memory(GiB)": 67.97,
      "step": 5940,
      "train_speed(iter/s)": 1.648281
    },
    {
      "acc": 0.64474354,
      "epoch": 0.15081177067478438,
      "grad_norm": 7.875,
      "learning_rate": 9.982363770769323e-06,
      "loss": 1.67810211,
      "memory(GiB)": 67.97,
      "step": 5945,
      "train_speed(iter/s)": 1.648428
    },
    {
      "acc": 0.64211788,
      "epoch": 0.15093860984271942,
      "grad_norm": 5.78125,
      "learning_rate": 9.98227566392246e-06,
      "loss": 1.68489761,
      "memory(GiB)": 67.97,
      "step": 5950,
      "train_speed(iter/s)": 1.648557
    },
    {
      "acc": 0.64047265,
      "epoch": 0.1510654490106545,
      "grad_norm": 7.25,
      "learning_rate": 9.982187337932871e-06,
      "loss": 1.67838707,
      "memory(GiB)": 67.97,
      "step": 5955,
      "train_speed(iter/s)": 1.648693
    },
    {
      "acc": 0.62222662,
      "epoch": 0.15119228817858954,
      "grad_norm": 4.96875,
      "learning_rate": 9.98209879280444e-06,
      "loss": 1.76731186,
      "memory(GiB)": 67.97,
      "step": 5960,
      "train_speed(iter/s)": 1.648837
    },
    {
      "acc": 0.63446589,
      "epoch": 0.1513191273465246,
      "grad_norm": 6.15625,
      "learning_rate": 9.982010028541057e-06,
      "loss": 1.73758335,
      "memory(GiB)": 67.97,
      "step": 5965,
      "train_speed(iter/s)": 1.64898
    },
    {
      "acc": 0.64323282,
      "epoch": 0.15144596651445966,
      "grad_norm": 6.46875,
      "learning_rate": 9.981921045146633e-06,
      "loss": 1.66124744,
      "memory(GiB)": 67.97,
      "step": 5970,
      "train_speed(iter/s)": 1.649117
    },
    {
      "acc": 0.64692993,
      "epoch": 0.15157280568239473,
      "grad_norm": 5.59375,
      "learning_rate": 9.981831842625079e-06,
      "loss": 1.64683552,
      "memory(GiB)": 67.97,
      "step": 5975,
      "train_speed(iter/s)": 1.649254
    },
    {
      "acc": 0.64014034,
      "epoch": 0.15169964485032977,
      "grad_norm": 5.59375,
      "learning_rate": 9.981742420980316e-06,
      "loss": 1.7176815,
      "memory(GiB)": 67.97,
      "step": 5980,
      "train_speed(iter/s)": 1.649399
    },
    {
      "acc": 0.64055929,
      "epoch": 0.15182648401826485,
      "grad_norm": 6.8125,
      "learning_rate": 9.981652780216281e-06,
      "loss": 1.70856915,
      "memory(GiB)": 67.97,
      "step": 5985,
      "train_speed(iter/s)": 1.649535
    },
    {
      "acc": 0.6607892,
      "epoch": 0.1519533231861999,
      "grad_norm": 8.0,
      "learning_rate": 9.981562920336915e-06,
      "loss": 1.61497307,
      "memory(GiB)": 67.97,
      "step": 5990,
      "train_speed(iter/s)": 1.64968
    },
    {
      "acc": 0.62692347,
      "epoch": 0.15208016235413496,
      "grad_norm": 5.96875,
      "learning_rate": 9.98147284134617e-06,
      "loss": 1.7615097,
      "memory(GiB)": 67.97,
      "step": 5995,
      "train_speed(iter/s)": 1.649817
    },
    {
      "acc": 0.63626213,
      "epoch": 0.15220700152207,
      "grad_norm": 6.4375,
      "learning_rate": 9.981382543248011e-06,
      "loss": 1.73356247,
      "memory(GiB)": 67.97,
      "step": 6000,
      "train_speed(iter/s)": 1.649948
    },
    {
      "epoch": 0.15220700152207,
      "eval_acc": 0.631341715794068,
      "eval_loss": 1.6586743593215942,
      "eval_runtime": 58.2386,
      "eval_samples_per_second": 109.378,
      "eval_steps_per_second": 27.353,
      "step": 6000
    },
    {
      "acc": 0.64247751,
      "epoch": 0.15233384069000508,
      "grad_norm": 5.65625,
      "learning_rate": 9.981292026046406e-06,
      "loss": 1.67188072,
      "memory(GiB)": 67.97,
      "step": 6005,
      "train_speed(iter/s)": 1.622245
    },
    {
      "acc": 0.63753009,
      "epoch": 0.15246067985794012,
      "grad_norm": 6.46875,
      "learning_rate": 9.981201289745337e-06,
      "loss": 1.69720078,
      "memory(GiB)": 67.97,
      "step": 6010,
      "train_speed(iter/s)": 1.622371
    },
    {
      "acc": 0.63781967,
      "epoch": 0.1525875190258752,
      "grad_norm": 5.71875,
      "learning_rate": 9.981110334348796e-06,
      "loss": 1.73426647,
      "memory(GiB)": 67.97,
      "step": 6015,
      "train_speed(iter/s)": 1.622529
    },
    {
      "acc": 0.65269566,
      "epoch": 0.15271435819381024,
      "grad_norm": 6.84375,
      "learning_rate": 9.981019159860782e-06,
      "loss": 1.63263931,
      "memory(GiB)": 67.97,
      "step": 6020,
      "train_speed(iter/s)": 1.62268
    },
    {
      "acc": 0.64456654,
      "epoch": 0.1528411973617453,
      "grad_norm": 7.65625,
      "learning_rate": 9.98092776628531e-06,
      "loss": 1.69595184,
      "memory(GiB)": 67.97,
      "step": 6025,
      "train_speed(iter/s)": 1.622826
    },
    {
      "acc": 0.64955573,
      "epoch": 0.15296803652968036,
      "grad_norm": 4.96875,
      "learning_rate": 9.980836153626396e-06,
      "loss": 1.64122925,
      "memory(GiB)": 67.97,
      "step": 6030,
      "train_speed(iter/s)": 1.622974
    },
    {
      "acc": 0.65374017,
      "epoch": 0.15309487569761543,
      "grad_norm": 5.75,
      "learning_rate": 9.980744321888068e-06,
      "loss": 1.68965683,
      "memory(GiB)": 67.97,
      "step": 6035,
      "train_speed(iter/s)": 1.623118
    },
    {
      "acc": 0.63661962,
      "epoch": 0.15322171486555047,
      "grad_norm": 5.15625,
      "learning_rate": 9.98065227107437e-06,
      "loss": 1.78869629,
      "memory(GiB)": 67.97,
      "step": 6040,
      "train_speed(iter/s)": 1.623279
    },
    {
      "acc": 0.63980098,
      "epoch": 0.15334855403348555,
      "grad_norm": 5.15625,
      "learning_rate": 9.980560001189346e-06,
      "loss": 1.71096458,
      "memory(GiB)": 67.97,
      "step": 6045,
      "train_speed(iter/s)": 1.623415
    },
    {
      "acc": 0.62682972,
      "epoch": 0.1534753932014206,
      "grad_norm": 5.5,
      "learning_rate": 9.980467512237058e-06,
      "loss": 1.76352234,
      "memory(GiB)": 67.97,
      "step": 6050,
      "train_speed(iter/s)": 1.623576
    },
    {
      "acc": 0.63099384,
      "epoch": 0.15360223236935566,
      "grad_norm": 6.40625,
      "learning_rate": 9.98037480422157e-06,
      "loss": 1.69793587,
      "memory(GiB)": 77.59,
      "step": 6055,
      "train_speed(iter/s)": 1.623726
    },
    {
      "acc": 0.65371971,
      "epoch": 0.1537290715372907,
      "grad_norm": 6.34375,
      "learning_rate": 9.980281877146964e-06,
      "loss": 1.68482018,
      "memory(GiB)": 77.59,
      "step": 6060,
      "train_speed(iter/s)": 1.623876
    },
    {
      "acc": 0.64076004,
      "epoch": 0.15385591070522578,
      "grad_norm": 5.09375,
      "learning_rate": 9.980188731017327e-06,
      "loss": 1.7227747,
      "memory(GiB)": 77.59,
      "step": 6065,
      "train_speed(iter/s)": 1.624032
    },
    {
      "acc": 0.65239887,
      "epoch": 0.15398274987316082,
      "grad_norm": 5.96875,
      "learning_rate": 9.980095365836753e-06,
      "loss": 1.65667305,
      "memory(GiB)": 77.59,
      "step": 6070,
      "train_speed(iter/s)": 1.624173
    },
    {
      "acc": 0.64093332,
      "epoch": 0.1541095890410959,
      "grad_norm": 6.28125,
      "learning_rate": 9.980001781609353e-06,
      "loss": 1.7067337,
      "memory(GiB)": 77.59,
      "step": 6075,
      "train_speed(iter/s)": 1.62433
    },
    {
      "acc": 0.63491664,
      "epoch": 0.15423642820903094,
      "grad_norm": 5.9375,
      "learning_rate": 9.979907978339236e-06,
      "loss": 1.71492023,
      "memory(GiB)": 77.59,
      "step": 6080,
      "train_speed(iter/s)": 1.624484
    },
    {
      "acc": 0.6486021,
      "epoch": 0.154363267376966,
      "grad_norm": 6.375,
      "learning_rate": 9.979813956030535e-06,
      "loss": 1.65320702,
      "memory(GiB)": 77.59,
      "step": 6085,
      "train_speed(iter/s)": 1.624629
    },
    {
      "acc": 0.6361412,
      "epoch": 0.15449010654490106,
      "grad_norm": 6.34375,
      "learning_rate": 9.979719714687384e-06,
      "loss": 1.68614998,
      "memory(GiB)": 77.59,
      "step": 6090,
      "train_speed(iter/s)": 1.624776
    },
    {
      "acc": 0.63105526,
      "epoch": 0.15461694571283613,
      "grad_norm": 5.1875,
      "learning_rate": 9.979625254313924e-06,
      "loss": 1.78558273,
      "memory(GiB)": 77.59,
      "step": 6095,
      "train_speed(iter/s)": 1.624916
    },
    {
      "acc": 0.63914604,
      "epoch": 0.15474378488077117,
      "grad_norm": 7.21875,
      "learning_rate": 9.979530574914316e-06,
      "loss": 1.69465752,
      "memory(GiB)": 77.59,
      "step": 6100,
      "train_speed(iter/s)": 1.625067
    },
    {
      "acc": 0.64542117,
      "epoch": 0.15487062404870625,
      "grad_norm": 6.90625,
      "learning_rate": 9.97943567649272e-06,
      "loss": 1.65333576,
      "memory(GiB)": 77.59,
      "step": 6105,
      "train_speed(iter/s)": 1.625207
    },
    {
      "acc": 0.63427954,
      "epoch": 0.1549974632166413,
      "grad_norm": 5.9375,
      "learning_rate": 9.979340559053311e-06,
      "loss": 1.65659218,
      "memory(GiB)": 77.59,
      "step": 6110,
      "train_speed(iter/s)": 1.625348
    },
    {
      "acc": 0.63331575,
      "epoch": 0.15512430238457636,
      "grad_norm": 4.59375,
      "learning_rate": 9.979245222600273e-06,
      "loss": 1.70867023,
      "memory(GiB)": 77.59,
      "step": 6115,
      "train_speed(iter/s)": 1.625493
    },
    {
      "acc": 0.6256835,
      "epoch": 0.1552511415525114,
      "grad_norm": 6.09375,
      "learning_rate": 9.979149667137801e-06,
      "loss": 1.76142693,
      "memory(GiB)": 77.59,
      "step": 6120,
      "train_speed(iter/s)": 1.625644
    },
    {
      "acc": 0.62907953,
      "epoch": 0.15537798072044648,
      "grad_norm": 5.8125,
      "learning_rate": 9.979053892670094e-06,
      "loss": 1.73337746,
      "memory(GiB)": 77.59,
      "step": 6125,
      "train_speed(iter/s)": 1.625785
    },
    {
      "acc": 0.63550453,
      "epoch": 0.15550481988838152,
      "grad_norm": 5.1875,
      "learning_rate": 9.978957899201369e-06,
      "loss": 1.7557785,
      "memory(GiB)": 77.59,
      "step": 6130,
      "train_speed(iter/s)": 1.625932
    },
    {
      "acc": 0.64810534,
      "epoch": 0.1556316590563166,
      "grad_norm": 4.96875,
      "learning_rate": 9.978861686735845e-06,
      "loss": 1.69074402,
      "memory(GiB)": 77.59,
      "step": 6135,
      "train_speed(iter/s)": 1.626081
    },
    {
      "acc": 0.64021478,
      "epoch": 0.15575849822425164,
      "grad_norm": 7.65625,
      "learning_rate": 9.978765255277756e-06,
      "loss": 1.7314827,
      "memory(GiB)": 77.59,
      "step": 6140,
      "train_speed(iter/s)": 1.626227
    },
    {
      "acc": 0.64839134,
      "epoch": 0.1558853373921867,
      "grad_norm": 5.09375,
      "learning_rate": 9.97866860483134e-06,
      "loss": 1.70773048,
      "memory(GiB)": 77.59,
      "step": 6145,
      "train_speed(iter/s)": 1.626378
    },
    {
      "acc": 0.65473585,
      "epoch": 0.15601217656012176,
      "grad_norm": 5.5625,
      "learning_rate": 9.978571735400853e-06,
      "loss": 1.65567513,
      "memory(GiB)": 77.59,
      "step": 6150,
      "train_speed(iter/s)": 1.626529
    },
    {
      "acc": 0.63746767,
      "epoch": 0.15613901572805683,
      "grad_norm": 6.625,
      "learning_rate": 9.978474646990552e-06,
      "loss": 1.66863117,
      "memory(GiB)": 77.59,
      "step": 6155,
      "train_speed(iter/s)": 1.626681
    },
    {
      "acc": 0.64006233,
      "epoch": 0.15626585489599187,
      "grad_norm": 6.28125,
      "learning_rate": 9.97837733960471e-06,
      "loss": 1.73263168,
      "memory(GiB)": 77.59,
      "step": 6160,
      "train_speed(iter/s)": 1.626811
    },
    {
      "acc": 0.65888743,
      "epoch": 0.15639269406392695,
      "grad_norm": 6.53125,
      "learning_rate": 9.978279813247605e-06,
      "loss": 1.64967499,
      "memory(GiB)": 77.59,
      "step": 6165,
      "train_speed(iter/s)": 1.626952
    },
    {
      "acc": 0.64943757,
      "epoch": 0.156519533231862,
      "grad_norm": 5.03125,
      "learning_rate": 9.978182067923528e-06,
      "loss": 1.65488949,
      "memory(GiB)": 77.59,
      "step": 6170,
      "train_speed(iter/s)": 1.627093
    },
    {
      "acc": 0.63030982,
      "epoch": 0.15664637239979706,
      "grad_norm": 4.71875,
      "learning_rate": 9.978084103636778e-06,
      "loss": 1.79546776,
      "memory(GiB)": 77.59,
      "step": 6175,
      "train_speed(iter/s)": 1.627229
    },
    {
      "acc": 0.65145741,
      "epoch": 0.1567732115677321,
      "grad_norm": 6.21875,
      "learning_rate": 9.977985920391661e-06,
      "loss": 1.67122612,
      "memory(GiB)": 77.59,
      "step": 6180,
      "train_speed(iter/s)": 1.627374
    },
    {
      "acc": 0.64416523,
      "epoch": 0.15690005073566718,
      "grad_norm": 5.71875,
      "learning_rate": 9.977887518192501e-06,
      "loss": 1.74429035,
      "memory(GiB)": 77.59,
      "step": 6185,
      "train_speed(iter/s)": 1.627524
    },
    {
      "acc": 0.61545467,
      "epoch": 0.15702688990360222,
      "grad_norm": 4.65625,
      "learning_rate": 9.977788897043622e-06,
      "loss": 1.7866478,
      "memory(GiB)": 77.59,
      "step": 6190,
      "train_speed(iter/s)": 1.627676
    },
    {
      "acc": 0.64552336,
      "epoch": 0.1571537290715373,
      "grad_norm": 5.90625,
      "learning_rate": 9.977690056949363e-06,
      "loss": 1.68018341,
      "memory(GiB)": 77.59,
      "step": 6195,
      "train_speed(iter/s)": 1.62782
    },
    {
      "acc": 0.6357573,
      "epoch": 0.15728056823947234,
      "grad_norm": 7.15625,
      "learning_rate": 9.977590997914072e-06,
      "loss": 1.72546806,
      "memory(GiB)": 77.59,
      "step": 6200,
      "train_speed(iter/s)": 1.627954
    },
    {
      "acc": 0.63645563,
      "epoch": 0.1574074074074074,
      "grad_norm": 5.3125,
      "learning_rate": 9.977491719942106e-06,
      "loss": 1.67075195,
      "memory(GiB)": 77.59,
      "step": 6205,
      "train_speed(iter/s)": 1.628094
    },
    {
      "acc": 0.63589287,
      "epoch": 0.15753424657534246,
      "grad_norm": 6.125,
      "learning_rate": 9.97739222303783e-06,
      "loss": 1.71371651,
      "memory(GiB)": 77.59,
      "step": 6210,
      "train_speed(iter/s)": 1.628236
    },
    {
      "acc": 0.6513371,
      "epoch": 0.15766108574327753,
      "grad_norm": 6.8125,
      "learning_rate": 9.977292507205623e-06,
      "loss": 1.67914371,
      "memory(GiB)": 77.59,
      "step": 6215,
      "train_speed(iter/s)": 1.628381
    },
    {
      "acc": 0.61512079,
      "epoch": 0.15778792491121257,
      "grad_norm": 6.34375,
      "learning_rate": 9.977192572449868e-06,
      "loss": 1.77316208,
      "memory(GiB)": 77.59,
      "step": 6220,
      "train_speed(iter/s)": 1.628521
    },
    {
      "acc": 0.64252591,
      "epoch": 0.15791476407914765,
      "grad_norm": 5.9375,
      "learning_rate": 9.977092418774962e-06,
      "loss": 1.69516239,
      "memory(GiB)": 77.59,
      "step": 6225,
      "train_speed(iter/s)": 1.628678
    },
    {
      "acc": 0.62649403,
      "epoch": 0.1580416032470827,
      "grad_norm": 5.90625,
      "learning_rate": 9.976992046185313e-06,
      "loss": 1.76648808,
      "memory(GiB)": 77.59,
      "step": 6230,
      "train_speed(iter/s)": 1.628821
    },
    {
      "acc": 0.634657,
      "epoch": 0.15816844241501776,
      "grad_norm": 5.09375,
      "learning_rate": 9.97689145468533e-06,
      "loss": 1.72153091,
      "memory(GiB)": 77.59,
      "step": 6235,
      "train_speed(iter/s)": 1.628963
    },
    {
      "acc": 0.65538073,
      "epoch": 0.1582952815829528,
      "grad_norm": 5.71875,
      "learning_rate": 9.976790644279442e-06,
      "loss": 1.62780285,
      "memory(GiB)": 77.59,
      "step": 6240,
      "train_speed(iter/s)": 1.629107
    },
    {
      "acc": 0.64440994,
      "epoch": 0.15842212075088788,
      "grad_norm": 5.65625,
      "learning_rate": 9.976689614972082e-06,
      "loss": 1.63728638,
      "memory(GiB)": 77.59,
      "step": 6245,
      "train_speed(iter/s)": 1.62925
    },
    {
      "acc": 0.64431801,
      "epoch": 0.15854895991882292,
      "grad_norm": 7.5,
      "learning_rate": 9.976588366767693e-06,
      "loss": 1.71953621,
      "memory(GiB)": 77.59,
      "step": 6250,
      "train_speed(iter/s)": 1.629398
    },
    {
      "acc": 0.63995633,
      "epoch": 0.158675799086758,
      "grad_norm": 6.21875,
      "learning_rate": 9.976486899670729e-06,
      "loss": 1.6575222,
      "memory(GiB)": 77.59,
      "step": 6255,
      "train_speed(iter/s)": 1.629546
    },
    {
      "acc": 0.64340801,
      "epoch": 0.15880263825469304,
      "grad_norm": 5.75,
      "learning_rate": 9.976385213685652e-06,
      "loss": 1.7093338,
      "memory(GiB)": 77.59,
      "step": 6260,
      "train_speed(iter/s)": 1.629692
    },
    {
      "acc": 0.63619251,
      "epoch": 0.1589294774226281,
      "grad_norm": 5.84375,
      "learning_rate": 9.976283308816937e-06,
      "loss": 1.70851097,
      "memory(GiB)": 77.59,
      "step": 6265,
      "train_speed(iter/s)": 1.629836
    },
    {
      "acc": 0.65177736,
      "epoch": 0.15905631659056316,
      "grad_norm": 6.6875,
      "learning_rate": 9.976181185069063e-06,
      "loss": 1.62205124,
      "memory(GiB)": 77.59,
      "step": 6270,
      "train_speed(iter/s)": 1.629983
    },
    {
      "acc": 0.65391154,
      "epoch": 0.15918315575849823,
      "grad_norm": 4.28125,
      "learning_rate": 9.976078842446522e-06,
      "loss": 1.64793148,
      "memory(GiB)": 77.59,
      "step": 6275,
      "train_speed(iter/s)": 1.63012
    },
    {
      "acc": 0.6346415,
      "epoch": 0.15930999492643327,
      "grad_norm": 5.15625,
      "learning_rate": 9.97597628095382e-06,
      "loss": 1.8150074,
      "memory(GiB)": 77.59,
      "step": 6280,
      "train_speed(iter/s)": 1.630261
    },
    {
      "acc": 0.62557349,
      "epoch": 0.15943683409436835,
      "grad_norm": 6.21875,
      "learning_rate": 9.975873500595464e-06,
      "loss": 1.81581287,
      "memory(GiB)": 77.59,
      "step": 6285,
      "train_speed(iter/s)": 1.630399
    },
    {
      "acc": 0.64790401,
      "epoch": 0.1595636732623034,
      "grad_norm": 6.0,
      "learning_rate": 9.975770501375974e-06,
      "loss": 1.72666969,
      "memory(GiB)": 77.59,
      "step": 6290,
      "train_speed(iter/s)": 1.630542
    },
    {
      "acc": 0.64370136,
      "epoch": 0.15969051243023846,
      "grad_norm": 5.8125,
      "learning_rate": 9.975667283299884e-06,
      "loss": 1.68165207,
      "memory(GiB)": 77.59,
      "step": 6295,
      "train_speed(iter/s)": 1.630673
    },
    {
      "acc": 0.6499074,
      "epoch": 0.1598173515981735,
      "grad_norm": 6.65625,
      "learning_rate": 9.975563846371732e-06,
      "loss": 1.68750763,
      "memory(GiB)": 77.59,
      "step": 6300,
      "train_speed(iter/s)": 1.630803
    },
    {
      "acc": 0.64219446,
      "epoch": 0.15994419076610858,
      "grad_norm": 5.21875,
      "learning_rate": 9.975460190596068e-06,
      "loss": 1.69849129,
      "memory(GiB)": 77.59,
      "step": 6305,
      "train_speed(iter/s)": 1.630933
    },
    {
      "acc": 0.63774624,
      "epoch": 0.16007102993404362,
      "grad_norm": 7.09375,
      "learning_rate": 9.975356315977451e-06,
      "loss": 1.70805016,
      "memory(GiB)": 77.59,
      "step": 6310,
      "train_speed(iter/s)": 1.631077
    },
    {
      "acc": 0.64253635,
      "epoch": 0.1601978691019787,
      "grad_norm": 5.78125,
      "learning_rate": 9.975252222520449e-06,
      "loss": 1.66821289,
      "memory(GiB)": 77.59,
      "step": 6315,
      "train_speed(iter/s)": 1.631216
    },
    {
      "acc": 0.63843169,
      "epoch": 0.16032470826991374,
      "grad_norm": 4.9375,
      "learning_rate": 9.97514791022964e-06,
      "loss": 1.70587273,
      "memory(GiB)": 77.59,
      "step": 6320,
      "train_speed(iter/s)": 1.63135
    },
    {
      "acc": 0.64853153,
      "epoch": 0.1604515474378488,
      "grad_norm": 5.65625,
      "learning_rate": 9.975043379109617e-06,
      "loss": 1.68425255,
      "memory(GiB)": 77.59,
      "step": 6325,
      "train_speed(iter/s)": 1.631478
    },
    {
      "acc": 0.6516242,
      "epoch": 0.16057838660578386,
      "grad_norm": 4.96875,
      "learning_rate": 9.974938629164973e-06,
      "loss": 1.66455574,
      "memory(GiB)": 77.59,
      "step": 6330,
      "train_speed(iter/s)": 1.631611
    },
    {
      "acc": 0.64136925,
      "epoch": 0.16070522577371893,
      "grad_norm": 4.65625,
      "learning_rate": 9.974833660400315e-06,
      "loss": 1.72513733,
      "memory(GiB)": 77.59,
      "step": 6335,
      "train_speed(iter/s)": 1.631749
    },
    {
      "acc": 0.63354859,
      "epoch": 0.16083206494165397,
      "grad_norm": 6.4375,
      "learning_rate": 9.974728472820264e-06,
      "loss": 1.6803978,
      "memory(GiB)": 77.59,
      "step": 6340,
      "train_speed(iter/s)": 1.631896
    },
    {
      "acc": 0.63999481,
      "epoch": 0.16095890410958905,
      "grad_norm": 4.5625,
      "learning_rate": 9.97462306642944e-06,
      "loss": 1.70625572,
      "memory(GiB)": 77.59,
      "step": 6345,
      "train_speed(iter/s)": 1.632022
    },
    {
      "acc": 0.66038437,
      "epoch": 0.1610857432775241,
      "grad_norm": 5.09375,
      "learning_rate": 9.974517441232487e-06,
      "loss": 1.61866665,
      "memory(GiB)": 77.59,
      "step": 6350,
      "train_speed(iter/s)": 1.632153
    },
    {
      "acc": 0.63697634,
      "epoch": 0.16121258244545916,
      "grad_norm": 5.90625,
      "learning_rate": 9.974411597234046e-06,
      "loss": 1.70938187,
      "memory(GiB)": 77.59,
      "step": 6355,
      "train_speed(iter/s)": 1.632282
    },
    {
      "acc": 0.63928413,
      "epoch": 0.1613394216133942,
      "grad_norm": 6.71875,
      "learning_rate": 9.974305534438774e-06,
      "loss": 1.649403,
      "memory(GiB)": 77.59,
      "step": 6360,
      "train_speed(iter/s)": 1.632414
    },
    {
      "acc": 0.65040388,
      "epoch": 0.16146626078132928,
      "grad_norm": 5.4375,
      "learning_rate": 9.974199252851338e-06,
      "loss": 1.68915596,
      "memory(GiB)": 77.59,
      "step": 6365,
      "train_speed(iter/s)": 1.632562
    },
    {
      "acc": 0.65095468,
      "epoch": 0.16159309994926432,
      "grad_norm": 6.0625,
      "learning_rate": 9.974092752476408e-06,
      "loss": 1.65609608,
      "memory(GiB)": 77.59,
      "step": 6370,
      "train_speed(iter/s)": 1.632692
    },
    {
      "acc": 0.65658998,
      "epoch": 0.1617199391171994,
      "grad_norm": 5.46875,
      "learning_rate": 9.973986033318673e-06,
      "loss": 1.65106831,
      "memory(GiB)": 77.59,
      "step": 6375,
      "train_speed(iter/s)": 1.632822
    },
    {
      "acc": 0.66161289,
      "epoch": 0.16184677828513444,
      "grad_norm": 5.75,
      "learning_rate": 9.973879095382824e-06,
      "loss": 1.59875546,
      "memory(GiB)": 77.59,
      "step": 6380,
      "train_speed(iter/s)": 1.632944
    },
    {
      "acc": 0.65416441,
      "epoch": 0.1619736174530695,
      "grad_norm": 6.375,
      "learning_rate": 9.973771938673564e-06,
      "loss": 1.71432629,
      "memory(GiB)": 77.59,
      "step": 6385,
      "train_speed(iter/s)": 1.633085
    },
    {
      "acc": 0.64049058,
      "epoch": 0.16210045662100456,
      "grad_norm": 6.125,
      "learning_rate": 9.973664563195609e-06,
      "loss": 1.75938148,
      "memory(GiB)": 77.59,
      "step": 6390,
      "train_speed(iter/s)": 1.633218
    },
    {
      "acc": 0.64086409,
      "epoch": 0.16222729578893963,
      "grad_norm": 6.125,
      "learning_rate": 9.973556968953682e-06,
      "loss": 1.69508114,
      "memory(GiB)": 77.59,
      "step": 6395,
      "train_speed(iter/s)": 1.633346
    },
    {
      "acc": 0.63516617,
      "epoch": 0.16235413495687467,
      "grad_norm": 5.65625,
      "learning_rate": 9.973449155952512e-06,
      "loss": 1.6902504,
      "memory(GiB)": 77.59,
      "step": 6400,
      "train_speed(iter/s)": 1.633477
    },
    {
      "acc": 0.64499741,
      "epoch": 0.16248097412480975,
      "grad_norm": 5.53125,
      "learning_rate": 9.973341124196847e-06,
      "loss": 1.64758034,
      "memory(GiB)": 77.59,
      "step": 6405,
      "train_speed(iter/s)": 1.633611
    },
    {
      "acc": 0.65194874,
      "epoch": 0.1626078132927448,
      "grad_norm": 5.0625,
      "learning_rate": 9.973232873691431e-06,
      "loss": 1.62782841,
      "memory(GiB)": 77.59,
      "step": 6410,
      "train_speed(iter/s)": 1.633744
    },
    {
      "acc": 0.63939381,
      "epoch": 0.16273465246067986,
      "grad_norm": 4.90625,
      "learning_rate": 9.973124404441031e-06,
      "loss": 1.67741985,
      "memory(GiB)": 77.59,
      "step": 6415,
      "train_speed(iter/s)": 1.633866
    },
    {
      "acc": 0.63335905,
      "epoch": 0.1628614916286149,
      "grad_norm": 5.59375,
      "learning_rate": 9.973015716450416e-06,
      "loss": 1.79659672,
      "memory(GiB)": 77.59,
      "step": 6420,
      "train_speed(iter/s)": 1.633989
    },
    {
      "acc": 0.65132799,
      "epoch": 0.16298833079654998,
      "grad_norm": 5.84375,
      "learning_rate": 9.972906809724367e-06,
      "loss": 1.63987732,
      "memory(GiB)": 77.59,
      "step": 6425,
      "train_speed(iter/s)": 1.634111
    },
    {
      "acc": 0.65189471,
      "epoch": 0.16311516996448502,
      "grad_norm": 7.59375,
      "learning_rate": 9.972797684267674e-06,
      "loss": 1.60783386,
      "memory(GiB)": 77.59,
      "step": 6430,
      "train_speed(iter/s)": 1.634248
    },
    {
      "acc": 0.65726333,
      "epoch": 0.1632420091324201,
      "grad_norm": 5.25,
      "learning_rate": 9.972688340085137e-06,
      "loss": 1.61662788,
      "memory(GiB)": 77.59,
      "step": 6435,
      "train_speed(iter/s)": 1.634369
    },
    {
      "acc": 0.64713655,
      "epoch": 0.16336884830035514,
      "grad_norm": 7.15625,
      "learning_rate": 9.972578777181565e-06,
      "loss": 1.6686718,
      "memory(GiB)": 77.59,
      "step": 6440,
      "train_speed(iter/s)": 1.63449
    },
    {
      "acc": 0.6449069,
      "epoch": 0.1634956874682902,
      "grad_norm": 5.84375,
      "learning_rate": 9.972468995561778e-06,
      "loss": 1.65861053,
      "memory(GiB)": 77.59,
      "step": 6445,
      "train_speed(iter/s)": 1.634632
    },
    {
      "acc": 0.64491663,
      "epoch": 0.16362252663622526,
      "grad_norm": 5.5625,
      "learning_rate": 9.972358995230604e-06,
      "loss": 1.60342979,
      "memory(GiB)": 77.59,
      "step": 6450,
      "train_speed(iter/s)": 1.634765
    },
    {
      "acc": 0.65160785,
      "epoch": 0.16374936580416033,
      "grad_norm": 6.59375,
      "learning_rate": 9.97224877619288e-06,
      "loss": 1.68284607,
      "memory(GiB)": 77.59,
      "step": 6455,
      "train_speed(iter/s)": 1.634898
    },
    {
      "acc": 0.63881249,
      "epoch": 0.16387620497209537,
      "grad_norm": 5.5625,
      "learning_rate": 9.972138338453457e-06,
      "loss": 1.73002262,
      "memory(GiB)": 77.59,
      "step": 6460,
      "train_speed(iter/s)": 1.635021
    },
    {
      "acc": 0.6422462,
      "epoch": 0.16400304414003045,
      "grad_norm": 5.46875,
      "learning_rate": 9.972027682017191e-06,
      "loss": 1.68600445,
      "memory(GiB)": 77.59,
      "step": 6465,
      "train_speed(iter/s)": 1.635156
    },
    {
      "acc": 0.63023891,
      "epoch": 0.1641298833079655,
      "grad_norm": 4.6875,
      "learning_rate": 9.971916806888948e-06,
      "loss": 1.76903152,
      "memory(GiB)": 77.59,
      "step": 6470,
      "train_speed(iter/s)": 1.635292
    },
    {
      "acc": 0.63143888,
      "epoch": 0.16425672247590056,
      "grad_norm": 7.5,
      "learning_rate": 9.971805713073606e-06,
      "loss": 1.75191917,
      "memory(GiB)": 77.59,
      "step": 6475,
      "train_speed(iter/s)": 1.635419
    },
    {
      "acc": 0.63922138,
      "epoch": 0.1643835616438356,
      "grad_norm": 4.78125,
      "learning_rate": 9.971694400576053e-06,
      "loss": 1.64910049,
      "memory(GiB)": 77.59,
      "step": 6480,
      "train_speed(iter/s)": 1.635545
    },
    {
      "acc": 0.64952583,
      "epoch": 0.16451040081177068,
      "grad_norm": 4.78125,
      "learning_rate": 9.971582869401182e-06,
      "loss": 1.68407669,
      "memory(GiB)": 77.59,
      "step": 6485,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.64700904,
      "epoch": 0.16463723997970572,
      "grad_norm": 5.03125,
      "learning_rate": 9.9714711195539e-06,
      "loss": 1.67678223,
      "memory(GiB)": 77.59,
      "step": 6490,
      "train_speed(iter/s)": 1.635802
    },
    {
      "acc": 0.63749566,
      "epoch": 0.1647640791476408,
      "grad_norm": 7.375,
      "learning_rate": 9.97135915103912e-06,
      "loss": 1.68261909,
      "memory(GiB)": 77.59,
      "step": 6495,
      "train_speed(iter/s)": 1.635945
    },
    {
      "acc": 0.6402482,
      "epoch": 0.16489091831557584,
      "grad_norm": 5.6875,
      "learning_rate": 9.971246963861772e-06,
      "loss": 1.6840168,
      "memory(GiB)": 77.59,
      "step": 6500,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.63845644,
      "epoch": 0.1650177574835109,
      "grad_norm": 4.875,
      "learning_rate": 9.971134558026786e-06,
      "loss": 1.67110195,
      "memory(GiB)": 77.59,
      "step": 6505,
      "train_speed(iter/s)": 1.636196
    },
    {
      "acc": 0.64929929,
      "epoch": 0.16514459665144596,
      "grad_norm": 5.625,
      "learning_rate": 9.971021933539108e-06,
      "loss": 1.64482193,
      "memory(GiB)": 77.59,
      "step": 6510,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.65398116,
      "epoch": 0.16527143581938103,
      "grad_norm": 6.21875,
      "learning_rate": 9.97090909040369e-06,
      "loss": 1.64348888,
      "memory(GiB)": 77.59,
      "step": 6515,
      "train_speed(iter/s)": 1.636455
    },
    {
      "acc": 0.64077563,
      "epoch": 0.16539827498731607,
      "grad_norm": 5.09375,
      "learning_rate": 9.970796028625499e-06,
      "loss": 1.71602669,
      "memory(GiB)": 77.59,
      "step": 6520,
      "train_speed(iter/s)": 1.636582
    },
    {
      "acc": 0.64576349,
      "epoch": 0.16552511415525115,
      "grad_norm": 6.4375,
      "learning_rate": 9.970682748209505e-06,
      "loss": 1.71717949,
      "memory(GiB)": 77.59,
      "step": 6525,
      "train_speed(iter/s)": 1.636705
    },
    {
      "acc": 0.64918218,
      "epoch": 0.1656519533231862,
      "grad_norm": 4.65625,
      "learning_rate": 9.97056924916069e-06,
      "loss": 1.67604523,
      "memory(GiB)": 77.59,
      "step": 6530,
      "train_speed(iter/s)": 1.636822
    },
    {
      "acc": 0.66022949,
      "epoch": 0.16577879249112126,
      "grad_norm": 5.5,
      "learning_rate": 9.970455531484049e-06,
      "loss": 1.6372364,
      "memory(GiB)": 77.59,
      "step": 6535,
      "train_speed(iter/s)": 1.63695
    },
    {
      "acc": 0.64934483,
      "epoch": 0.1659056316590563,
      "grad_norm": 5.71875,
      "learning_rate": 9.97034159518458e-06,
      "loss": 1.66848183,
      "memory(GiB)": 77.59,
      "step": 6540,
      "train_speed(iter/s)": 1.637073
    },
    {
      "acc": 0.64757705,
      "epoch": 0.16603247082699138,
      "grad_norm": 6.875,
      "learning_rate": 9.9702274402673e-06,
      "loss": 1.68137646,
      "memory(GiB)": 77.59,
      "step": 6545,
      "train_speed(iter/s)": 1.637207
    },
    {
      "acc": 0.65756893,
      "epoch": 0.16615930999492642,
      "grad_norm": 5.40625,
      "learning_rate": 9.970113066737223e-06,
      "loss": 1.67345085,
      "memory(GiB)": 77.59,
      "step": 6550,
      "train_speed(iter/s)": 1.637336
    },
    {
      "acc": 0.64228902,
      "epoch": 0.1662861491628615,
      "grad_norm": 7.03125,
      "learning_rate": 9.969998474599386e-06,
      "loss": 1.63843899,
      "memory(GiB)": 77.59,
      "step": 6555,
      "train_speed(iter/s)": 1.637455
    },
    {
      "acc": 0.65614533,
      "epoch": 0.16641298833079654,
      "grad_norm": 6.90625,
      "learning_rate": 9.969883663858826e-06,
      "loss": 1.61360035,
      "memory(GiB)": 77.59,
      "step": 6560,
      "train_speed(iter/s)": 1.637579
    },
    {
      "acc": 0.65491066,
      "epoch": 0.1665398274987316,
      "grad_norm": 6.21875,
      "learning_rate": 9.969768634520593e-06,
      "loss": 1.66851044,
      "memory(GiB)": 77.59,
      "step": 6565,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.66400509,
      "epoch": 0.16666666666666666,
      "grad_norm": 4.78125,
      "learning_rate": 9.969653386589749e-06,
      "loss": 1.58823271,
      "memory(GiB)": 77.59,
      "step": 6570,
      "train_speed(iter/s)": 1.637838
    },
    {
      "acc": 0.65091476,
      "epoch": 0.16679350583460173,
      "grad_norm": 5.4375,
      "learning_rate": 9.96953792007136e-06,
      "loss": 1.68033352,
      "memory(GiB)": 77.59,
      "step": 6575,
      "train_speed(iter/s)": 1.637971
    },
    {
      "acc": 0.63005195,
      "epoch": 0.16692034500253677,
      "grad_norm": 4.59375,
      "learning_rate": 9.969422234970506e-06,
      "loss": 1.72654209,
      "memory(GiB)": 77.59,
      "step": 6580,
      "train_speed(iter/s)": 1.638099
    },
    {
      "acc": 0.6641572,
      "epoch": 0.16704718417047185,
      "grad_norm": 5.34375,
      "learning_rate": 9.969306331292273e-06,
      "loss": 1.61240158,
      "memory(GiB)": 77.59,
      "step": 6585,
      "train_speed(iter/s)": 1.638216
    },
    {
      "acc": 0.6374177,
      "epoch": 0.1671740233384069,
      "grad_norm": 5.78125,
      "learning_rate": 9.969190209041764e-06,
      "loss": 1.75764637,
      "memory(GiB)": 77.59,
      "step": 6590,
      "train_speed(iter/s)": 1.638344
    },
    {
      "acc": 0.63630428,
      "epoch": 0.16730086250634196,
      "grad_norm": 6.75,
      "learning_rate": 9.969073868224082e-06,
      "loss": 1.70809555,
      "memory(GiB)": 77.59,
      "step": 6595,
      "train_speed(iter/s)": 1.638476
    },
    {
      "acc": 0.63657174,
      "epoch": 0.167427701674277,
      "grad_norm": 5.125,
      "learning_rate": 9.968957308844346e-06,
      "loss": 1.77841892,
      "memory(GiB)": 77.59,
      "step": 6600,
      "train_speed(iter/s)": 1.638607
    },
    {
      "acc": 0.64168973,
      "epoch": 0.16755454084221208,
      "grad_norm": 6.625,
      "learning_rate": 9.968840530907684e-06,
      "loss": 1.71487465,
      "memory(GiB)": 77.59,
      "step": 6605,
      "train_speed(iter/s)": 1.638724
    },
    {
      "acc": 0.62594371,
      "epoch": 0.16768138001014712,
      "grad_norm": 7.625,
      "learning_rate": 9.96872353441923e-06,
      "loss": 1.73303661,
      "memory(GiB)": 77.59,
      "step": 6610,
      "train_speed(iter/s)": 1.638842
    },
    {
      "acc": 0.6411756,
      "epoch": 0.1678082191780822,
      "grad_norm": 6.75,
      "learning_rate": 9.968606319384131e-06,
      "loss": 1.73566246,
      "memory(GiB)": 77.59,
      "step": 6615,
      "train_speed(iter/s)": 1.63896
    },
    {
      "acc": 0.64131818,
      "epoch": 0.16793505834601724,
      "grad_norm": 7.3125,
      "learning_rate": 9.968488885807544e-06,
      "loss": 1.77030144,
      "memory(GiB)": 77.59,
      "step": 6620,
      "train_speed(iter/s)": 1.639089
    },
    {
      "acc": 0.64899211,
      "epoch": 0.1680618975139523,
      "grad_norm": 4.96875,
      "learning_rate": 9.968371233694633e-06,
      "loss": 1.67177773,
      "memory(GiB)": 77.59,
      "step": 6625,
      "train_speed(iter/s)": 1.639219
    },
    {
      "acc": 0.64630604,
      "epoch": 0.16818873668188736,
      "grad_norm": 5.75,
      "learning_rate": 9.968253363050573e-06,
      "loss": 1.69663029,
      "memory(GiB)": 77.59,
      "step": 6630,
      "train_speed(iter/s)": 1.63935
    },
    {
      "acc": 0.64754329,
      "epoch": 0.16831557584982243,
      "grad_norm": 4.625,
      "learning_rate": 9.968135273880547e-06,
      "loss": 1.64831924,
      "memory(GiB)": 77.59,
      "step": 6635,
      "train_speed(iter/s)": 1.639468
    },
    {
      "acc": 0.64454737,
      "epoch": 0.16844241501775747,
      "grad_norm": 6.75,
      "learning_rate": 9.968016966189753e-06,
      "loss": 1.67571716,
      "memory(GiB)": 77.59,
      "step": 6640,
      "train_speed(iter/s)": 1.639597
    },
    {
      "acc": 0.64654708,
      "epoch": 0.16856925418569255,
      "grad_norm": 6.3125,
      "learning_rate": 9.96789843998339e-06,
      "loss": 1.59341631,
      "memory(GiB)": 77.59,
      "step": 6645,
      "train_speed(iter/s)": 1.639715
    },
    {
      "acc": 0.64665756,
      "epoch": 0.1686960933536276,
      "grad_norm": 7.21875,
      "learning_rate": 9.967779695266675e-06,
      "loss": 1.66016655,
      "memory(GiB)": 77.59,
      "step": 6650,
      "train_speed(iter/s)": 1.639837
    },
    {
      "acc": 0.64671593,
      "epoch": 0.16882293252156266,
      "grad_norm": 5.96875,
      "learning_rate": 9.967660732044828e-06,
      "loss": 1.68118477,
      "memory(GiB)": 77.59,
      "step": 6655,
      "train_speed(iter/s)": 1.639958
    },
    {
      "acc": 0.645193,
      "epoch": 0.1689497716894977,
      "grad_norm": 5.0,
      "learning_rate": 9.967541550323085e-06,
      "loss": 1.69771557,
      "memory(GiB)": 77.59,
      "step": 6660,
      "train_speed(iter/s)": 1.640085
    },
    {
      "acc": 0.65040121,
      "epoch": 0.16907661085743278,
      "grad_norm": 6.21875,
      "learning_rate": 9.967422150106685e-06,
      "loss": 1.69082069,
      "memory(GiB)": 77.59,
      "step": 6665,
      "train_speed(iter/s)": 1.640218
    },
    {
      "acc": 0.64032044,
      "epoch": 0.16920345002536782,
      "grad_norm": 5.0625,
      "learning_rate": 9.96730253140088e-06,
      "loss": 1.67483864,
      "memory(GiB)": 77.59,
      "step": 6670,
      "train_speed(iter/s)": 1.640341
    },
    {
      "acc": 0.64160852,
      "epoch": 0.1693302891933029,
      "grad_norm": 5.96875,
      "learning_rate": 9.967182694210933e-06,
      "loss": 1.71246223,
      "memory(GiB)": 77.59,
      "step": 6675,
      "train_speed(iter/s)": 1.640468
    },
    {
      "acc": 0.6497797,
      "epoch": 0.16945712836123794,
      "grad_norm": 4.71875,
      "learning_rate": 9.967062638542116e-06,
      "loss": 1.67274075,
      "memory(GiB)": 77.59,
      "step": 6680,
      "train_speed(iter/s)": 1.640595
    },
    {
      "acc": 0.64639378,
      "epoch": 0.169583967529173,
      "grad_norm": 5.625,
      "learning_rate": 9.966942364399706e-06,
      "loss": 1.69283409,
      "memory(GiB)": 77.59,
      "step": 6685,
      "train_speed(iter/s)": 1.640716
    },
    {
      "acc": 0.63445115,
      "epoch": 0.16971080669710806,
      "grad_norm": 5.75,
      "learning_rate": 9.966821871788995e-06,
      "loss": 1.72606163,
      "memory(GiB)": 77.59,
      "step": 6690,
      "train_speed(iter/s)": 1.640843
    },
    {
      "acc": 0.64796643,
      "epoch": 0.16983764586504313,
      "grad_norm": 6.0625,
      "learning_rate": 9.966701160715283e-06,
      "loss": 1.67875748,
      "memory(GiB)": 77.59,
      "step": 6695,
      "train_speed(iter/s)": 1.640951
    },
    {
      "acc": 0.65945635,
      "epoch": 0.16996448503297817,
      "grad_norm": 6.28125,
      "learning_rate": 9.96658023118388e-06,
      "loss": 1.64298096,
      "memory(GiB)": 77.59,
      "step": 6700,
      "train_speed(iter/s)": 1.641083
    },
    {
      "acc": 0.64791965,
      "epoch": 0.17009132420091325,
      "grad_norm": 6.09375,
      "learning_rate": 9.966459083200102e-06,
      "loss": 1.70205956,
      "memory(GiB)": 77.59,
      "step": 6705,
      "train_speed(iter/s)": 1.641207
    },
    {
      "acc": 0.63008356,
      "epoch": 0.1702181633688483,
      "grad_norm": 5.03125,
      "learning_rate": 9.966337716769283e-06,
      "loss": 1.74058895,
      "memory(GiB)": 77.59,
      "step": 6710,
      "train_speed(iter/s)": 1.641333
    },
    {
      "acc": 0.63434973,
      "epoch": 0.17034500253678336,
      "grad_norm": 6.0,
      "learning_rate": 9.966216131896755e-06,
      "loss": 1.73385658,
      "memory(GiB)": 77.59,
      "step": 6715,
      "train_speed(iter/s)": 1.641439
    },
    {
      "acc": 0.63197145,
      "epoch": 0.1704718417047184,
      "grad_norm": 6.5625,
      "learning_rate": 9.966094328587871e-06,
      "loss": 1.70180206,
      "memory(GiB)": 77.59,
      "step": 6720,
      "train_speed(iter/s)": 1.641552
    },
    {
      "acc": 0.63856096,
      "epoch": 0.17059868087265348,
      "grad_norm": 5.375,
      "learning_rate": 9.965972306847986e-06,
      "loss": 1.74872761,
      "memory(GiB)": 77.59,
      "step": 6725,
      "train_speed(iter/s)": 1.641663
    },
    {
      "acc": 0.66174555,
      "epoch": 0.17072552004058852,
      "grad_norm": 5.71875,
      "learning_rate": 9.965850066682468e-06,
      "loss": 1.57597809,
      "memory(GiB)": 77.59,
      "step": 6730,
      "train_speed(iter/s)": 1.641786
    },
    {
      "acc": 0.63363771,
      "epoch": 0.1708523592085236,
      "grad_norm": 5.28125,
      "learning_rate": 9.965727608096692e-06,
      "loss": 1.67796688,
      "memory(GiB)": 77.59,
      "step": 6735,
      "train_speed(iter/s)": 1.641917
    },
    {
      "acc": 0.65201859,
      "epoch": 0.17097919837645864,
      "grad_norm": 4.5625,
      "learning_rate": 9.965604931096045e-06,
      "loss": 1.72040901,
      "memory(GiB)": 77.59,
      "step": 6740,
      "train_speed(iter/s)": 1.642042
    },
    {
      "acc": 0.63816962,
      "epoch": 0.1711060375443937,
      "grad_norm": 6.6875,
      "learning_rate": 9.965482035685925e-06,
      "loss": 1.7437851,
      "memory(GiB)": 77.59,
      "step": 6745,
      "train_speed(iter/s)": 1.642158
    },
    {
      "acc": 0.65284352,
      "epoch": 0.17123287671232876,
      "grad_norm": 5.5625,
      "learning_rate": 9.965358921871735e-06,
      "loss": 1.64097538,
      "memory(GiB)": 77.59,
      "step": 6750,
      "train_speed(iter/s)": 1.642289
    },
    {
      "acc": 0.65702648,
      "epoch": 0.17135971588026383,
      "grad_norm": 5.21875,
      "learning_rate": 9.965235589658891e-06,
      "loss": 1.61880646,
      "memory(GiB)": 77.59,
      "step": 6755,
      "train_speed(iter/s)": 1.64242
    },
    {
      "acc": 0.64878421,
      "epoch": 0.17148655504819887,
      "grad_norm": 4.15625,
      "learning_rate": 9.965112039052817e-06,
      "loss": 1.67541847,
      "memory(GiB)": 77.59,
      "step": 6760,
      "train_speed(iter/s)": 1.64254
    },
    {
      "acc": 0.63958874,
      "epoch": 0.17161339421613395,
      "grad_norm": 5.875,
      "learning_rate": 9.964988270058948e-06,
      "loss": 1.72893887,
      "memory(GiB)": 77.59,
      "step": 6765,
      "train_speed(iter/s)": 1.642656
    },
    {
      "acc": 0.63764372,
      "epoch": 0.171740233384069,
      "grad_norm": 5.34375,
      "learning_rate": 9.96486428268273e-06,
      "loss": 1.65488224,
      "memory(GiB)": 77.59,
      "step": 6770,
      "train_speed(iter/s)": 1.642774
    },
    {
      "acc": 0.64337626,
      "epoch": 0.17186707255200406,
      "grad_norm": 6.375,
      "learning_rate": 9.964740076929612e-06,
      "loss": 1.66642056,
      "memory(GiB)": 77.59,
      "step": 6775,
      "train_speed(iter/s)": 1.642889
    },
    {
      "acc": 0.63530145,
      "epoch": 0.1719939117199391,
      "grad_norm": 6.1875,
      "learning_rate": 9.964615652805059e-06,
      "loss": 1.72618942,
      "memory(GiB)": 77.59,
      "step": 6780,
      "train_speed(iter/s)": 1.642997
    },
    {
      "acc": 0.65174122,
      "epoch": 0.17212075088787418,
      "grad_norm": 6.0625,
      "learning_rate": 9.964491010314545e-06,
      "loss": 1.65879211,
      "memory(GiB)": 77.59,
      "step": 6785,
      "train_speed(iter/s)": 1.643111
    },
    {
      "acc": 0.64636569,
      "epoch": 0.17224759005580922,
      "grad_norm": 5.40625,
      "learning_rate": 9.964366149463552e-06,
      "loss": 1.69336262,
      "memory(GiB)": 77.59,
      "step": 6790,
      "train_speed(iter/s)": 1.643226
    },
    {
      "acc": 0.66044865,
      "epoch": 0.1723744292237443,
      "grad_norm": 5.03125,
      "learning_rate": 9.96424107025757e-06,
      "loss": 1.61055202,
      "memory(GiB)": 77.59,
      "step": 6795,
      "train_speed(iter/s)": 1.643337
    },
    {
      "acc": 0.6528614,
      "epoch": 0.17250126839167934,
      "grad_norm": 4.84375,
      "learning_rate": 9.964115772702104e-06,
      "loss": 1.61430187,
      "memory(GiB)": 77.59,
      "step": 6800,
      "train_speed(iter/s)": 1.643454
    },
    {
      "acc": 0.64503965,
      "epoch": 0.1726281075596144,
      "grad_norm": 5.4375,
      "learning_rate": 9.963990256802662e-06,
      "loss": 1.72916031,
      "memory(GiB)": 77.59,
      "step": 6805,
      "train_speed(iter/s)": 1.64357
    },
    {
      "acc": 0.6588366,
      "epoch": 0.17275494672754946,
      "grad_norm": 7.34375,
      "learning_rate": 9.963864522564765e-06,
      "loss": 1.61271057,
      "memory(GiB)": 77.59,
      "step": 6810,
      "train_speed(iter/s)": 1.643692
    },
    {
      "acc": 0.6429822,
      "epoch": 0.17288178589548453,
      "grad_norm": 5.09375,
      "learning_rate": 9.963738569993945e-06,
      "loss": 1.71824512,
      "memory(GiB)": 77.59,
      "step": 6815,
      "train_speed(iter/s)": 1.643811
    },
    {
      "acc": 0.65590649,
      "epoch": 0.17300862506341957,
      "grad_norm": 7.03125,
      "learning_rate": 9.963612399095743e-06,
      "loss": 1.67555981,
      "memory(GiB)": 77.59,
      "step": 6820,
      "train_speed(iter/s)": 1.643927
    },
    {
      "acc": 0.629282,
      "epoch": 0.17313546423135465,
      "grad_norm": 6.125,
      "learning_rate": 9.963486009875705e-06,
      "loss": 1.75797329,
      "memory(GiB)": 77.59,
      "step": 6825,
      "train_speed(iter/s)": 1.644038
    },
    {
      "acc": 0.64625211,
      "epoch": 0.1732623033992897,
      "grad_norm": 5.65625,
      "learning_rate": 9.963359402339393e-06,
      "loss": 1.68408127,
      "memory(GiB)": 77.59,
      "step": 6830,
      "train_speed(iter/s)": 1.644154
    },
    {
      "acc": 0.64075089,
      "epoch": 0.17338914256722476,
      "grad_norm": 7.375,
      "learning_rate": 9.963232576492373e-06,
      "loss": 1.75103569,
      "memory(GiB)": 77.59,
      "step": 6835,
      "train_speed(iter/s)": 1.644274
    },
    {
      "acc": 0.63705473,
      "epoch": 0.1735159817351598,
      "grad_norm": 5.25,
      "learning_rate": 9.963105532340226e-06,
      "loss": 1.75616817,
      "memory(GiB)": 77.59,
      "step": 6840,
      "train_speed(iter/s)": 1.644384
    },
    {
      "acc": 0.634588,
      "epoch": 0.17364282090309488,
      "grad_norm": 6.40625,
      "learning_rate": 9.962978269888538e-06,
      "loss": 1.7275013,
      "memory(GiB)": 77.59,
      "step": 6845,
      "train_speed(iter/s)": 1.644502
    },
    {
      "acc": 0.63084083,
      "epoch": 0.17376966007102992,
      "grad_norm": 4.71875,
      "learning_rate": 9.96285078914291e-06,
      "loss": 1.67146854,
      "memory(GiB)": 77.59,
      "step": 6850,
      "train_speed(iter/s)": 1.644614
    },
    {
      "acc": 0.63827963,
      "epoch": 0.173896499238965,
      "grad_norm": 5.4375,
      "learning_rate": 9.962723090108944e-06,
      "loss": 1.70191002,
      "memory(GiB)": 77.59,
      "step": 6855,
      "train_speed(iter/s)": 1.644737
    },
    {
      "acc": 0.64026279,
      "epoch": 0.17402333840690004,
      "grad_norm": 5.625,
      "learning_rate": 9.962595172792261e-06,
      "loss": 1.70468407,
      "memory(GiB)": 77.59,
      "step": 6860,
      "train_speed(iter/s)": 1.644858
    },
    {
      "acc": 0.64106016,
      "epoch": 0.1741501775748351,
      "grad_norm": 5.0,
      "learning_rate": 9.962467037198487e-06,
      "loss": 1.63270378,
      "memory(GiB)": 77.59,
      "step": 6865,
      "train_speed(iter/s)": 1.644964
    },
    {
      "acc": 0.64408913,
      "epoch": 0.17427701674277016,
      "grad_norm": 5.03125,
      "learning_rate": 9.962338683333254e-06,
      "loss": 1.7004734,
      "memory(GiB)": 77.59,
      "step": 6870,
      "train_speed(iter/s)": 1.645078
    },
    {
      "acc": 0.63785305,
      "epoch": 0.17440385591070523,
      "grad_norm": 5.0,
      "learning_rate": 9.962210111202212e-06,
      "loss": 1.77585373,
      "memory(GiB)": 77.59,
      "step": 6875,
      "train_speed(iter/s)": 1.645195
    },
    {
      "acc": 0.64036989,
      "epoch": 0.17453069507864027,
      "grad_norm": 6.90625,
      "learning_rate": 9.962081320811015e-06,
      "loss": 1.68967934,
      "memory(GiB)": 77.59,
      "step": 6880,
      "train_speed(iter/s)": 1.645308
    },
    {
      "acc": 0.65590191,
      "epoch": 0.17465753424657535,
      "grad_norm": 5.84375,
      "learning_rate": 9.961952312165327e-06,
      "loss": 1.67022934,
      "memory(GiB)": 77.59,
      "step": 6885,
      "train_speed(iter/s)": 1.645427
    },
    {
      "acc": 0.6510498,
      "epoch": 0.1747843734145104,
      "grad_norm": 4.875,
      "learning_rate": 9.961823085270823e-06,
      "loss": 1.65727654,
      "memory(GiB)": 77.59,
      "step": 6890,
      "train_speed(iter/s)": 1.645527
    },
    {
      "acc": 0.65662556,
      "epoch": 0.17491121258244546,
      "grad_norm": 5.09375,
      "learning_rate": 9.961693640133187e-06,
      "loss": 1.5743947,
      "memory(GiB)": 77.59,
      "step": 6895,
      "train_speed(iter/s)": 1.645636
    },
    {
      "acc": 0.64352493,
      "epoch": 0.1750380517503805,
      "grad_norm": 5.40625,
      "learning_rate": 9.961563976758112e-06,
      "loss": 1.67347393,
      "memory(GiB)": 77.59,
      "step": 6900,
      "train_speed(iter/s)": 1.645748
    },
    {
      "acc": 0.63922806,
      "epoch": 0.17516489091831558,
      "grad_norm": 6.0,
      "learning_rate": 9.961434095151301e-06,
      "loss": 1.71597672,
      "memory(GiB)": 77.59,
      "step": 6905,
      "train_speed(iter/s)": 1.645871
    },
    {
      "acc": 0.64139996,
      "epoch": 0.17529173008625062,
      "grad_norm": 4.375,
      "learning_rate": 9.961303995318467e-06,
      "loss": 1.63297272,
      "memory(GiB)": 77.59,
      "step": 6910,
      "train_speed(iter/s)": 1.645984
    },
    {
      "acc": 0.65589032,
      "epoch": 0.1754185692541857,
      "grad_norm": 18.0,
      "learning_rate": 9.961173677265334e-06,
      "loss": 1.6429985,
      "memory(GiB)": 77.59,
      "step": 6915,
      "train_speed(iter/s)": 1.646098
    },
    {
      "acc": 0.6408843,
      "epoch": 0.17554540842212074,
      "grad_norm": 4.78125,
      "learning_rate": 9.961043140997632e-06,
      "loss": 1.65616589,
      "memory(GiB)": 77.59,
      "step": 6920,
      "train_speed(iter/s)": 1.646209
    },
    {
      "acc": 0.63362069,
      "epoch": 0.1756722475900558,
      "grad_norm": 6.125,
      "learning_rate": 9.960912386521104e-06,
      "loss": 1.6962431,
      "memory(GiB)": 77.59,
      "step": 6925,
      "train_speed(iter/s)": 1.646316
    },
    {
      "acc": 0.64199467,
      "epoch": 0.17579908675799086,
      "grad_norm": 5.75,
      "learning_rate": 9.9607814138415e-06,
      "loss": 1.69304123,
      "memory(GiB)": 77.59,
      "step": 6930,
      "train_speed(iter/s)": 1.646432
    },
    {
      "acc": 0.64293036,
      "epoch": 0.17592592592592593,
      "grad_norm": 7.8125,
      "learning_rate": 9.96065022296458e-06,
      "loss": 1.70573483,
      "memory(GiB)": 77.59,
      "step": 6935,
      "train_speed(iter/s)": 1.646544
    },
    {
      "acc": 0.65458007,
      "epoch": 0.17605276509386097,
      "grad_norm": 6.1875,
      "learning_rate": 9.960518813896117e-06,
      "loss": 1.62631474,
      "memory(GiB)": 77.59,
      "step": 6940,
      "train_speed(iter/s)": 1.64369
    },
    {
      "acc": 0.64092197,
      "epoch": 0.17617960426179605,
      "grad_norm": 5.9375,
      "learning_rate": 9.960387186641887e-06,
      "loss": 1.6703289,
      "memory(GiB)": 77.59,
      "step": 6945,
      "train_speed(iter/s)": 1.643804
    },
    {
      "acc": 0.64519463,
      "epoch": 0.1763064434297311,
      "grad_norm": 6.125,
      "learning_rate": 9.960255341207686e-06,
      "loss": 1.63410873,
      "memory(GiB)": 77.59,
      "step": 6950,
      "train_speed(iter/s)": 1.641823
    },
    {
      "acc": 0.66517429,
      "epoch": 0.17643328259766616,
      "grad_norm": 5.1875,
      "learning_rate": 9.960123277599305e-06,
      "loss": 1.63524628,
      "memory(GiB)": 77.59,
      "step": 6955,
      "train_speed(iter/s)": 1.641933
    },
    {
      "acc": 0.63006377,
      "epoch": 0.1765601217656012,
      "grad_norm": 6.40625,
      "learning_rate": 9.959990995822559e-06,
      "loss": 1.70026054,
      "memory(GiB)": 77.59,
      "step": 6960,
      "train_speed(iter/s)": 1.642052
    },
    {
      "acc": 0.65416689,
      "epoch": 0.17668696093353628,
      "grad_norm": 4.59375,
      "learning_rate": 9.959858495883263e-06,
      "loss": 1.64462242,
      "memory(GiB)": 77.59,
      "step": 6965,
      "train_speed(iter/s)": 1.642174
    },
    {
      "acc": 0.64853497,
      "epoch": 0.17681380010147132,
      "grad_norm": 4.75,
      "learning_rate": 9.959725777787249e-06,
      "loss": 1.61428261,
      "memory(GiB)": 77.59,
      "step": 6970,
      "train_speed(iter/s)": 1.64229
    },
    {
      "acc": 0.64460812,
      "epoch": 0.1769406392694064,
      "grad_norm": 6.09375,
      "learning_rate": 9.959592841540349e-06,
      "loss": 1.68385105,
      "memory(GiB)": 77.59,
      "step": 6975,
      "train_speed(iter/s)": 1.642392
    },
    {
      "acc": 0.63454218,
      "epoch": 0.17706747843734144,
      "grad_norm": 5.5625,
      "learning_rate": 9.959459687148414e-06,
      "loss": 1.66434441,
      "memory(GiB)": 77.59,
      "step": 6980,
      "train_speed(iter/s)": 1.642514
    },
    {
      "acc": 0.62219305,
      "epoch": 0.1771943176052765,
      "grad_norm": 6.40625,
      "learning_rate": 9.959326314617299e-06,
      "loss": 1.72570801,
      "memory(GiB)": 77.59,
      "step": 6985,
      "train_speed(iter/s)": 1.642618
    },
    {
      "acc": 0.63553267,
      "epoch": 0.17732115677321156,
      "grad_norm": 6.0,
      "learning_rate": 9.95919272395287e-06,
      "loss": 1.71557579,
      "memory(GiB)": 77.59,
      "step": 6990,
      "train_speed(iter/s)": 1.64272
    },
    {
      "acc": 0.63350801,
      "epoch": 0.17744799594114663,
      "grad_norm": 6.1875,
      "learning_rate": 9.959058915161006e-06,
      "loss": 1.79244995,
      "memory(GiB)": 77.59,
      "step": 6995,
      "train_speed(iter/s)": 1.642833
    },
    {
      "acc": 0.62339063,
      "epoch": 0.17757483510908167,
      "grad_norm": 5.3125,
      "learning_rate": 9.95892488824759e-06,
      "loss": 1.75188446,
      "memory(GiB)": 77.59,
      "step": 7000,
      "train_speed(iter/s)": 1.64294
    },
    {
      "epoch": 0.17757483510908167,
      "eval_acc": 0.633713196215154,
      "eval_loss": 1.6434897184371948,
      "eval_runtime": 58.7221,
      "eval_samples_per_second": 108.477,
      "eval_steps_per_second": 27.128,
      "step": 7000
    },
    {
      "acc": 0.66383352,
      "epoch": 0.17770167427701675,
      "grad_norm": 5.96875,
      "learning_rate": 9.958790643218515e-06,
      "loss": 1.55810013,
      "memory(GiB)": 77.59,
      "step": 7005,
      "train_speed(iter/s)": 1.616317
    },
    {
      "acc": 0.64305129,
      "epoch": 0.1778285134449518,
      "grad_norm": 4.4375,
      "learning_rate": 9.95865618007969e-06,
      "loss": 1.63794308,
      "memory(GiB)": 77.59,
      "step": 7010,
      "train_speed(iter/s)": 1.616434
    },
    {
      "acc": 0.64964619,
      "epoch": 0.17795535261288686,
      "grad_norm": 5.1875,
      "learning_rate": 9.958521498837029e-06,
      "loss": 1.67577801,
      "memory(GiB)": 77.59,
      "step": 7015,
      "train_speed(iter/s)": 1.616566
    },
    {
      "acc": 0.64845467,
      "epoch": 0.1780821917808219,
      "grad_norm": 6.875,
      "learning_rate": 9.95838659949645e-06,
      "loss": 1.7144001,
      "memory(GiB)": 77.59,
      "step": 7020,
      "train_speed(iter/s)": 1.61669
    },
    {
      "acc": 0.66691551,
      "epoch": 0.17820903094875698,
      "grad_norm": 5.9375,
      "learning_rate": 9.958251482063894e-06,
      "loss": 1.64887276,
      "memory(GiB)": 77.59,
      "step": 7025,
      "train_speed(iter/s)": 1.616815
    },
    {
      "acc": 0.63838854,
      "epoch": 0.17833587011669202,
      "grad_norm": 5.75,
      "learning_rate": 9.9581161465453e-06,
      "loss": 1.69448376,
      "memory(GiB)": 77.59,
      "step": 7030,
      "train_speed(iter/s)": 1.616931
    },
    {
      "acc": 0.63211827,
      "epoch": 0.1784627092846271,
      "grad_norm": 6.09375,
      "learning_rate": 9.957980592946621e-06,
      "loss": 1.69647293,
      "memory(GiB)": 77.59,
      "step": 7035,
      "train_speed(iter/s)": 1.617065
    },
    {
      "acc": 0.66187387,
      "epoch": 0.17858954845256214,
      "grad_norm": 5.53125,
      "learning_rate": 9.957844821273822e-06,
      "loss": 1.61571198,
      "memory(GiB)": 77.59,
      "step": 7040,
      "train_speed(iter/s)": 1.617194
    },
    {
      "acc": 0.65497189,
      "epoch": 0.1787163876204972,
      "grad_norm": 5.1875,
      "learning_rate": 9.95770883153287e-06,
      "loss": 1.58073425,
      "memory(GiB)": 77.59,
      "step": 7045,
      "train_speed(iter/s)": 1.617318
    },
    {
      "acc": 0.64070177,
      "epoch": 0.17884322678843226,
      "grad_norm": 5.6875,
      "learning_rate": 9.957572623729749e-06,
      "loss": 1.7325016,
      "memory(GiB)": 77.59,
      "step": 7050,
      "train_speed(iter/s)": 1.617443
    },
    {
      "acc": 0.63557272,
      "epoch": 0.17897006595636733,
      "grad_norm": 6.71875,
      "learning_rate": 9.957436197870451e-06,
      "loss": 1.69866238,
      "memory(GiB)": 77.59,
      "step": 7055,
      "train_speed(iter/s)": 1.617574
    },
    {
      "acc": 0.64438353,
      "epoch": 0.17909690512430237,
      "grad_norm": 7.09375,
      "learning_rate": 9.957299553960975e-06,
      "loss": 1.73529491,
      "memory(GiB)": 77.59,
      "step": 7060,
      "train_speed(iter/s)": 1.617703
    },
    {
      "acc": 0.64680119,
      "epoch": 0.17922374429223745,
      "grad_norm": 6.125,
      "learning_rate": 9.957162692007334e-06,
      "loss": 1.62102757,
      "memory(GiB)": 77.59,
      "step": 7065,
      "train_speed(iter/s)": 1.617841
    },
    {
      "acc": 0.65200863,
      "epoch": 0.1793505834601725,
      "grad_norm": 6.25,
      "learning_rate": 9.957025612015543e-06,
      "loss": 1.68627396,
      "memory(GiB)": 77.59,
      "step": 7070,
      "train_speed(iter/s)": 1.61797
    },
    {
      "acc": 0.65332608,
      "epoch": 0.17947742262810756,
      "grad_norm": 5.0,
      "learning_rate": 9.956888313991636e-06,
      "loss": 1.61336212,
      "memory(GiB)": 77.59,
      "step": 7075,
      "train_speed(iter/s)": 1.618104
    },
    {
      "acc": 0.64589319,
      "epoch": 0.1796042617960426,
      "grad_norm": 5.96875,
      "learning_rate": 9.956750797941648e-06,
      "loss": 1.69664879,
      "memory(GiB)": 77.59,
      "step": 7080,
      "train_speed(iter/s)": 1.618232
    },
    {
      "acc": 0.63732195,
      "epoch": 0.17973110096397768,
      "grad_norm": 5.0625,
      "learning_rate": 9.95661306387163e-06,
      "loss": 1.73729515,
      "memory(GiB)": 77.59,
      "step": 7085,
      "train_speed(iter/s)": 1.618362
    },
    {
      "acc": 0.64629126,
      "epoch": 0.17985794013191272,
      "grad_norm": 5.03125,
      "learning_rate": 9.95647511178764e-06,
      "loss": 1.5979641,
      "memory(GiB)": 77.59,
      "step": 7090,
      "train_speed(iter/s)": 1.618487
    },
    {
      "acc": 0.64954901,
      "epoch": 0.1799847792998478,
      "grad_norm": 6.34375,
      "learning_rate": 9.956336941695747e-06,
      "loss": 1.70026016,
      "memory(GiB)": 77.59,
      "step": 7095,
      "train_speed(iter/s)": 1.618612
    },
    {
      "acc": 0.64863548,
      "epoch": 0.18011161846778284,
      "grad_norm": 6.3125,
      "learning_rate": 9.956198553602026e-06,
      "loss": 1.64024582,
      "memory(GiB)": 77.59,
      "step": 7100,
      "train_speed(iter/s)": 1.618741
    },
    {
      "acc": 0.61645384,
      "epoch": 0.1802384576357179,
      "grad_norm": 5.1875,
      "learning_rate": 9.956059947512563e-06,
      "loss": 1.78149529,
      "memory(GiB)": 77.59,
      "step": 7105,
      "train_speed(iter/s)": 1.618868
    },
    {
      "acc": 0.63665099,
      "epoch": 0.18036529680365296,
      "grad_norm": 5.84375,
      "learning_rate": 9.95592112343346e-06,
      "loss": 1.67411232,
      "memory(GiB)": 77.59,
      "step": 7110,
      "train_speed(iter/s)": 1.618993
    },
    {
      "acc": 0.64703655,
      "epoch": 0.18049213597158803,
      "grad_norm": 5.9375,
      "learning_rate": 9.955782081370818e-06,
      "loss": 1.62678795,
      "memory(GiB)": 77.59,
      "step": 7115,
      "train_speed(iter/s)": 1.619117
    },
    {
      "acc": 0.65716386,
      "epoch": 0.18061897513952307,
      "grad_norm": 5.0625,
      "learning_rate": 9.955642821330752e-06,
      "loss": 1.67519836,
      "memory(GiB)": 77.59,
      "step": 7120,
      "train_speed(iter/s)": 1.61924
    },
    {
      "acc": 0.66499462,
      "epoch": 0.18074581430745815,
      "grad_norm": 5.53125,
      "learning_rate": 9.95550334331939e-06,
      "loss": 1.59537916,
      "memory(GiB)": 77.59,
      "step": 7125,
      "train_speed(iter/s)": 1.619351
    },
    {
      "acc": 0.65082536,
      "epoch": 0.1808726534753932,
      "grad_norm": 5.28125,
      "learning_rate": 9.955363647342868e-06,
      "loss": 1.60954227,
      "memory(GiB)": 77.59,
      "step": 7130,
      "train_speed(iter/s)": 1.619464
    },
    {
      "acc": 0.63223467,
      "epoch": 0.18099949264332826,
      "grad_norm": 6.5,
      "learning_rate": 9.955223733407327e-06,
      "loss": 1.71288204,
      "memory(GiB)": 77.59,
      "step": 7135,
      "train_speed(iter/s)": 1.619588
    },
    {
      "acc": 0.64555407,
      "epoch": 0.1811263318112633,
      "grad_norm": 5.5,
      "learning_rate": 9.955083601518924e-06,
      "loss": 1.71014099,
      "memory(GiB)": 77.59,
      "step": 7140,
      "train_speed(iter/s)": 1.619707
    },
    {
      "acc": 0.62587733,
      "epoch": 0.18125317097919838,
      "grad_norm": 5.5,
      "learning_rate": 9.95494325168382e-06,
      "loss": 1.74481926,
      "memory(GiB)": 77.59,
      "step": 7145,
      "train_speed(iter/s)": 1.619828
    },
    {
      "acc": 0.63875594,
      "epoch": 0.18138001014713342,
      "grad_norm": 5.28125,
      "learning_rate": 9.954802683908192e-06,
      "loss": 1.74068375,
      "memory(GiB)": 77.59,
      "step": 7150,
      "train_speed(iter/s)": 1.619954
    },
    {
      "acc": 0.64579372,
      "epoch": 0.1815068493150685,
      "grad_norm": 5.28125,
      "learning_rate": 9.954661898198216e-06,
      "loss": 1.66888332,
      "memory(GiB)": 77.59,
      "step": 7155,
      "train_speed(iter/s)": 1.620074
    },
    {
      "acc": 0.63566279,
      "epoch": 0.18163368848300354,
      "grad_norm": 5.9375,
      "learning_rate": 9.954520894560092e-06,
      "loss": 1.71961861,
      "memory(GiB)": 77.59,
      "step": 7160,
      "train_speed(iter/s)": 1.620204
    },
    {
      "acc": 0.63956718,
      "epoch": 0.1817605276509386,
      "grad_norm": 4.6875,
      "learning_rate": 9.954379673000018e-06,
      "loss": 1.70595989,
      "memory(GiB)": 77.59,
      "step": 7165,
      "train_speed(iter/s)": 1.620323
    },
    {
      "acc": 0.64019213,
      "epoch": 0.18188736681887366,
      "grad_norm": 5.3125,
      "learning_rate": 9.954238233524208e-06,
      "loss": 1.72009506,
      "memory(GiB)": 77.59,
      "step": 7170,
      "train_speed(iter/s)": 1.620443
    },
    {
      "acc": 0.64163303,
      "epoch": 0.18201420598680873,
      "grad_norm": 5.78125,
      "learning_rate": 9.954096576138879e-06,
      "loss": 1.73176422,
      "memory(GiB)": 77.59,
      "step": 7175,
      "train_speed(iter/s)": 1.620558
    },
    {
      "acc": 0.6387723,
      "epoch": 0.18214104515474377,
      "grad_norm": 5.375,
      "learning_rate": 9.953954700850264e-06,
      "loss": 1.71393185,
      "memory(GiB)": 77.59,
      "step": 7180,
      "train_speed(iter/s)": 1.620686
    },
    {
      "acc": 0.63552141,
      "epoch": 0.18226788432267885,
      "grad_norm": 4.84375,
      "learning_rate": 9.953812607664607e-06,
      "loss": 1.72599831,
      "memory(GiB)": 77.59,
      "step": 7185,
      "train_speed(iter/s)": 1.6208
    },
    {
      "acc": 0.65808129,
      "epoch": 0.1823947234906139,
      "grad_norm": 6.15625,
      "learning_rate": 9.95367029658815e-06,
      "loss": 1.67959442,
      "memory(GiB)": 77.59,
      "step": 7190,
      "train_speed(iter/s)": 1.620908
    },
    {
      "acc": 0.63875608,
      "epoch": 0.18252156265854896,
      "grad_norm": 6.4375,
      "learning_rate": 9.953527767627159e-06,
      "loss": 1.62620068,
      "memory(GiB)": 77.59,
      "step": 7195,
      "train_speed(iter/s)": 1.621024
    },
    {
      "acc": 0.66544304,
      "epoch": 0.182648401826484,
      "grad_norm": 4.96875,
      "learning_rate": 9.9533850207879e-06,
      "loss": 1.60160732,
      "memory(GiB)": 77.59,
      "step": 7200,
      "train_speed(iter/s)": 1.621144
    },
    {
      "acc": 0.65402861,
      "epoch": 0.18277524099441908,
      "grad_norm": 6.125,
      "learning_rate": 9.953242056076652e-06,
      "loss": 1.6854044,
      "memory(GiB)": 77.59,
      "step": 7205,
      "train_speed(iter/s)": 1.621263
    },
    {
      "acc": 0.63643265,
      "epoch": 0.18290208016235412,
      "grad_norm": 6.34375,
      "learning_rate": 9.953098873499705e-06,
      "loss": 1.73724136,
      "memory(GiB)": 77.59,
      "step": 7210,
      "train_speed(iter/s)": 1.62139
    },
    {
      "acc": 0.650736,
      "epoch": 0.1830289193302892,
      "grad_norm": 5.15625,
      "learning_rate": 9.952955473063356e-06,
      "loss": 1.61893234,
      "memory(GiB)": 77.59,
      "step": 7215,
      "train_speed(iter/s)": 1.621515
    },
    {
      "acc": 0.63614092,
      "epoch": 0.18315575849822424,
      "grad_norm": 5.40625,
      "learning_rate": 9.952811854773911e-06,
      "loss": 1.72106552,
      "memory(GiB)": 77.59,
      "step": 7220,
      "train_speed(iter/s)": 1.621638
    },
    {
      "acc": 0.63416362,
      "epoch": 0.1832825976661593,
      "grad_norm": 6.3125,
      "learning_rate": 9.952668018637687e-06,
      "loss": 1.7122797,
      "memory(GiB)": 77.59,
      "step": 7225,
      "train_speed(iter/s)": 1.621759
    },
    {
      "acc": 0.64628859,
      "epoch": 0.18340943683409436,
      "grad_norm": 5.625,
      "learning_rate": 9.952523964661014e-06,
      "loss": 1.66565475,
      "memory(GiB)": 77.59,
      "step": 7230,
      "train_speed(iter/s)": 1.621883
    },
    {
      "acc": 0.64756575,
      "epoch": 0.18353627600202943,
      "grad_norm": 6.15625,
      "learning_rate": 9.952379692850222e-06,
      "loss": 1.6767374,
      "memory(GiB)": 77.59,
      "step": 7235,
      "train_speed(iter/s)": 1.622007
    },
    {
      "acc": 0.65332966,
      "epoch": 0.18366311516996447,
      "grad_norm": 9.3125,
      "learning_rate": 9.952235203211663e-06,
      "loss": 1.62064381,
      "memory(GiB)": 77.59,
      "step": 7240,
      "train_speed(iter/s)": 1.622129
    },
    {
      "acc": 0.65426602,
      "epoch": 0.18378995433789955,
      "grad_norm": 5.25,
      "learning_rate": 9.952090495751689e-06,
      "loss": 1.64998131,
      "memory(GiB)": 77.59,
      "step": 7245,
      "train_speed(iter/s)": 1.622253
    },
    {
      "acc": 0.64379072,
      "epoch": 0.1839167935058346,
      "grad_norm": 6.0625,
      "learning_rate": 9.951945570476666e-06,
      "loss": 1.65906219,
      "memory(GiB)": 77.59,
      "step": 7250,
      "train_speed(iter/s)": 1.622381
    },
    {
      "acc": 0.62811546,
      "epoch": 0.18404363267376966,
      "grad_norm": 6.0,
      "learning_rate": 9.951800427392968e-06,
      "loss": 1.74646511,
      "memory(GiB)": 77.59,
      "step": 7255,
      "train_speed(iter/s)": 1.622507
    },
    {
      "acc": 0.64205914,
      "epoch": 0.1841704718417047,
      "grad_norm": 5.78125,
      "learning_rate": 9.951655066506977e-06,
      "loss": 1.75297375,
      "memory(GiB)": 77.59,
      "step": 7260,
      "train_speed(iter/s)": 1.622634
    },
    {
      "acc": 0.63896675,
      "epoch": 0.18429731100963978,
      "grad_norm": 5.4375,
      "learning_rate": 9.951509487825091e-06,
      "loss": 1.64735317,
      "memory(GiB)": 77.59,
      "step": 7265,
      "train_speed(iter/s)": 1.622749
    },
    {
      "acc": 0.6320116,
      "epoch": 0.18442415017757482,
      "grad_norm": 6.59375,
      "learning_rate": 9.95136369135371e-06,
      "loss": 1.73418388,
      "memory(GiB)": 77.59,
      "step": 7270,
      "train_speed(iter/s)": 1.622871
    },
    {
      "acc": 0.64414997,
      "epoch": 0.1845509893455099,
      "grad_norm": 5.25,
      "learning_rate": 9.951217677099248e-06,
      "loss": 1.69243927,
      "memory(GiB)": 77.59,
      "step": 7275,
      "train_speed(iter/s)": 1.622996
    },
    {
      "acc": 0.6418952,
      "epoch": 0.18467782851344494,
      "grad_norm": 5.28125,
      "learning_rate": 9.951071445068125e-06,
      "loss": 1.65538559,
      "memory(GiB)": 77.59,
      "step": 7280,
      "train_speed(iter/s)": 1.623118
    },
    {
      "acc": 0.63668184,
      "epoch": 0.18480466768138,
      "grad_norm": 6.1875,
      "learning_rate": 9.950924995266778e-06,
      "loss": 1.70533619,
      "memory(GiB)": 77.59,
      "step": 7285,
      "train_speed(iter/s)": 1.623243
    },
    {
      "acc": 0.6396327,
      "epoch": 0.18493150684931506,
      "grad_norm": 6.0,
      "learning_rate": 9.950778327701643e-06,
      "loss": 1.6748024,
      "memory(GiB)": 77.59,
      "step": 7290,
      "train_speed(iter/s)": 1.623367
    },
    {
      "acc": 0.65611219,
      "epoch": 0.18505834601725013,
      "grad_norm": 4.6875,
      "learning_rate": 9.950631442379175e-06,
      "loss": 1.66338844,
      "memory(GiB)": 77.59,
      "step": 7295,
      "train_speed(iter/s)": 1.623488
    },
    {
      "acc": 0.64618397,
      "epoch": 0.18518518518518517,
      "grad_norm": 6.03125,
      "learning_rate": 9.950484339305832e-06,
      "loss": 1.64510574,
      "memory(GiB)": 77.59,
      "step": 7300,
      "train_speed(iter/s)": 1.623595
    },
    {
      "acc": 0.65088601,
      "epoch": 0.18531202435312025,
      "grad_norm": 5.6875,
      "learning_rate": 9.950337018488086e-06,
      "loss": 1.70006027,
      "memory(GiB)": 77.59,
      "step": 7305,
      "train_speed(iter/s)": 1.623709
    },
    {
      "acc": 0.63807602,
      "epoch": 0.1854388635210553,
      "grad_norm": 5.0,
      "learning_rate": 9.950189479932417e-06,
      "loss": 1.69802361,
      "memory(GiB)": 77.59,
      "step": 7310,
      "train_speed(iter/s)": 1.623823
    },
    {
      "acc": 0.64677753,
      "epoch": 0.18556570268899036,
      "grad_norm": 5.28125,
      "learning_rate": 9.950041723645312e-06,
      "loss": 1.66775341,
      "memory(GiB)": 77.59,
      "step": 7315,
      "train_speed(iter/s)": 1.623944
    },
    {
      "acc": 0.64219332,
      "epoch": 0.1856925418569254,
      "grad_norm": 5.0,
      "learning_rate": 9.949893749633273e-06,
      "loss": 1.71438885,
      "memory(GiB)": 77.59,
      "step": 7320,
      "train_speed(iter/s)": 1.62406
    },
    {
      "acc": 0.64291296,
      "epoch": 0.18581938102486048,
      "grad_norm": 5.8125,
      "learning_rate": 9.949745557902806e-06,
      "loss": 1.73521843,
      "memory(GiB)": 77.59,
      "step": 7325,
      "train_speed(iter/s)": 1.624177
    },
    {
      "acc": 0.64869943,
      "epoch": 0.18594622019279552,
      "grad_norm": 4.53125,
      "learning_rate": 9.949597148460433e-06,
      "loss": 1.69890499,
      "memory(GiB)": 77.59,
      "step": 7330,
      "train_speed(iter/s)": 1.624298
    },
    {
      "acc": 0.63394394,
      "epoch": 0.1860730593607306,
      "grad_norm": 5.71875,
      "learning_rate": 9.949448521312676e-06,
      "loss": 1.79771175,
      "memory(GiB)": 77.59,
      "step": 7335,
      "train_speed(iter/s)": 1.624418
    },
    {
      "acc": 0.65008349,
      "epoch": 0.18619989852866564,
      "grad_norm": 7.4375,
      "learning_rate": 9.949299676466077e-06,
      "loss": 1.70881271,
      "memory(GiB)": 77.59,
      "step": 7340,
      "train_speed(iter/s)": 1.624544
    },
    {
      "acc": 0.6357439,
      "epoch": 0.1863267376966007,
      "grad_norm": 6.15625,
      "learning_rate": 9.94915061392718e-06,
      "loss": 1.66827774,
      "memory(GiB)": 77.59,
      "step": 7345,
      "train_speed(iter/s)": 1.624663
    },
    {
      "acc": 0.65615187,
      "epoch": 0.18645357686453576,
      "grad_norm": 5.40625,
      "learning_rate": 9.949001333702543e-06,
      "loss": 1.63151321,
      "memory(GiB)": 77.59,
      "step": 7350,
      "train_speed(iter/s)": 1.624785
    },
    {
      "acc": 0.63377533,
      "epoch": 0.18658041603247083,
      "grad_norm": 5.90625,
      "learning_rate": 9.948851835798732e-06,
      "loss": 1.7970108,
      "memory(GiB)": 77.59,
      "step": 7355,
      "train_speed(iter/s)": 1.624897
    },
    {
      "acc": 0.62856464,
      "epoch": 0.18670725520040587,
      "grad_norm": 5.03125,
      "learning_rate": 9.948702120222323e-06,
      "loss": 1.73523903,
      "memory(GiB)": 77.59,
      "step": 7360,
      "train_speed(iter/s)": 1.625016
    },
    {
      "acc": 0.66290793,
      "epoch": 0.18683409436834095,
      "grad_norm": 6.0,
      "learning_rate": 9.9485521869799e-06,
      "loss": 1.63394623,
      "memory(GiB)": 77.59,
      "step": 7365,
      "train_speed(iter/s)": 1.625131
    },
    {
      "acc": 0.66382289,
      "epoch": 0.186960933536276,
      "grad_norm": 5.28125,
      "learning_rate": 9.948402036078057e-06,
      "loss": 1.57046251,
      "memory(GiB)": 77.59,
      "step": 7370,
      "train_speed(iter/s)": 1.62526
    },
    {
      "acc": 0.63160882,
      "epoch": 0.18708777270421106,
      "grad_norm": 6.5625,
      "learning_rate": 9.948251667523401e-06,
      "loss": 1.72319946,
      "memory(GiB)": 77.59,
      "step": 7375,
      "train_speed(iter/s)": 1.625377
    },
    {
      "acc": 0.63488054,
      "epoch": 0.1872146118721461,
      "grad_norm": 5.375,
      "learning_rate": 9.948101081322544e-06,
      "loss": 1.62635155,
      "memory(GiB)": 77.59,
      "step": 7380,
      "train_speed(iter/s)": 1.625497
    },
    {
      "acc": 0.6456624,
      "epoch": 0.18734145104008118,
      "grad_norm": 4.9375,
      "learning_rate": 9.947950277482109e-06,
      "loss": 1.65658779,
      "memory(GiB)": 77.59,
      "step": 7385,
      "train_speed(iter/s)": 1.62562
    },
    {
      "acc": 0.64808559,
      "epoch": 0.18746829020801623,
      "grad_norm": 5.75,
      "learning_rate": 9.94779925600873e-06,
      "loss": 1.68540001,
      "memory(GiB)": 77.59,
      "step": 7390,
      "train_speed(iter/s)": 1.625739
    },
    {
      "acc": 0.66459303,
      "epoch": 0.1875951293759513,
      "grad_norm": 5.25,
      "learning_rate": 9.947648016909048e-06,
      "loss": 1.63555069,
      "memory(GiB)": 77.59,
      "step": 7395,
      "train_speed(iter/s)": 1.625863
    },
    {
      "acc": 0.64715314,
      "epoch": 0.18772196854388634,
      "grad_norm": 6.3125,
      "learning_rate": 9.947496560189717e-06,
      "loss": 1.66416512,
      "memory(GiB)": 77.59,
      "step": 7400,
      "train_speed(iter/s)": 1.625984
    },
    {
      "acc": 0.64939599,
      "epoch": 0.18784880771182141,
      "grad_norm": 5.25,
      "learning_rate": 9.9473448858574e-06,
      "loss": 1.68471889,
      "memory(GiB)": 77.59,
      "step": 7405,
      "train_speed(iter/s)": 1.626106
    },
    {
      "acc": 0.62002168,
      "epoch": 0.18797564687975646,
      "grad_norm": 5.96875,
      "learning_rate": 9.947192993918765e-06,
      "loss": 1.78501797,
      "memory(GiB)": 77.59,
      "step": 7410,
      "train_speed(iter/s)": 1.626217
    },
    {
      "acc": 0.632866,
      "epoch": 0.18810248604769153,
      "grad_norm": 5.96875,
      "learning_rate": 9.947040884380496e-06,
      "loss": 1.70102005,
      "memory(GiB)": 77.59,
      "step": 7415,
      "train_speed(iter/s)": 1.62633
    },
    {
      "acc": 0.65509653,
      "epoch": 0.18822932521562658,
      "grad_norm": 6.40625,
      "learning_rate": 9.946888557249281e-06,
      "loss": 1.65170479,
      "memory(GiB)": 77.59,
      "step": 7420,
      "train_speed(iter/s)": 1.626448
    },
    {
      "acc": 0.64921708,
      "epoch": 0.18835616438356165,
      "grad_norm": 5.40625,
      "learning_rate": 9.946736012531821e-06,
      "loss": 1.66297741,
      "memory(GiB)": 77.59,
      "step": 7425,
      "train_speed(iter/s)": 1.626577
    },
    {
      "acc": 0.64842544,
      "epoch": 0.1884830035514967,
      "grad_norm": 5.1875,
      "learning_rate": 9.946583250234826e-06,
      "loss": 1.65580101,
      "memory(GiB)": 77.59,
      "step": 7430,
      "train_speed(iter/s)": 1.626698
    },
    {
      "acc": 0.64534473,
      "epoch": 0.18860984271943176,
      "grad_norm": 4.84375,
      "learning_rate": 9.946430270365015e-06,
      "loss": 1.61940098,
      "memory(GiB)": 77.59,
      "step": 7435,
      "train_speed(iter/s)": 1.626814
    },
    {
      "acc": 0.65221057,
      "epoch": 0.1887366818873668,
      "grad_norm": 5.71875,
      "learning_rate": 9.946277072929115e-06,
      "loss": 1.64499588,
      "memory(GiB)": 77.59,
      "step": 7440,
      "train_speed(iter/s)": 1.626932
    },
    {
      "acc": 0.64241719,
      "epoch": 0.18886352105530188,
      "grad_norm": 5.3125,
      "learning_rate": 9.946123657933867e-06,
      "loss": 1.69488716,
      "memory(GiB)": 77.59,
      "step": 7445,
      "train_speed(iter/s)": 1.627039
    },
    {
      "acc": 0.64012222,
      "epoch": 0.18899036022323693,
      "grad_norm": 5.71875,
      "learning_rate": 9.945970025386018e-06,
      "loss": 1.62594109,
      "memory(GiB)": 77.59,
      "step": 7450,
      "train_speed(iter/s)": 1.627153
    },
    {
      "acc": 0.64713979,
      "epoch": 0.189117199391172,
      "grad_norm": 6.46875,
      "learning_rate": 9.945816175292326e-06,
      "loss": 1.6352438,
      "memory(GiB)": 77.59,
      "step": 7455,
      "train_speed(iter/s)": 1.627272
    },
    {
      "acc": 0.64514918,
      "epoch": 0.18924403855910704,
      "grad_norm": 5.3125,
      "learning_rate": 9.945662107659554e-06,
      "loss": 1.71115894,
      "memory(GiB)": 77.59,
      "step": 7460,
      "train_speed(iter/s)": 1.627388
    },
    {
      "acc": 0.65765886,
      "epoch": 0.18937087772704211,
      "grad_norm": 7.625,
      "learning_rate": 9.945507822494485e-06,
      "loss": 1.63517151,
      "memory(GiB)": 77.59,
      "step": 7465,
      "train_speed(iter/s)": 1.627502
    },
    {
      "acc": 0.64353065,
      "epoch": 0.18949771689497716,
      "grad_norm": 5.96875,
      "learning_rate": 9.9453533198039e-06,
      "loss": 1.68638687,
      "memory(GiB)": 77.59,
      "step": 7470,
      "train_speed(iter/s)": 1.627615
    },
    {
      "acc": 0.63392801,
      "epoch": 0.18962455606291223,
      "grad_norm": 5.3125,
      "learning_rate": 9.945198599594598e-06,
      "loss": 1.66831703,
      "memory(GiB)": 77.59,
      "step": 7475,
      "train_speed(iter/s)": 1.627736
    },
    {
      "acc": 0.63619165,
      "epoch": 0.18975139523084728,
      "grad_norm": 4.84375,
      "learning_rate": 9.945043661873381e-06,
      "loss": 1.73271637,
      "memory(GiB)": 77.59,
      "step": 7480,
      "train_speed(iter/s)": 1.627849
    },
    {
      "acc": 0.65396185,
      "epoch": 0.18987823439878235,
      "grad_norm": 4.84375,
      "learning_rate": 9.944888506647066e-06,
      "loss": 1.6267189,
      "memory(GiB)": 77.59,
      "step": 7485,
      "train_speed(iter/s)": 1.627967
    },
    {
      "acc": 0.65445547,
      "epoch": 0.1900050735667174,
      "grad_norm": 5.875,
      "learning_rate": 9.944733133922479e-06,
      "loss": 1.65859509,
      "memory(GiB)": 77.59,
      "step": 7490,
      "train_speed(iter/s)": 1.628086
    },
    {
      "acc": 0.64349761,
      "epoch": 0.19013191273465246,
      "grad_norm": 6.0625,
      "learning_rate": 9.944577543706451e-06,
      "loss": 1.66820087,
      "memory(GiB)": 77.59,
      "step": 7495,
      "train_speed(iter/s)": 1.628209
    },
    {
      "acc": 0.63627405,
      "epoch": 0.1902587519025875,
      "grad_norm": 5.40625,
      "learning_rate": 9.944421736005825e-06,
      "loss": 1.66228065,
      "memory(GiB)": 77.59,
      "step": 7500,
      "train_speed(iter/s)": 1.628326
    },
    {
      "acc": 0.63509989,
      "epoch": 0.19038559107052258,
      "grad_norm": 5.1875,
      "learning_rate": 9.944265710827459e-06,
      "loss": 1.75539074,
      "memory(GiB)": 77.59,
      "step": 7505,
      "train_speed(iter/s)": 1.628447
    },
    {
      "acc": 0.64721031,
      "epoch": 0.19051243023845763,
      "grad_norm": 5.15625,
      "learning_rate": 9.944109468178208e-06,
      "loss": 1.64554634,
      "memory(GiB)": 77.59,
      "step": 7510,
      "train_speed(iter/s)": 1.628563
    },
    {
      "acc": 0.64745188,
      "epoch": 0.1906392694063927,
      "grad_norm": 5.0,
      "learning_rate": 9.943953008064953e-06,
      "loss": 1.67710266,
      "memory(GiB)": 77.59,
      "step": 7515,
      "train_speed(iter/s)": 1.62868
    },
    {
      "acc": 0.64476213,
      "epoch": 0.19076610857432774,
      "grad_norm": 6.34375,
      "learning_rate": 9.94379633049457e-06,
      "loss": 1.68845291,
      "memory(GiB)": 77.59,
      "step": 7520,
      "train_speed(iter/s)": 1.6288
    },
    {
      "acc": 0.6497797,
      "epoch": 0.19089294774226281,
      "grad_norm": 4.9375,
      "learning_rate": 9.943639435473952e-06,
      "loss": 1.69892578,
      "memory(GiB)": 77.59,
      "step": 7525,
      "train_speed(iter/s)": 1.628916
    },
    {
      "acc": 0.65878482,
      "epoch": 0.19101978691019786,
      "grad_norm": 5.40625,
      "learning_rate": 9.94348232301e-06,
      "loss": 1.63249588,
      "memory(GiB)": 77.59,
      "step": 7530,
      "train_speed(iter/s)": 1.629027
    },
    {
      "acc": 0.63181257,
      "epoch": 0.19114662607813293,
      "grad_norm": 6.0625,
      "learning_rate": 9.943324993109624e-06,
      "loss": 1.71280022,
      "memory(GiB)": 77.59,
      "step": 7535,
      "train_speed(iter/s)": 1.629141
    },
    {
      "acc": 0.65265141,
      "epoch": 0.19127346524606798,
      "grad_norm": 7.25,
      "learning_rate": 9.943167445779745e-06,
      "loss": 1.61639557,
      "memory(GiB)": 77.59,
      "step": 7540,
      "train_speed(iter/s)": 1.629258
    },
    {
      "acc": 0.67400455,
      "epoch": 0.19140030441400305,
      "grad_norm": 6.5625,
      "learning_rate": 9.94300968102729e-06,
      "loss": 1.60792465,
      "memory(GiB)": 77.59,
      "step": 7545,
      "train_speed(iter/s)": 1.629374
    },
    {
      "acc": 0.65444889,
      "epoch": 0.1915271435819381,
      "grad_norm": 5.53125,
      "learning_rate": 9.942851698859204e-06,
      "loss": 1.64874229,
      "memory(GiB)": 77.59,
      "step": 7550,
      "train_speed(iter/s)": 1.629491
    },
    {
      "acc": 0.65789003,
      "epoch": 0.19165398274987316,
      "grad_norm": 6.0625,
      "learning_rate": 9.94269349928243e-06,
      "loss": 1.60055962,
      "memory(GiB)": 77.59,
      "step": 7555,
      "train_speed(iter/s)": 1.629602
    },
    {
      "acc": 0.6613224,
      "epoch": 0.1917808219178082,
      "grad_norm": 5.1875,
      "learning_rate": 9.942535082303927e-06,
      "loss": 1.58959074,
      "memory(GiB)": 77.59,
      "step": 7560,
      "train_speed(iter/s)": 1.629712
    },
    {
      "acc": 0.64439492,
      "epoch": 0.19190766108574328,
      "grad_norm": 5.75,
      "learning_rate": 9.942376447930666e-06,
      "loss": 1.68322124,
      "memory(GiB)": 77.59,
      "step": 7565,
      "train_speed(iter/s)": 1.629823
    },
    {
      "acc": 0.65919485,
      "epoch": 0.19203450025367833,
      "grad_norm": 4.9375,
      "learning_rate": 9.942217596169623e-06,
      "loss": 1.57794476,
      "memory(GiB)": 77.59,
      "step": 7570,
      "train_speed(iter/s)": 1.629936
    },
    {
      "acc": 0.62779818,
      "epoch": 0.1921613394216134,
      "grad_norm": 4.96875,
      "learning_rate": 9.942058527027785e-06,
      "loss": 1.73289337,
      "memory(GiB)": 77.59,
      "step": 7575,
      "train_speed(iter/s)": 1.630051
    },
    {
      "acc": 0.62824359,
      "epoch": 0.19228817858954844,
      "grad_norm": 6.65625,
      "learning_rate": 9.941899240512147e-06,
      "loss": 1.69513054,
      "memory(GiB)": 77.59,
      "step": 7580,
      "train_speed(iter/s)": 1.630168
    },
    {
      "acc": 0.63809719,
      "epoch": 0.19241501775748351,
      "grad_norm": 5.5,
      "learning_rate": 9.941739736629716e-06,
      "loss": 1.67402,
      "memory(GiB)": 77.59,
      "step": 7585,
      "train_speed(iter/s)": 1.630282
    },
    {
      "acc": 0.64863114,
      "epoch": 0.19254185692541856,
      "grad_norm": 5.375,
      "learning_rate": 9.941580015387509e-06,
      "loss": 1.61916904,
      "memory(GiB)": 77.59,
      "step": 7590,
      "train_speed(iter/s)": 1.630397
    },
    {
      "acc": 0.62263918,
      "epoch": 0.19266869609335363,
      "grad_norm": 8.9375,
      "learning_rate": 9.94142007679255e-06,
      "loss": 1.76624832,
      "memory(GiB)": 77.59,
      "step": 7595,
      "train_speed(iter/s)": 1.630518
    },
    {
      "acc": 0.66040039,
      "epoch": 0.19279553526128868,
      "grad_norm": 7.21875,
      "learning_rate": 9.941259920851874e-06,
      "loss": 1.6323246,
      "memory(GiB)": 77.59,
      "step": 7600,
      "train_speed(iter/s)": 1.630631
    },
    {
      "acc": 0.64601893,
      "epoch": 0.19292237442922375,
      "grad_norm": 7.6875,
      "learning_rate": 9.941099547572527e-06,
      "loss": 1.67999172,
      "memory(GiB)": 77.59,
      "step": 7605,
      "train_speed(iter/s)": 1.630755
    },
    {
      "acc": 0.65333796,
      "epoch": 0.1930492135971588,
      "grad_norm": 5.09375,
      "learning_rate": 9.94093895696156e-06,
      "loss": 1.65717735,
      "memory(GiB)": 77.59,
      "step": 7610,
      "train_speed(iter/s)": 1.630869
    },
    {
      "acc": 0.63890438,
      "epoch": 0.19317605276509386,
      "grad_norm": 6.15625,
      "learning_rate": 9.940778149026038e-06,
      "loss": 1.69429169,
      "memory(GiB)": 77.59,
      "step": 7615,
      "train_speed(iter/s)": 1.630981
    },
    {
      "acc": 0.6394372,
      "epoch": 0.1933028919330289,
      "grad_norm": 5.4375,
      "learning_rate": 9.940617123773036e-06,
      "loss": 1.69591408,
      "memory(GiB)": 77.59,
      "step": 7620,
      "train_speed(iter/s)": 1.631098
    },
    {
      "acc": 0.65708766,
      "epoch": 0.19342973110096398,
      "grad_norm": 5.75,
      "learning_rate": 9.940455881209632e-06,
      "loss": 1.57650871,
      "memory(GiB)": 77.59,
      "step": 7625,
      "train_speed(iter/s)": 1.631219
    },
    {
      "acc": 0.64314799,
      "epoch": 0.19355657026889903,
      "grad_norm": 5.59375,
      "learning_rate": 9.940294421342922e-06,
      "loss": 1.60519085,
      "memory(GiB)": 77.59,
      "step": 7630,
      "train_speed(iter/s)": 1.631332
    },
    {
      "acc": 0.6334199,
      "epoch": 0.1936834094368341,
      "grad_norm": 6.15625,
      "learning_rate": 9.940132744180007e-06,
      "loss": 1.7166523,
      "memory(GiB)": 77.59,
      "step": 7635,
      "train_speed(iter/s)": 1.631445
    },
    {
      "acc": 0.65944061,
      "epoch": 0.19381024860476914,
      "grad_norm": 5.5625,
      "learning_rate": 9.939970849727995e-06,
      "loss": 1.61455383,
      "memory(GiB)": 77.59,
      "step": 7640,
      "train_speed(iter/s)": 1.631554
    },
    {
      "acc": 0.63408394,
      "epoch": 0.19393708777270421,
      "grad_norm": 5.6875,
      "learning_rate": 9.939808737994013e-06,
      "loss": 1.72968178,
      "memory(GiB)": 77.59,
      "step": 7645,
      "train_speed(iter/s)": 1.631674
    },
    {
      "acc": 0.6466785,
      "epoch": 0.19406392694063926,
      "grad_norm": 5.90625,
      "learning_rate": 9.939646408985186e-06,
      "loss": 1.61666393,
      "memory(GiB)": 77.59,
      "step": 7650,
      "train_speed(iter/s)": 1.631785
    },
    {
      "acc": 0.65346975,
      "epoch": 0.19419076610857433,
      "grad_norm": 5.9375,
      "learning_rate": 9.939483862708658e-06,
      "loss": 1.62012177,
      "memory(GiB)": 77.59,
      "step": 7655,
      "train_speed(iter/s)": 1.631903
    },
    {
      "acc": 0.64003558,
      "epoch": 0.19431760527650938,
      "grad_norm": 5.75,
      "learning_rate": 9.939321099171575e-06,
      "loss": 1.64041882,
      "memory(GiB)": 77.59,
      "step": 7660,
      "train_speed(iter/s)": 1.632007
    },
    {
      "acc": 0.64093351,
      "epoch": 0.19444444444444445,
      "grad_norm": 4.9375,
      "learning_rate": 9.939158118381097e-06,
      "loss": 1.72127285,
      "memory(GiB)": 77.59,
      "step": 7665,
      "train_speed(iter/s)": 1.63212
    },
    {
      "acc": 0.63992558,
      "epoch": 0.1945712836123795,
      "grad_norm": 6.0625,
      "learning_rate": 9.938994920344395e-06,
      "loss": 1.68768444,
      "memory(GiB)": 77.59,
      "step": 7670,
      "train_speed(iter/s)": 1.632232
    },
    {
      "acc": 0.64807968,
      "epoch": 0.19469812278031456,
      "grad_norm": 5.0,
      "learning_rate": 9.938831505068645e-06,
      "loss": 1.62658596,
      "memory(GiB)": 77.59,
      "step": 7675,
      "train_speed(iter/s)": 1.632355
    },
    {
      "acc": 0.64729881,
      "epoch": 0.1948249619482496,
      "grad_norm": 5.375,
      "learning_rate": 9.938667872561035e-06,
      "loss": 1.58645487,
      "memory(GiB)": 77.59,
      "step": 7680,
      "train_speed(iter/s)": 1.632472
    },
    {
      "acc": 0.6605937,
      "epoch": 0.19495180111618468,
      "grad_norm": 5.46875,
      "learning_rate": 9.938504022828762e-06,
      "loss": 1.58746214,
      "memory(GiB)": 77.59,
      "step": 7685,
      "train_speed(iter/s)": 1.632586
    },
    {
      "acc": 0.64638605,
      "epoch": 0.19507864028411973,
      "grad_norm": 4.53125,
      "learning_rate": 9.938339955879033e-06,
      "loss": 1.6381422,
      "memory(GiB)": 77.59,
      "step": 7690,
      "train_speed(iter/s)": 1.632697
    },
    {
      "acc": 0.64010434,
      "epoch": 0.1952054794520548,
      "grad_norm": 5.21875,
      "learning_rate": 9.938175671719064e-06,
      "loss": 1.67854919,
      "memory(GiB)": 77.59,
      "step": 7695,
      "train_speed(iter/s)": 1.632807
    },
    {
      "acc": 0.62463942,
      "epoch": 0.19533231861998984,
      "grad_norm": 5.78125,
      "learning_rate": 9.938011170356083e-06,
      "loss": 1.65680466,
      "memory(GiB)": 77.59,
      "step": 7700,
      "train_speed(iter/s)": 1.63292
    },
    {
      "acc": 0.63911147,
      "epoch": 0.19545915778792491,
      "grad_norm": 8.5,
      "learning_rate": 9.937846451797324e-06,
      "loss": 1.67437267,
      "memory(GiB)": 77.59,
      "step": 7705,
      "train_speed(iter/s)": 1.633029
    },
    {
      "acc": 0.65099292,
      "epoch": 0.19558599695585996,
      "grad_norm": 7.03125,
      "learning_rate": 9.93768151605003e-06,
      "loss": 1.61787701,
      "memory(GiB)": 77.59,
      "step": 7710,
      "train_speed(iter/s)": 1.633139
    },
    {
      "acc": 0.65595174,
      "epoch": 0.19571283612379503,
      "grad_norm": 6.125,
      "learning_rate": 9.93751636312146e-06,
      "loss": 1.62009678,
      "memory(GiB)": 77.59,
      "step": 7715,
      "train_speed(iter/s)": 1.63325
    },
    {
      "acc": 0.63047824,
      "epoch": 0.19583967529173008,
      "grad_norm": 6.90625,
      "learning_rate": 9.937350993018875e-06,
      "loss": 1.75598621,
      "memory(GiB)": 77.59,
      "step": 7720,
      "train_speed(iter/s)": 1.633362
    },
    {
      "acc": 0.6408926,
      "epoch": 0.19596651445966515,
      "grad_norm": 4.84375,
      "learning_rate": 9.93718540574955e-06,
      "loss": 1.65693874,
      "memory(GiB)": 77.59,
      "step": 7725,
      "train_speed(iter/s)": 1.633478
    },
    {
      "acc": 0.63349919,
      "epoch": 0.1960933536276002,
      "grad_norm": 5.0,
      "learning_rate": 9.937019601320768e-06,
      "loss": 1.72434883,
      "memory(GiB)": 77.59,
      "step": 7730,
      "train_speed(iter/s)": 1.633592
    },
    {
      "acc": 0.62782717,
      "epoch": 0.19622019279553526,
      "grad_norm": 5.71875,
      "learning_rate": 9.936853579739823e-06,
      "loss": 1.67874832,
      "memory(GiB)": 77.59,
      "step": 7735,
      "train_speed(iter/s)": 1.633704
    },
    {
      "acc": 0.62826281,
      "epoch": 0.1963470319634703,
      "grad_norm": 6.9375,
      "learning_rate": 9.936687341014015e-06,
      "loss": 1.75626316,
      "memory(GiB)": 77.59,
      "step": 7740,
      "train_speed(iter/s)": 1.633807
    },
    {
      "acc": 0.65873413,
      "epoch": 0.19647387113140538,
      "grad_norm": 5.40625,
      "learning_rate": 9.936520885150655e-06,
      "loss": 1.5933773,
      "memory(GiB)": 77.59,
      "step": 7745,
      "train_speed(iter/s)": 1.633913
    },
    {
      "acc": 0.64515133,
      "epoch": 0.19660071029934043,
      "grad_norm": 5.1875,
      "learning_rate": 9.936354212157068e-06,
      "loss": 1.71064873,
      "memory(GiB)": 77.59,
      "step": 7750,
      "train_speed(iter/s)": 1.634017
    },
    {
      "acc": 0.64298263,
      "epoch": 0.1967275494672755,
      "grad_norm": 4.5,
      "learning_rate": 9.936187322040584e-06,
      "loss": 1.68684635,
      "memory(GiB)": 77.59,
      "step": 7755,
      "train_speed(iter/s)": 1.634132
    },
    {
      "acc": 0.63161044,
      "epoch": 0.19685438863521054,
      "grad_norm": 5.125,
      "learning_rate": 9.936020214808544e-06,
      "loss": 1.72244186,
      "memory(GiB)": 77.59,
      "step": 7760,
      "train_speed(iter/s)": 1.634242
    },
    {
      "acc": 0.65020905,
      "epoch": 0.19698122780314561,
      "grad_norm": 5.5,
      "learning_rate": 9.935852890468297e-06,
      "loss": 1.6733654,
      "memory(GiB)": 77.59,
      "step": 7765,
      "train_speed(iter/s)": 1.634352
    },
    {
      "acc": 0.63110623,
      "epoch": 0.19710806697108066,
      "grad_norm": 8.125,
      "learning_rate": 9.935685349027201e-06,
      "loss": 1.68286629,
      "memory(GiB)": 77.59,
      "step": 7770,
      "train_speed(iter/s)": 1.634463
    },
    {
      "acc": 0.63578558,
      "epoch": 0.19723490613901573,
      "grad_norm": 5.34375,
      "learning_rate": 9.935517590492627e-06,
      "loss": 1.67285881,
      "memory(GiB)": 77.59,
      "step": 7775,
      "train_speed(iter/s)": 1.634571
    },
    {
      "acc": 0.6535295,
      "epoch": 0.19736174530695078,
      "grad_norm": 5.25,
      "learning_rate": 9.935349614871957e-06,
      "loss": 1.61393757,
      "memory(GiB)": 77.59,
      "step": 7780,
      "train_speed(iter/s)": 1.634683
    },
    {
      "acc": 0.65299621,
      "epoch": 0.19748858447488585,
      "grad_norm": 5.3125,
      "learning_rate": 9.935181422172574e-06,
      "loss": 1.66883202,
      "memory(GiB)": 77.59,
      "step": 7785,
      "train_speed(iter/s)": 1.634797
    },
    {
      "acc": 0.63826895,
      "epoch": 0.1976154236428209,
      "grad_norm": 6.5625,
      "learning_rate": 9.935013012401878e-06,
      "loss": 1.77757244,
      "memory(GiB)": 77.59,
      "step": 7790,
      "train_speed(iter/s)": 1.634914
    },
    {
      "acc": 0.65492325,
      "epoch": 0.19774226281075596,
      "grad_norm": 5.75,
      "learning_rate": 9.934844385567275e-06,
      "loss": 1.61685257,
      "memory(GiB)": 77.59,
      "step": 7795,
      "train_speed(iter/s)": 1.635024
    },
    {
      "acc": 0.64512129,
      "epoch": 0.197869101978691,
      "grad_norm": 6.3125,
      "learning_rate": 9.934675541676186e-06,
      "loss": 1.65668201,
      "memory(GiB)": 77.59,
      "step": 7800,
      "train_speed(iter/s)": 1.63514
    },
    {
      "acc": 0.65817451,
      "epoch": 0.19799594114662608,
      "grad_norm": 6.84375,
      "learning_rate": 9.934506480736034e-06,
      "loss": 1.62153244,
      "memory(GiB)": 77.59,
      "step": 7805,
      "train_speed(iter/s)": 1.635252
    },
    {
      "acc": 0.65557733,
      "epoch": 0.19812278031456113,
      "grad_norm": 5.6875,
      "learning_rate": 9.934337202754257e-06,
      "loss": 1.6400877,
      "memory(GiB)": 77.59,
      "step": 7810,
      "train_speed(iter/s)": 1.635368
    },
    {
      "acc": 0.65059519,
      "epoch": 0.1982496194824962,
      "grad_norm": 6.25,
      "learning_rate": 9.934167707738298e-06,
      "loss": 1.71084137,
      "memory(GiB)": 77.59,
      "step": 7815,
      "train_speed(iter/s)": 1.635488
    },
    {
      "acc": 0.64949894,
      "epoch": 0.19837645865043124,
      "grad_norm": 6.875,
      "learning_rate": 9.933997995695615e-06,
      "loss": 1.66692123,
      "memory(GiB)": 77.59,
      "step": 7820,
      "train_speed(iter/s)": 1.635587
    },
    {
      "acc": 0.6385705,
      "epoch": 0.19850329781836631,
      "grad_norm": 5.28125,
      "learning_rate": 9.93382806663367e-06,
      "loss": 1.71105919,
      "memory(GiB)": 77.59,
      "step": 7825,
      "train_speed(iter/s)": 1.635702
    },
    {
      "acc": 0.64712896,
      "epoch": 0.19863013698630136,
      "grad_norm": 5.65625,
      "learning_rate": 9.933657920559939e-06,
      "loss": 1.65789585,
      "memory(GiB)": 77.59,
      "step": 7830,
      "train_speed(iter/s)": 1.635812
    },
    {
      "acc": 0.64820595,
      "epoch": 0.19875697615423643,
      "grad_norm": 5.34375,
      "learning_rate": 9.933487557481905e-06,
      "loss": 1.67896671,
      "memory(GiB)": 77.59,
      "step": 7835,
      "train_speed(iter/s)": 1.635923
    },
    {
      "acc": 0.65184479,
      "epoch": 0.19888381532217148,
      "grad_norm": 5.28125,
      "learning_rate": 9.933316977407063e-06,
      "loss": 1.72082691,
      "memory(GiB)": 77.59,
      "step": 7840,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.66542158,
      "epoch": 0.19901065449010655,
      "grad_norm": 5.0625,
      "learning_rate": 9.933146180342914e-06,
      "loss": 1.56001062,
      "memory(GiB)": 77.59,
      "step": 7845,
      "train_speed(iter/s)": 1.636133
    },
    {
      "acc": 0.65702295,
      "epoch": 0.1991374936580416,
      "grad_norm": 5.15625,
      "learning_rate": 9.932975166296972e-06,
      "loss": 1.65709152,
      "memory(GiB)": 77.59,
      "step": 7850,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.64513555,
      "epoch": 0.19926433282597666,
      "grad_norm": 6.5,
      "learning_rate": 9.932803935276757e-06,
      "loss": 1.6050312,
      "memory(GiB)": 77.59,
      "step": 7855,
      "train_speed(iter/s)": 1.636336
    },
    {
      "acc": 0.63779025,
      "epoch": 0.1993911719939117,
      "grad_norm": 6.59375,
      "learning_rate": 9.932632487289802e-06,
      "loss": 1.73334351,
      "memory(GiB)": 77.59,
      "step": 7860,
      "train_speed(iter/s)": 1.636443
    },
    {
      "acc": 0.64397364,
      "epoch": 0.19951801116184678,
      "grad_norm": 5.4375,
      "learning_rate": 9.932460822343649e-06,
      "loss": 1.66076736,
      "memory(GiB)": 77.59,
      "step": 7865,
      "train_speed(iter/s)": 1.636548
    },
    {
      "acc": 0.63360085,
      "epoch": 0.19964485032978183,
      "grad_norm": 6.03125,
      "learning_rate": 9.932288940445845e-06,
      "loss": 1.79628563,
      "memory(GiB)": 77.59,
      "step": 7870,
      "train_speed(iter/s)": 1.636653
    },
    {
      "acc": 0.6418045,
      "epoch": 0.1997716894977169,
      "grad_norm": 6.5625,
      "learning_rate": 9.932116841603954e-06,
      "loss": 1.61534061,
      "memory(GiB)": 77.59,
      "step": 7875,
      "train_speed(iter/s)": 1.636764
    },
    {
      "acc": 0.65211296,
      "epoch": 0.19989852866565194,
      "grad_norm": 7.21875,
      "learning_rate": 9.931944525825542e-06,
      "loss": 1.64864426,
      "memory(GiB)": 77.59,
      "step": 7880,
      "train_speed(iter/s)": 1.636877
    },
    {
      "acc": 0.64422231,
      "epoch": 0.20002536783358701,
      "grad_norm": 6.53125,
      "learning_rate": 9.931771993118191e-06,
      "loss": 1.61573906,
      "memory(GiB)": 77.59,
      "step": 7885,
      "train_speed(iter/s)": 1.636974
    },
    {
      "acc": 0.64482603,
      "epoch": 0.20015220700152206,
      "grad_norm": 4.46875,
      "learning_rate": 9.931599243489489e-06,
      "loss": 1.69481201,
      "memory(GiB)": 77.59,
      "step": 7890,
      "train_speed(iter/s)": 1.637078
    },
    {
      "acc": 0.64780335,
      "epoch": 0.20027904616945713,
      "grad_norm": 5.78125,
      "learning_rate": 9.931426276947037e-06,
      "loss": 1.65161324,
      "memory(GiB)": 77.59,
      "step": 7895,
      "train_speed(iter/s)": 1.637189
    },
    {
      "acc": 0.63699322,
      "epoch": 0.20040588533739218,
      "grad_norm": 6.0625,
      "learning_rate": 9.931253093498437e-06,
      "loss": 1.65363426,
      "memory(GiB)": 77.59,
      "step": 7900,
      "train_speed(iter/s)": 1.637298
    },
    {
      "acc": 0.63848782,
      "epoch": 0.20053272450532725,
      "grad_norm": 4.34375,
      "learning_rate": 9.93107969315131e-06,
      "loss": 1.6675106,
      "memory(GiB)": 77.59,
      "step": 7905,
      "train_speed(iter/s)": 1.637403
    },
    {
      "acc": 0.65617108,
      "epoch": 0.2006595636732623,
      "grad_norm": 5.09375,
      "learning_rate": 9.930906075913281e-06,
      "loss": 1.66813869,
      "memory(GiB)": 77.59,
      "step": 7910,
      "train_speed(iter/s)": 1.637508
    },
    {
      "acc": 0.65991898,
      "epoch": 0.20078640284119736,
      "grad_norm": 4.8125,
      "learning_rate": 9.93073224179199e-06,
      "loss": 1.62511177,
      "memory(GiB)": 77.59,
      "step": 7915,
      "train_speed(iter/s)": 1.637602
    },
    {
      "acc": 0.64154482,
      "epoch": 0.2009132420091324,
      "grad_norm": 6.40625,
      "learning_rate": 9.93055819079508e-06,
      "loss": 1.72127247,
      "memory(GiB)": 77.59,
      "step": 7920,
      "train_speed(iter/s)": 1.637705
    },
    {
      "acc": 0.65048213,
      "epoch": 0.20104008117706748,
      "grad_norm": 4.59375,
      "learning_rate": 9.930383922930207e-06,
      "loss": 1.72336464,
      "memory(GiB)": 77.59,
      "step": 7925,
      "train_speed(iter/s)": 1.637805
    },
    {
      "acc": 0.63912992,
      "epoch": 0.20116692034500253,
      "grad_norm": 5.8125,
      "learning_rate": 9.930209438205038e-06,
      "loss": 1.70175743,
      "memory(GiB)": 77.59,
      "step": 7930,
      "train_speed(iter/s)": 1.637907
    },
    {
      "acc": 0.64162908,
      "epoch": 0.2012937595129376,
      "grad_norm": 5.0625,
      "learning_rate": 9.930034736627245e-06,
      "loss": 1.65782051,
      "memory(GiB)": 77.59,
      "step": 7935,
      "train_speed(iter/s)": 1.638017
    },
    {
      "acc": 0.64904346,
      "epoch": 0.20142059868087264,
      "grad_norm": 5.5,
      "learning_rate": 9.929859818204514e-06,
      "loss": 1.6428133,
      "memory(GiB)": 77.59,
      "step": 7940,
      "train_speed(iter/s)": 1.638115
    },
    {
      "acc": 0.66294374,
      "epoch": 0.20154743784880771,
      "grad_norm": 5.0,
      "learning_rate": 9.929684682944538e-06,
      "loss": 1.51369743,
      "memory(GiB)": 77.59,
      "step": 7945,
      "train_speed(iter/s)": 1.638226
    },
    {
      "acc": 0.63926039,
      "epoch": 0.20167427701674276,
      "grad_norm": 5.46875,
      "learning_rate": 9.929509330855018e-06,
      "loss": 1.70389633,
      "memory(GiB)": 77.59,
      "step": 7950,
      "train_speed(iter/s)": 1.638328
    },
    {
      "acc": 0.64644952,
      "epoch": 0.20180111618467783,
      "grad_norm": 5.3125,
      "learning_rate": 9.929333761943672e-06,
      "loss": 1.72466908,
      "memory(GiB)": 77.59,
      "step": 7955,
      "train_speed(iter/s)": 1.63843
    },
    {
      "acc": 0.63113956,
      "epoch": 0.20192795535261288,
      "grad_norm": 6.1875,
      "learning_rate": 9.929157976218218e-06,
      "loss": 1.75687828,
      "memory(GiB)": 77.59,
      "step": 7960,
      "train_speed(iter/s)": 1.638536
    },
    {
      "acc": 0.64263639,
      "epoch": 0.20205479452054795,
      "grad_norm": 6.375,
      "learning_rate": 9.928981973686388e-06,
      "loss": 1.72279224,
      "memory(GiB)": 77.59,
      "step": 7965,
      "train_speed(iter/s)": 1.638639
    },
    {
      "acc": 0.65860739,
      "epoch": 0.202181633688483,
      "grad_norm": 4.6875,
      "learning_rate": 9.928805754355926e-06,
      "loss": 1.57535896,
      "memory(GiB)": 77.59,
      "step": 7970,
      "train_speed(iter/s)": 1.638737
    },
    {
      "acc": 0.64154053,
      "epoch": 0.20230847285641806,
      "grad_norm": 6.59375,
      "learning_rate": 9.92862931823458e-06,
      "loss": 1.65813713,
      "memory(GiB)": 77.59,
      "step": 7975,
      "train_speed(iter/s)": 1.638847
    },
    {
      "acc": 0.65051098,
      "epoch": 0.2024353120243531,
      "grad_norm": 5.84375,
      "learning_rate": 9.928452665330113e-06,
      "loss": 1.68977509,
      "memory(GiB)": 77.59,
      "step": 7980,
      "train_speed(iter/s)": 1.638947
    },
    {
      "acc": 0.63657284,
      "epoch": 0.20256215119228818,
      "grad_norm": 5.4375,
      "learning_rate": 9.928275795650293e-06,
      "loss": 1.70972595,
      "memory(GiB)": 77.59,
      "step": 7985,
      "train_speed(iter/s)": 1.639046
    },
    {
      "acc": 0.65277157,
      "epoch": 0.20268899036022323,
      "grad_norm": 5.21875,
      "learning_rate": 9.928098709202901e-06,
      "loss": 1.6477808,
      "memory(GiB)": 77.59,
      "step": 7990,
      "train_speed(iter/s)": 1.639146
    },
    {
      "acc": 0.63990397,
      "epoch": 0.2028158295281583,
      "grad_norm": 7.6875,
      "learning_rate": 9.927921405995727e-06,
      "loss": 1.70589218,
      "memory(GiB)": 77.59,
      "step": 7995,
      "train_speed(iter/s)": 1.639244
    },
    {
      "acc": 0.6625926,
      "epoch": 0.20294266869609334,
      "grad_norm": 5.4375,
      "learning_rate": 9.927743886036566e-06,
      "loss": 1.67445526,
      "memory(GiB)": 77.59,
      "step": 8000,
      "train_speed(iter/s)": 1.639354
    },
    {
      "epoch": 0.20294266869609334,
      "eval_acc": 0.6357697600745108,
      "eval_loss": 1.6331746578216553,
      "eval_runtime": 57.8788,
      "eval_samples_per_second": 110.058,
      "eval_steps_per_second": 27.523,
      "step": 8000
    },
    {
      "acc": 0.65977569,
      "epoch": 0.20306950786402841,
      "grad_norm": 5.1875,
      "learning_rate": 9.927566149333228e-06,
      "loss": 1.58529568,
      "memory(GiB)": 77.59,
      "step": 8005,
      "train_speed(iter/s)": 1.618865
    },
    {
      "acc": 0.63945141,
      "epoch": 0.20319634703196346,
      "grad_norm": 5.78125,
      "learning_rate": 9.92738819589353e-06,
      "loss": 1.66217422,
      "memory(GiB)": 77.59,
      "step": 8010,
      "train_speed(iter/s)": 1.618984
    },
    {
      "acc": 0.6445693,
      "epoch": 0.20332318619989853,
      "grad_norm": 6.09375,
      "learning_rate": 9.927210025725301e-06,
      "loss": 1.64672394,
      "memory(GiB)": 77.59,
      "step": 8015,
      "train_speed(iter/s)": 1.619103
    },
    {
      "acc": 0.64517527,
      "epoch": 0.20345002536783358,
      "grad_norm": 6.34375,
      "learning_rate": 9.927031638836377e-06,
      "loss": 1.68545208,
      "memory(GiB)": 77.59,
      "step": 8020,
      "train_speed(iter/s)": 1.619226
    },
    {
      "acc": 0.64619102,
      "epoch": 0.20357686453576865,
      "grad_norm": 6.53125,
      "learning_rate": 9.926853035234603e-06,
      "loss": 1.67822075,
      "memory(GiB)": 77.59,
      "step": 8025,
      "train_speed(iter/s)": 1.619352
    },
    {
      "acc": 0.65307193,
      "epoch": 0.2037037037037037,
      "grad_norm": 6.25,
      "learning_rate": 9.926674214927836e-06,
      "loss": 1.72326927,
      "memory(GiB)": 77.59,
      "step": 8030,
      "train_speed(iter/s)": 1.619473
    },
    {
      "acc": 0.64413338,
      "epoch": 0.20383054287163876,
      "grad_norm": 4.71875,
      "learning_rate": 9.926495177923941e-06,
      "loss": 1.63698082,
      "memory(GiB)": 77.59,
      "step": 8035,
      "train_speed(iter/s)": 1.61959
    },
    {
      "acc": 0.64417505,
      "epoch": 0.2039573820395738,
      "grad_norm": 6.15625,
      "learning_rate": 9.926315924230794e-06,
      "loss": 1.67357693,
      "memory(GiB)": 77.59,
      "step": 8040,
      "train_speed(iter/s)": 1.619706
    },
    {
      "acc": 0.64143553,
      "epoch": 0.20408422120750888,
      "grad_norm": 5.40625,
      "learning_rate": 9.926136453856277e-06,
      "loss": 1.6808012,
      "memory(GiB)": 77.59,
      "step": 8045,
      "train_speed(iter/s)": 1.61982
    },
    {
      "acc": 0.6483737,
      "epoch": 0.20421106037544393,
      "grad_norm": 5.4375,
      "learning_rate": 9.925956766808286e-06,
      "loss": 1.68206596,
      "memory(GiB)": 77.59,
      "step": 8050,
      "train_speed(iter/s)": 1.61994
    },
    {
      "acc": 0.63125157,
      "epoch": 0.204337899543379,
      "grad_norm": 4.8125,
      "learning_rate": 9.925776863094723e-06,
      "loss": 1.7031517,
      "memory(GiB)": 77.59,
      "step": 8055,
      "train_speed(iter/s)": 1.620065
    },
    {
      "acc": 0.63740573,
      "epoch": 0.20446473871131404,
      "grad_norm": 5.375,
      "learning_rate": 9.9255967427235e-06,
      "loss": 1.71787262,
      "memory(GiB)": 77.59,
      "step": 8060,
      "train_speed(iter/s)": 1.620187
    },
    {
      "acc": 0.64998183,
      "epoch": 0.20459157787924911,
      "grad_norm": 5.46875,
      "learning_rate": 9.925416405702544e-06,
      "loss": 1.65524864,
      "memory(GiB)": 77.59,
      "step": 8065,
      "train_speed(iter/s)": 1.62031
    },
    {
      "acc": 0.64478092,
      "epoch": 0.20471841704718416,
      "grad_norm": 5.34375,
      "learning_rate": 9.925235852039783e-06,
      "loss": 1.69534702,
      "memory(GiB)": 77.59,
      "step": 8070,
      "train_speed(iter/s)": 1.620429
    },
    {
      "acc": 0.63976612,
      "epoch": 0.20484525621511923,
      "grad_norm": 8.5,
      "learning_rate": 9.92505508174316e-06,
      "loss": 1.73726997,
      "memory(GiB)": 77.59,
      "step": 8075,
      "train_speed(iter/s)": 1.620547
    },
    {
      "acc": 0.65227184,
      "epoch": 0.20497209538305428,
      "grad_norm": 8.5,
      "learning_rate": 9.924874094820625e-06,
      "loss": 1.64289227,
      "memory(GiB)": 77.59,
      "step": 8080,
      "train_speed(iter/s)": 1.620659
    },
    {
      "acc": 0.64804726,
      "epoch": 0.20509893455098935,
      "grad_norm": 7.5625,
      "learning_rate": 9.924692891280139e-06,
      "loss": 1.68954716,
      "memory(GiB)": 77.59,
      "step": 8085,
      "train_speed(iter/s)": 1.620774
    },
    {
      "acc": 0.65150881,
      "epoch": 0.2052257737189244,
      "grad_norm": 5.5625,
      "learning_rate": 9.924511471129673e-06,
      "loss": 1.63041382,
      "memory(GiB)": 77.59,
      "step": 8090,
      "train_speed(iter/s)": 1.62089
    },
    {
      "acc": 0.64884038,
      "epoch": 0.20535261288685946,
      "grad_norm": 5.4375,
      "learning_rate": 9.924329834377206e-06,
      "loss": 1.64233952,
      "memory(GiB)": 77.59,
      "step": 8095,
      "train_speed(iter/s)": 1.621001
    },
    {
      "acc": 0.64271021,
      "epoch": 0.2054794520547945,
      "grad_norm": 6.5,
      "learning_rate": 9.924147981030728e-06,
      "loss": 1.65378113,
      "memory(GiB)": 77.59,
      "step": 8100,
      "train_speed(iter/s)": 1.621119
    },
    {
      "acc": 0.66110854,
      "epoch": 0.20560629122272958,
      "grad_norm": 5.1875,
      "learning_rate": 9.923965911098235e-06,
      "loss": 1.61195717,
      "memory(GiB)": 77.59,
      "step": 8105,
      "train_speed(iter/s)": 1.621205
    },
    {
      "acc": 0.63061852,
      "epoch": 0.20573313039066463,
      "grad_norm": 7.15625,
      "learning_rate": 9.92378362458774e-06,
      "loss": 1.75626564,
      "memory(GiB)": 77.59,
      "step": 8110,
      "train_speed(iter/s)": 1.621326
    },
    {
      "acc": 0.65140057,
      "epoch": 0.2058599695585997,
      "grad_norm": 5.1875,
      "learning_rate": 9.923601121507256e-06,
      "loss": 1.67508411,
      "memory(GiB)": 77.59,
      "step": 8115,
      "train_speed(iter/s)": 1.621448
    },
    {
      "acc": 0.63245001,
      "epoch": 0.20598680872653474,
      "grad_norm": 6.34375,
      "learning_rate": 9.923418401864812e-06,
      "loss": 1.69823914,
      "memory(GiB)": 77.59,
      "step": 8120,
      "train_speed(iter/s)": 1.62157
    },
    {
      "acc": 0.64150181,
      "epoch": 0.20611364789446981,
      "grad_norm": 5.875,
      "learning_rate": 9.923235465668447e-06,
      "loss": 1.63199577,
      "memory(GiB)": 77.59,
      "step": 8125,
      "train_speed(iter/s)": 1.621693
    },
    {
      "acc": 0.65600052,
      "epoch": 0.20624048706240486,
      "grad_norm": 5.8125,
      "learning_rate": 9.923052312926204e-06,
      "loss": 1.65171051,
      "memory(GiB)": 77.59,
      "step": 8130,
      "train_speed(iter/s)": 1.621816
    },
    {
      "acc": 0.64646649,
      "epoch": 0.20636732623033993,
      "grad_norm": 5.0625,
      "learning_rate": 9.922868943646142e-06,
      "loss": 1.62688026,
      "memory(GiB)": 77.59,
      "step": 8135,
      "train_speed(iter/s)": 1.621935
    },
    {
      "acc": 0.6379848,
      "epoch": 0.20649416539827498,
      "grad_norm": 7.21875,
      "learning_rate": 9.922685357836324e-06,
      "loss": 1.67142372,
      "memory(GiB)": 77.59,
      "step": 8140,
      "train_speed(iter/s)": 1.622048
    },
    {
      "acc": 0.64262109,
      "epoch": 0.20662100456621005,
      "grad_norm": 5.71875,
      "learning_rate": 9.922501555504827e-06,
      "loss": 1.67993088,
      "memory(GiB)": 77.59,
      "step": 8145,
      "train_speed(iter/s)": 1.622164
    },
    {
      "acc": 0.64655733,
      "epoch": 0.2067478437341451,
      "grad_norm": 5.5625,
      "learning_rate": 9.922317536659733e-06,
      "loss": 1.6251852,
      "memory(GiB)": 77.59,
      "step": 8150,
      "train_speed(iter/s)": 1.622278
    },
    {
      "acc": 0.65152316,
      "epoch": 0.20687468290208016,
      "grad_norm": 5.125,
      "learning_rate": 9.922133301309136e-06,
      "loss": 1.67091179,
      "memory(GiB)": 77.59,
      "step": 8155,
      "train_speed(iter/s)": 1.622401
    },
    {
      "acc": 0.6381959,
      "epoch": 0.2070015220700152,
      "grad_norm": 6.4375,
      "learning_rate": 9.921948849461142e-06,
      "loss": 1.67875366,
      "memory(GiB)": 77.59,
      "step": 8160,
      "train_speed(iter/s)": 1.622511
    },
    {
      "acc": 0.65721121,
      "epoch": 0.20712836123795028,
      "grad_norm": 5.40625,
      "learning_rate": 9.921764181123864e-06,
      "loss": 1.5984417,
      "memory(GiB)": 77.59,
      "step": 8165,
      "train_speed(iter/s)": 1.622626
    },
    {
      "acc": 0.66035037,
      "epoch": 0.20725520040588533,
      "grad_norm": 7.40625,
      "learning_rate": 9.921579296305421e-06,
      "loss": 1.60754204,
      "memory(GiB)": 77.59,
      "step": 8170,
      "train_speed(iter/s)": 1.622728
    },
    {
      "acc": 0.64061174,
      "epoch": 0.2073820395738204,
      "grad_norm": 6.4375,
      "learning_rate": 9.921394195013949e-06,
      "loss": 1.62264938,
      "memory(GiB)": 77.59,
      "step": 8175,
      "train_speed(iter/s)": 1.622846
    },
    {
      "acc": 0.64834986,
      "epoch": 0.20750887874175544,
      "grad_norm": 5.21875,
      "learning_rate": 9.921208877257586e-06,
      "loss": 1.68956242,
      "memory(GiB)": 77.59,
      "step": 8180,
      "train_speed(iter/s)": 1.622959
    },
    {
      "acc": 0.63978705,
      "epoch": 0.20763571790969051,
      "grad_norm": 8.3125,
      "learning_rate": 9.921023343044486e-06,
      "loss": 1.6844799,
      "memory(GiB)": 77.59,
      "step": 8185,
      "train_speed(iter/s)": 1.623068
    },
    {
      "acc": 0.6520762,
      "epoch": 0.20776255707762556,
      "grad_norm": 7.28125,
      "learning_rate": 9.92083759238281e-06,
      "loss": 1.58932476,
      "memory(GiB)": 77.59,
      "step": 8190,
      "train_speed(iter/s)": 1.623191
    },
    {
      "acc": 0.64670382,
      "epoch": 0.20788939624556063,
      "grad_norm": 5.4375,
      "learning_rate": 9.920651625280725e-06,
      "loss": 1.7187542,
      "memory(GiB)": 77.59,
      "step": 8195,
      "train_speed(iter/s)": 1.623307
    },
    {
      "acc": 0.64202294,
      "epoch": 0.20801623541349568,
      "grad_norm": 6.09375,
      "learning_rate": 9.920465441746412e-06,
      "loss": 1.68442764,
      "memory(GiB)": 77.59,
      "step": 8200,
      "train_speed(iter/s)": 1.62343
    },
    {
      "acc": 0.64233842,
      "epoch": 0.20814307458143075,
      "grad_norm": 5.09375,
      "learning_rate": 9.920279041788062e-06,
      "loss": 1.69617329,
      "memory(GiB)": 77.59,
      "step": 8205,
      "train_speed(iter/s)": 1.623548
    },
    {
      "acc": 0.65611525,
      "epoch": 0.2082699137493658,
      "grad_norm": 5.59375,
      "learning_rate": 9.920092425413871e-06,
      "loss": 1.65394211,
      "memory(GiB)": 77.59,
      "step": 8210,
      "train_speed(iter/s)": 1.623658
    },
    {
      "acc": 0.65303197,
      "epoch": 0.20839675291730086,
      "grad_norm": 6.0,
      "learning_rate": 9.919905592632048e-06,
      "loss": 1.56160412,
      "memory(GiB)": 77.59,
      "step": 8215,
      "train_speed(iter/s)": 1.623766
    },
    {
      "acc": 0.63766236,
      "epoch": 0.2085235920852359,
      "grad_norm": 6.28125,
      "learning_rate": 9.919718543450813e-06,
      "loss": 1.63620415,
      "memory(GiB)": 77.59,
      "step": 8220,
      "train_speed(iter/s)": 1.623885
    },
    {
      "acc": 0.64577684,
      "epoch": 0.20865043125317098,
      "grad_norm": 7.34375,
      "learning_rate": 9.919531277878391e-06,
      "loss": 1.67722473,
      "memory(GiB)": 77.59,
      "step": 8225,
      "train_speed(iter/s)": 1.623994
    },
    {
      "acc": 0.64646959,
      "epoch": 0.20877727042110603,
      "grad_norm": 9.4375,
      "learning_rate": 9.91934379592302e-06,
      "loss": 1.57216797,
      "memory(GiB)": 77.59,
      "step": 8230,
      "train_speed(iter/s)": 1.624109
    },
    {
      "acc": 0.65957918,
      "epoch": 0.2089041095890411,
      "grad_norm": 4.9375,
      "learning_rate": 9.919156097592944e-06,
      "loss": 1.55535278,
      "memory(GiB)": 77.59,
      "step": 8235,
      "train_speed(iter/s)": 1.624221
    },
    {
      "acc": 0.65380173,
      "epoch": 0.20903094875697614,
      "grad_norm": 6.0625,
      "learning_rate": 9.91896818289642e-06,
      "loss": 1.60029678,
      "memory(GiB)": 77.59,
      "step": 8240,
      "train_speed(iter/s)": 1.624338
    },
    {
      "acc": 0.65884752,
      "epoch": 0.20915778792491121,
      "grad_norm": 5.34375,
      "learning_rate": 9.918780051841716e-06,
      "loss": 1.65002918,
      "memory(GiB)": 77.59,
      "step": 8245,
      "train_speed(iter/s)": 1.624457
    },
    {
      "acc": 0.64723539,
      "epoch": 0.20928462709284626,
      "grad_norm": 5.21875,
      "learning_rate": 9.918591704437103e-06,
      "loss": 1.65435219,
      "memory(GiB)": 77.59,
      "step": 8250,
      "train_speed(iter/s)": 1.624565
    },
    {
      "acc": 0.64758787,
      "epoch": 0.20941146626078133,
      "grad_norm": 5.53125,
      "learning_rate": 9.918403140690866e-06,
      "loss": 1.6532177,
      "memory(GiB)": 77.59,
      "step": 8255,
      "train_speed(iter/s)": 1.624688
    },
    {
      "acc": 0.64280815,
      "epoch": 0.20953830542871638,
      "grad_norm": 5.125,
      "learning_rate": 9.918214360611302e-06,
      "loss": 1.68372955,
      "memory(GiB)": 77.59,
      "step": 8260,
      "train_speed(iter/s)": 1.624807
    },
    {
      "acc": 0.6420639,
      "epoch": 0.20966514459665145,
      "grad_norm": 7.09375,
      "learning_rate": 9.918025364206712e-06,
      "loss": 1.6575304,
      "memory(GiB)": 77.59,
      "step": 8265,
      "train_speed(iter/s)": 1.624924
    },
    {
      "acc": 0.64520044,
      "epoch": 0.2097919837645865,
      "grad_norm": 4.96875,
      "learning_rate": 9.917836151485407e-06,
      "loss": 1.65246849,
      "memory(GiB)": 77.59,
      "step": 8270,
      "train_speed(iter/s)": 1.62503
    },
    {
      "acc": 0.65880404,
      "epoch": 0.20991882293252156,
      "grad_norm": 5.0,
      "learning_rate": 9.917646722455713e-06,
      "loss": 1.6775856,
      "memory(GiB)": 77.59,
      "step": 8275,
      "train_speed(iter/s)": 1.62515
    },
    {
      "acc": 0.639642,
      "epoch": 0.2100456621004566,
      "grad_norm": 5.9375,
      "learning_rate": 9.91745707712596e-06,
      "loss": 1.62629128,
      "memory(GiB)": 77.59,
      "step": 8280,
      "train_speed(iter/s)": 1.625265
    },
    {
      "acc": 0.64745445,
      "epoch": 0.21017250126839168,
      "grad_norm": 6.0625,
      "learning_rate": 9.91726721550449e-06,
      "loss": 1.66536331,
      "memory(GiB)": 77.59,
      "step": 8285,
      "train_speed(iter/s)": 1.625385
    },
    {
      "acc": 0.63763294,
      "epoch": 0.21029934043632673,
      "grad_norm": 4.625,
      "learning_rate": 9.917077137599653e-06,
      "loss": 1.62223969,
      "memory(GiB)": 77.59,
      "step": 8290,
      "train_speed(iter/s)": 1.625505
    },
    {
      "acc": 0.63691978,
      "epoch": 0.2104261796042618,
      "grad_norm": 4.90625,
      "learning_rate": 9.916886843419811e-06,
      "loss": 1.72010422,
      "memory(GiB)": 77.59,
      "step": 8295,
      "train_speed(iter/s)": 1.625615
    },
    {
      "acc": 0.66086044,
      "epoch": 0.21055301877219684,
      "grad_norm": 5.96875,
      "learning_rate": 9.916696332973334e-06,
      "loss": 1.67197609,
      "memory(GiB)": 77.59,
      "step": 8300,
      "train_speed(iter/s)": 1.62573
    },
    {
      "acc": 0.6437602,
      "epoch": 0.21067985794013191,
      "grad_norm": 5.625,
      "learning_rate": 9.9165056062686e-06,
      "loss": 1.64001007,
      "memory(GiB)": 77.59,
      "step": 8305,
      "train_speed(iter/s)": 1.625846
    },
    {
      "acc": 0.63959589,
      "epoch": 0.21080669710806696,
      "grad_norm": 7.25,
      "learning_rate": 9.916314663314e-06,
      "loss": 1.68658485,
      "memory(GiB)": 77.59,
      "step": 8310,
      "train_speed(iter/s)": 1.625961
    },
    {
      "acc": 0.65336423,
      "epoch": 0.21093353627600203,
      "grad_norm": 5.0625,
      "learning_rate": 9.91612350411793e-06,
      "loss": 1.64935341,
      "memory(GiB)": 77.59,
      "step": 8315,
      "train_speed(iter/s)": 1.626078
    },
    {
      "acc": 0.66239243,
      "epoch": 0.21106037544393708,
      "grad_norm": 6.3125,
      "learning_rate": 9.9159321286888e-06,
      "loss": 1.60674877,
      "memory(GiB)": 77.59,
      "step": 8320,
      "train_speed(iter/s)": 1.626188
    },
    {
      "acc": 0.63801651,
      "epoch": 0.21118721461187215,
      "grad_norm": 5.28125,
      "learning_rate": 9.915740537035026e-06,
      "loss": 1.70842514,
      "memory(GiB)": 77.59,
      "step": 8325,
      "train_speed(iter/s)": 1.6263
    },
    {
      "acc": 0.64135876,
      "epoch": 0.2113140537798072,
      "grad_norm": 6.0625,
      "learning_rate": 9.915548729165036e-06,
      "loss": 1.70464973,
      "memory(GiB)": 77.59,
      "step": 8330,
      "train_speed(iter/s)": 1.62642
    },
    {
      "acc": 0.67257271,
      "epoch": 0.21144089294774226,
      "grad_norm": 5.96875,
      "learning_rate": 9.915356705087269e-06,
      "loss": 1.5961998,
      "memory(GiB)": 77.59,
      "step": 8335,
      "train_speed(iter/s)": 1.62653
    },
    {
      "acc": 0.63645921,
      "epoch": 0.2115677321156773,
      "grad_norm": 5.625,
      "learning_rate": 9.915164464810166e-06,
      "loss": 1.7077179,
      "memory(GiB)": 77.59,
      "step": 8340,
      "train_speed(iter/s)": 1.626642
    },
    {
      "acc": 0.65207982,
      "epoch": 0.21169457128361238,
      "grad_norm": 5.84375,
      "learning_rate": 9.914972008342186e-06,
      "loss": 1.63905373,
      "memory(GiB)": 77.59,
      "step": 8345,
      "train_speed(iter/s)": 1.626755
    },
    {
      "acc": 0.63130016,
      "epoch": 0.21182141045154743,
      "grad_norm": 5.375,
      "learning_rate": 9.914779335691793e-06,
      "loss": 1.72059326,
      "memory(GiB)": 77.59,
      "step": 8350,
      "train_speed(iter/s)": 1.626863
    },
    {
      "acc": 0.63687859,
      "epoch": 0.2119482496194825,
      "grad_norm": 5.0625,
      "learning_rate": 9.914586446867463e-06,
      "loss": 1.71129837,
      "memory(GiB)": 77.59,
      "step": 8355,
      "train_speed(iter/s)": 1.62697
    },
    {
      "acc": 0.65402355,
      "epoch": 0.21207508878741754,
      "grad_norm": 5.46875,
      "learning_rate": 9.914393341877678e-06,
      "loss": 1.67361412,
      "memory(GiB)": 77.59,
      "step": 8360,
      "train_speed(iter/s)": 1.627085
    },
    {
      "acc": 0.6403204,
      "epoch": 0.21220192795535261,
      "grad_norm": 6.28125,
      "learning_rate": 9.914200020730932e-06,
      "loss": 1.68642654,
      "memory(GiB)": 77.59,
      "step": 8365,
      "train_speed(iter/s)": 1.627195
    },
    {
      "acc": 0.65126219,
      "epoch": 0.21232876712328766,
      "grad_norm": 6.4375,
      "learning_rate": 9.914006483435732e-06,
      "loss": 1.63516178,
      "memory(GiB)": 77.59,
      "step": 8370,
      "train_speed(iter/s)": 1.627308
    },
    {
      "acc": 0.63316717,
      "epoch": 0.21245560629122273,
      "grad_norm": 5.40625,
      "learning_rate": 9.913812730000585e-06,
      "loss": 1.7060955,
      "memory(GiB)": 77.59,
      "step": 8375,
      "train_speed(iter/s)": 1.627413
    },
    {
      "acc": 0.65344415,
      "epoch": 0.21258244545915778,
      "grad_norm": 6.5625,
      "learning_rate": 9.913618760434015e-06,
      "loss": 1.66577759,
      "memory(GiB)": 77.59,
      "step": 8380,
      "train_speed(iter/s)": 1.627532
    },
    {
      "acc": 0.65479565,
      "epoch": 0.21270928462709285,
      "grad_norm": 5.78125,
      "learning_rate": 9.913424574744555e-06,
      "loss": 1.58334064,
      "memory(GiB)": 77.59,
      "step": 8385,
      "train_speed(iter/s)": 1.627642
    },
    {
      "acc": 0.65342941,
      "epoch": 0.2128361237950279,
      "grad_norm": 6.34375,
      "learning_rate": 9.913230172940744e-06,
      "loss": 1.64900208,
      "memory(GiB)": 77.59,
      "step": 8390,
      "train_speed(iter/s)": 1.627743
    },
    {
      "acc": 0.63078222,
      "epoch": 0.21296296296296297,
      "grad_norm": 5.3125,
      "learning_rate": 9.913035555031136e-06,
      "loss": 1.69271832,
      "memory(GiB)": 77.59,
      "step": 8395,
      "train_speed(iter/s)": 1.627863
    },
    {
      "acc": 0.6467721,
      "epoch": 0.213089802130898,
      "grad_norm": 5.59375,
      "learning_rate": 9.912840721024288e-06,
      "loss": 1.61202888,
      "memory(GiB)": 77.59,
      "step": 8400,
      "train_speed(iter/s)": 1.627975
    },
    {
      "acc": 0.66350784,
      "epoch": 0.21321664129883308,
      "grad_norm": 5.1875,
      "learning_rate": 9.91264567092877e-06,
      "loss": 1.53365574,
      "memory(GiB)": 77.59,
      "step": 8405,
      "train_speed(iter/s)": 1.62809
    },
    {
      "acc": 0.63582897,
      "epoch": 0.21334348046676813,
      "grad_norm": 6.71875,
      "learning_rate": 9.912450404753164e-06,
      "loss": 1.73253746,
      "memory(GiB)": 77.59,
      "step": 8410,
      "train_speed(iter/s)": 1.628207
    },
    {
      "acc": 0.65780821,
      "epoch": 0.2134703196347032,
      "grad_norm": 6.34375,
      "learning_rate": 9.912254922506057e-06,
      "loss": 1.63765602,
      "memory(GiB)": 77.59,
      "step": 8415,
      "train_speed(iter/s)": 1.628321
    },
    {
      "acc": 0.64422302,
      "epoch": 0.21359715880263824,
      "grad_norm": 5.40625,
      "learning_rate": 9.912059224196044e-06,
      "loss": 1.67829494,
      "memory(GiB)": 77.59,
      "step": 8420,
      "train_speed(iter/s)": 1.62844
    },
    {
      "acc": 0.64829693,
      "epoch": 0.21372399797057332,
      "grad_norm": 4.875,
      "learning_rate": 9.911863309831738e-06,
      "loss": 1.60129166,
      "memory(GiB)": 77.59,
      "step": 8425,
      "train_speed(iter/s)": 1.628549
    },
    {
      "acc": 0.6283473,
      "epoch": 0.21385083713850836,
      "grad_norm": 5.78125,
      "learning_rate": 9.911667179421753e-06,
      "loss": 1.76669254,
      "memory(GiB)": 77.59,
      "step": 8430,
      "train_speed(iter/s)": 1.628662
    },
    {
      "acc": 0.65893593,
      "epoch": 0.21397767630644343,
      "grad_norm": 4.875,
      "learning_rate": 9.911470832974717e-06,
      "loss": 1.60862846,
      "memory(GiB)": 77.59,
      "step": 8435,
      "train_speed(iter/s)": 1.628764
    },
    {
      "acc": 0.64142599,
      "epoch": 0.21410451547437848,
      "grad_norm": 5.21875,
      "learning_rate": 9.911274270499265e-06,
      "loss": 1.61372776,
      "memory(GiB)": 77.59,
      "step": 8440,
      "train_speed(iter/s)": 1.628875
    },
    {
      "acc": 0.65471873,
      "epoch": 0.21423135464231355,
      "grad_norm": 7.25,
      "learning_rate": 9.911077492004044e-06,
      "loss": 1.63499527,
      "memory(GiB)": 77.59,
      "step": 8445,
      "train_speed(iter/s)": 1.628982
    },
    {
      "acc": 0.64031177,
      "epoch": 0.2143581938102486,
      "grad_norm": 6.125,
      "learning_rate": 9.910880497497707e-06,
      "loss": 1.64578476,
      "memory(GiB)": 77.59,
      "step": 8450,
      "train_speed(iter/s)": 1.629098
    },
    {
      "acc": 0.64969273,
      "epoch": 0.21448503297818367,
      "grad_norm": 5.28125,
      "learning_rate": 9.910683286988922e-06,
      "loss": 1.64888611,
      "memory(GiB)": 77.59,
      "step": 8455,
      "train_speed(iter/s)": 1.629208
    },
    {
      "acc": 0.64128742,
      "epoch": 0.2146118721461187,
      "grad_norm": 5.78125,
      "learning_rate": 9.910485860486361e-06,
      "loss": 1.66696815,
      "memory(GiB)": 77.59,
      "step": 8460,
      "train_speed(iter/s)": 1.629319
    },
    {
      "acc": 0.63944397,
      "epoch": 0.21473871131405378,
      "grad_norm": 5.40625,
      "learning_rate": 9.910288217998707e-06,
      "loss": 1.70761452,
      "memory(GiB)": 77.59,
      "step": 8465,
      "train_speed(iter/s)": 1.629434
    },
    {
      "acc": 0.6441371,
      "epoch": 0.21486555048198883,
      "grad_norm": 5.6875,
      "learning_rate": 9.910090359534654e-06,
      "loss": 1.68477287,
      "memory(GiB)": 77.59,
      "step": 8470,
      "train_speed(iter/s)": 1.629546
    },
    {
      "acc": 0.65118241,
      "epoch": 0.2149923896499239,
      "grad_norm": 6.25,
      "learning_rate": 9.909892285102907e-06,
      "loss": 1.66041737,
      "memory(GiB)": 77.59,
      "step": 8475,
      "train_speed(iter/s)": 1.62966
    },
    {
      "acc": 0.65761256,
      "epoch": 0.21511922881785894,
      "grad_norm": 6.09375,
      "learning_rate": 9.909693994712174e-06,
      "loss": 1.61552715,
      "memory(GiB)": 77.59,
      "step": 8480,
      "train_speed(iter/s)": 1.629775
    },
    {
      "acc": 0.66708546,
      "epoch": 0.21524606798579402,
      "grad_norm": 5.5,
      "learning_rate": 9.909495488371181e-06,
      "loss": 1.56115389,
      "memory(GiB)": 77.59,
      "step": 8485,
      "train_speed(iter/s)": 1.629876
    },
    {
      "acc": 0.64924803,
      "epoch": 0.21537290715372906,
      "grad_norm": 6.0,
      "learning_rate": 9.909296766088657e-06,
      "loss": 1.64076824,
      "memory(GiB)": 77.59,
      "step": 8490,
      "train_speed(iter/s)": 1.629989
    },
    {
      "acc": 0.66240573,
      "epoch": 0.21549974632166413,
      "grad_norm": 6.21875,
      "learning_rate": 9.909097827873341e-06,
      "loss": 1.62462959,
      "memory(GiB)": 77.59,
      "step": 8495,
      "train_speed(iter/s)": 1.6301
    },
    {
      "acc": 0.65771203,
      "epoch": 0.21562658548959918,
      "grad_norm": 6.375,
      "learning_rate": 9.908898673733986e-06,
      "loss": 1.68465729,
      "memory(GiB)": 77.59,
      "step": 8500,
      "train_speed(iter/s)": 1.630212
    },
    {
      "acc": 0.65662766,
      "epoch": 0.21575342465753425,
      "grad_norm": 5.59375,
      "learning_rate": 9.90869930367935e-06,
      "loss": 1.62575798,
      "memory(GiB)": 77.59,
      "step": 8505,
      "train_speed(iter/s)": 1.63032
    },
    {
      "acc": 0.63108907,
      "epoch": 0.2158802638254693,
      "grad_norm": 6.46875,
      "learning_rate": 9.908499717718203e-06,
      "loss": 1.73423691,
      "memory(GiB)": 77.59,
      "step": 8510,
      "train_speed(iter/s)": 1.630434
    },
    {
      "acc": 0.65553608,
      "epoch": 0.21600710299340437,
      "grad_norm": 5.84375,
      "learning_rate": 9.908299915859325e-06,
      "loss": 1.66251068,
      "memory(GiB)": 77.59,
      "step": 8515,
      "train_speed(iter/s)": 1.63054
    },
    {
      "acc": 0.65806885,
      "epoch": 0.2161339421613394,
      "grad_norm": 5.25,
      "learning_rate": 9.908099898111502e-06,
      "loss": 1.61349487,
      "memory(GiB)": 77.59,
      "step": 8520,
      "train_speed(iter/s)": 1.630652
    },
    {
      "acc": 0.64546289,
      "epoch": 0.21626078132927448,
      "grad_norm": 5.90625,
      "learning_rate": 9.907899664483533e-06,
      "loss": 1.61516914,
      "memory(GiB)": 77.59,
      "step": 8525,
      "train_speed(iter/s)": 1.63076
    },
    {
      "acc": 0.62854815,
      "epoch": 0.21638762049720953,
      "grad_norm": 5.09375,
      "learning_rate": 9.907699214984223e-06,
      "loss": 1.68960381,
      "memory(GiB)": 77.59,
      "step": 8530,
      "train_speed(iter/s)": 1.63087
    },
    {
      "acc": 0.63813071,
      "epoch": 0.2165144596651446,
      "grad_norm": 5.5,
      "learning_rate": 9.90749854962239e-06,
      "loss": 1.66673012,
      "memory(GiB)": 77.59,
      "step": 8535,
      "train_speed(iter/s)": 1.630976
    },
    {
      "acc": 0.65178452,
      "epoch": 0.21664129883307964,
      "grad_norm": 6.53125,
      "learning_rate": 9.907297668406863e-06,
      "loss": 1.60834808,
      "memory(GiB)": 77.59,
      "step": 8540,
      "train_speed(iter/s)": 1.631087
    },
    {
      "acc": 0.63977785,
      "epoch": 0.21676813800101472,
      "grad_norm": 4.96875,
      "learning_rate": 9.907096571346474e-06,
      "loss": 1.66329422,
      "memory(GiB)": 77.59,
      "step": 8545,
      "train_speed(iter/s)": 1.631205
    },
    {
      "acc": 0.64295607,
      "epoch": 0.21689497716894976,
      "grad_norm": 4.75,
      "learning_rate": 9.906895258450067e-06,
      "loss": 1.6192564,
      "memory(GiB)": 77.59,
      "step": 8550,
      "train_speed(iter/s)": 1.631317
    },
    {
      "acc": 0.65605288,
      "epoch": 0.21702181633688483,
      "grad_norm": 4.875,
      "learning_rate": 9.9066937297265e-06,
      "loss": 1.60834084,
      "memory(GiB)": 77.59,
      "step": 8555,
      "train_speed(iter/s)": 1.631423
    },
    {
      "acc": 0.6422637,
      "epoch": 0.21714865550481988,
      "grad_norm": 5.40625,
      "learning_rate": 9.906491985184637e-06,
      "loss": 1.59745216,
      "memory(GiB)": 77.59,
      "step": 8560,
      "train_speed(iter/s)": 1.631526
    },
    {
      "acc": 0.62723713,
      "epoch": 0.21727549467275495,
      "grad_norm": 6.0,
      "learning_rate": 9.906290024833349e-06,
      "loss": 1.69743805,
      "memory(GiB)": 77.59,
      "step": 8565,
      "train_speed(iter/s)": 1.631633
    },
    {
      "acc": 0.64682379,
      "epoch": 0.21740233384069,
      "grad_norm": 6.5625,
      "learning_rate": 9.906087848681523e-06,
      "loss": 1.64991112,
      "memory(GiB)": 77.59,
      "step": 8570,
      "train_speed(iter/s)": 1.631743
    },
    {
      "acc": 0.64697676,
      "epoch": 0.21752917300862507,
      "grad_norm": 6.03125,
      "learning_rate": 9.905885456738046e-06,
      "loss": 1.66882057,
      "memory(GiB)": 77.59,
      "step": 8575,
      "train_speed(iter/s)": 1.631859
    },
    {
      "acc": 0.65591955,
      "epoch": 0.2176560121765601,
      "grad_norm": 5.90625,
      "learning_rate": 9.905682849011826e-06,
      "loss": 1.65404167,
      "memory(GiB)": 77.59,
      "step": 8580,
      "train_speed(iter/s)": 1.631971
    },
    {
      "acc": 0.64565945,
      "epoch": 0.21778285134449518,
      "grad_norm": 6.15625,
      "learning_rate": 9.905480025511772e-06,
      "loss": 1.70297546,
      "memory(GiB)": 77.59,
      "step": 8585,
      "train_speed(iter/s)": 1.632079
    },
    {
      "acc": 0.63967342,
      "epoch": 0.21790969051243023,
      "grad_norm": 5.40625,
      "learning_rate": 9.905276986246804e-06,
      "loss": 1.67091541,
      "memory(GiB)": 77.59,
      "step": 8590,
      "train_speed(iter/s)": 1.632187
    },
    {
      "acc": 0.64475427,
      "epoch": 0.2180365296803653,
      "grad_norm": 5.1875,
      "learning_rate": 9.905073731225854e-06,
      "loss": 1.5832325,
      "memory(GiB)": 77.59,
      "step": 8595,
      "train_speed(iter/s)": 1.632305
    },
    {
      "acc": 0.62980385,
      "epoch": 0.21816336884830034,
      "grad_norm": 5.9375,
      "learning_rate": 9.904870260457861e-06,
      "loss": 1.66110249,
      "memory(GiB)": 77.59,
      "step": 8600,
      "train_speed(iter/s)": 1.632419
    },
    {
      "acc": 0.6292099,
      "epoch": 0.21829020801623542,
      "grad_norm": 6.5,
      "learning_rate": 9.904666573951777e-06,
      "loss": 1.6999754,
      "memory(GiB)": 77.59,
      "step": 8605,
      "train_speed(iter/s)": 1.632536
    },
    {
      "acc": 0.63838439,
      "epoch": 0.21841704718417046,
      "grad_norm": 5.8125,
      "learning_rate": 9.904462671716559e-06,
      "loss": 1.70130424,
      "memory(GiB)": 77.59,
      "step": 8610,
      "train_speed(iter/s)": 1.632652
    },
    {
      "acc": 0.62956901,
      "epoch": 0.21854388635210553,
      "grad_norm": 5.21875,
      "learning_rate": 9.904258553761175e-06,
      "loss": 1.7441967,
      "memory(GiB)": 77.59,
      "step": 8615,
      "train_speed(iter/s)": 1.63277
    },
    {
      "acc": 0.64285345,
      "epoch": 0.21867072552004058,
      "grad_norm": 4.875,
      "learning_rate": 9.904054220094603e-06,
      "loss": 1.69572258,
      "memory(GiB)": 77.59,
      "step": 8620,
      "train_speed(iter/s)": 1.632882
    },
    {
      "acc": 0.63723063,
      "epoch": 0.21879756468797565,
      "grad_norm": 5.46875,
      "learning_rate": 9.903849670725833e-06,
      "loss": 1.68363056,
      "memory(GiB)": 77.59,
      "step": 8625,
      "train_speed(iter/s)": 1.632995
    },
    {
      "acc": 0.6453342,
      "epoch": 0.2189244038559107,
      "grad_norm": 5.75,
      "learning_rate": 9.903644905663861e-06,
      "loss": 1.63896809,
      "memory(GiB)": 77.59,
      "step": 8630,
      "train_speed(iter/s)": 1.633113
    },
    {
      "acc": 0.64380865,
      "epoch": 0.21905124302384577,
      "grad_norm": 7.1875,
      "learning_rate": 9.90343992491769e-06,
      "loss": 1.64266586,
      "memory(GiB)": 77.59,
      "step": 8635,
      "train_speed(iter/s)": 1.63322
    },
    {
      "acc": 0.6507225,
      "epoch": 0.2191780821917808,
      "grad_norm": 5.25,
      "learning_rate": 9.903234728496341e-06,
      "loss": 1.64144497,
      "memory(GiB)": 77.59,
      "step": 8640,
      "train_speed(iter/s)": 1.633332
    },
    {
      "acc": 0.63628936,
      "epoch": 0.21930492135971588,
      "grad_norm": 6.125,
      "learning_rate": 9.903029316408838e-06,
      "loss": 1.72334366,
      "memory(GiB)": 77.59,
      "step": 8645,
      "train_speed(iter/s)": 1.633444
    },
    {
      "acc": 0.65579467,
      "epoch": 0.21943176052765093,
      "grad_norm": 5.28125,
      "learning_rate": 9.902823688664214e-06,
      "loss": 1.62671928,
      "memory(GiB)": 77.59,
      "step": 8650,
      "train_speed(iter/s)": 1.633561
    },
    {
      "acc": 0.64384308,
      "epoch": 0.219558599695586,
      "grad_norm": 6.1875,
      "learning_rate": 9.902617845271514e-06,
      "loss": 1.63062286,
      "memory(GiB)": 77.59,
      "step": 8655,
      "train_speed(iter/s)": 1.633677
    },
    {
      "acc": 0.66443839,
      "epoch": 0.21968543886352104,
      "grad_norm": 5.03125,
      "learning_rate": 9.902411786239794e-06,
      "loss": 1.58239136,
      "memory(GiB)": 77.59,
      "step": 8660,
      "train_speed(iter/s)": 1.633788
    },
    {
      "acc": 0.63960915,
      "epoch": 0.21981227803145612,
      "grad_norm": 4.84375,
      "learning_rate": 9.902205511578114e-06,
      "loss": 1.66505508,
      "memory(GiB)": 77.59,
      "step": 8665,
      "train_speed(iter/s)": 1.633899
    },
    {
      "acc": 0.64205399,
      "epoch": 0.21993911719939116,
      "grad_norm": 6.75,
      "learning_rate": 9.90199902129555e-06,
      "loss": 1.6725647,
      "memory(GiB)": 77.59,
      "step": 8670,
      "train_speed(iter/s)": 1.634007
    },
    {
      "acc": 0.63029222,
      "epoch": 0.22006595636732623,
      "grad_norm": 5.71875,
      "learning_rate": 9.901792315401184e-06,
      "loss": 1.7220562,
      "memory(GiB)": 77.59,
      "step": 8675,
      "train_speed(iter/s)": 1.634111
    },
    {
      "acc": 0.65128794,
      "epoch": 0.22019279553526128,
      "grad_norm": 5.21875,
      "learning_rate": 9.901585393904104e-06,
      "loss": 1.57617397,
      "memory(GiB)": 77.59,
      "step": 8680,
      "train_speed(iter/s)": 1.634221
    },
    {
      "acc": 0.64128423,
      "epoch": 0.22031963470319635,
      "grad_norm": 5.09375,
      "learning_rate": 9.901378256813418e-06,
      "loss": 1.69571953,
      "memory(GiB)": 77.59,
      "step": 8685,
      "train_speed(iter/s)": 1.634332
    },
    {
      "acc": 0.65720778,
      "epoch": 0.2204464738711314,
      "grad_norm": 5.84375,
      "learning_rate": 9.901170904138232e-06,
      "loss": 1.60900707,
      "memory(GiB)": 77.59,
      "step": 8690,
      "train_speed(iter/s)": 1.634438
    },
    {
      "acc": 0.62904091,
      "epoch": 0.22057331303906647,
      "grad_norm": 5.3125,
      "learning_rate": 9.900963335887667e-06,
      "loss": 1.74302502,
      "memory(GiB)": 77.59,
      "step": 8695,
      "train_speed(iter/s)": 1.634543
    },
    {
      "acc": 0.64808598,
      "epoch": 0.2207001522070015,
      "grad_norm": 6.375,
      "learning_rate": 9.900755552070852e-06,
      "loss": 1.6748003,
      "memory(GiB)": 77.59,
      "step": 8700,
      "train_speed(iter/s)": 1.634646
    },
    {
      "acc": 0.65508118,
      "epoch": 0.22082699137493658,
      "grad_norm": 5.5,
      "learning_rate": 9.900547552696931e-06,
      "loss": 1.64467583,
      "memory(GiB)": 77.59,
      "step": 8705,
      "train_speed(iter/s)": 1.63475
    },
    {
      "acc": 0.65405536,
      "epoch": 0.22095383054287163,
      "grad_norm": 5.125,
      "learning_rate": 9.900339337775046e-06,
      "loss": 1.69861526,
      "memory(GiB)": 77.59,
      "step": 8710,
      "train_speed(iter/s)": 1.63486
    },
    {
      "acc": 0.64679093,
      "epoch": 0.2210806697108067,
      "grad_norm": 6.3125,
      "learning_rate": 9.90013090731436e-06,
      "loss": 1.67696075,
      "memory(GiB)": 77.59,
      "step": 8715,
      "train_speed(iter/s)": 1.634972
    },
    {
      "acc": 0.63485146,
      "epoch": 0.22120750887874174,
      "grad_norm": 6.5625,
      "learning_rate": 9.89992226132404e-06,
      "loss": 1.7033226,
      "memory(GiB)": 77.59,
      "step": 8720,
      "train_speed(iter/s)": 1.635081
    },
    {
      "acc": 0.64462056,
      "epoch": 0.22133434804667682,
      "grad_norm": 4.65625,
      "learning_rate": 9.899713399813261e-06,
      "loss": 1.64769821,
      "memory(GiB)": 77.59,
      "step": 8725,
      "train_speed(iter/s)": 1.635187
    },
    {
      "acc": 0.65201559,
      "epoch": 0.22146118721461186,
      "grad_norm": 5.6875,
      "learning_rate": 9.899504322791212e-06,
      "loss": 1.63786049,
      "memory(GiB)": 77.59,
      "step": 8730,
      "train_speed(iter/s)": 1.635296
    },
    {
      "acc": 0.64604015,
      "epoch": 0.22158802638254693,
      "grad_norm": 5.34375,
      "learning_rate": 9.899295030267086e-06,
      "loss": 1.70348663,
      "memory(GiB)": 77.59,
      "step": 8735,
      "train_speed(iter/s)": 1.635403
    },
    {
      "acc": 0.63902173,
      "epoch": 0.22171486555048198,
      "grad_norm": 5.6875,
      "learning_rate": 9.899085522250094e-06,
      "loss": 1.63704987,
      "memory(GiB)": 77.59,
      "step": 8740,
      "train_speed(iter/s)": 1.635511
    },
    {
      "acc": 0.65556135,
      "epoch": 0.22184170471841705,
      "grad_norm": 5.375,
      "learning_rate": 9.898875798749446e-06,
      "loss": 1.650494,
      "memory(GiB)": 77.59,
      "step": 8745,
      "train_speed(iter/s)": 1.63562
    },
    {
      "acc": 0.6451849,
      "epoch": 0.2219685438863521,
      "grad_norm": 5.65625,
      "learning_rate": 9.898665859774367e-06,
      "loss": 1.66049042,
      "memory(GiB)": 77.59,
      "step": 8750,
      "train_speed(iter/s)": 1.635732
    },
    {
      "acc": 0.64999762,
      "epoch": 0.22209538305428717,
      "grad_norm": 7.8125,
      "learning_rate": 9.898455705334095e-06,
      "loss": 1.6913868,
      "memory(GiB)": 77.59,
      "step": 8755,
      "train_speed(iter/s)": 1.63584
    },
    {
      "acc": 0.64375086,
      "epoch": 0.2222222222222222,
      "grad_norm": 5.15625,
      "learning_rate": 9.89824533543787e-06,
      "loss": 1.66471634,
      "memory(GiB)": 77.59,
      "step": 8760,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.63135786,
      "epoch": 0.22234906139015728,
      "grad_norm": 5.5,
      "learning_rate": 9.898034750094946e-06,
      "loss": 1.70278244,
      "memory(GiB)": 77.59,
      "step": 8765,
      "train_speed(iter/s)": 1.636057
    },
    {
      "acc": 0.64137568,
      "epoch": 0.22247590055809233,
      "grad_norm": 6.15625,
      "learning_rate": 9.897823949314586e-06,
      "loss": 1.71330738,
      "memory(GiB)": 77.59,
      "step": 8770,
      "train_speed(iter/s)": 1.636167
    },
    {
      "acc": 0.64283295,
      "epoch": 0.2226027397260274,
      "grad_norm": 6.09375,
      "learning_rate": 9.897612933106061e-06,
      "loss": 1.71886826,
      "memory(GiB)": 77.59,
      "step": 8775,
      "train_speed(iter/s)": 1.636282
    },
    {
      "acc": 0.64763718,
      "epoch": 0.22272957889396244,
      "grad_norm": 5.46875,
      "learning_rate": 9.897401701478654e-06,
      "loss": 1.62709312,
      "memory(GiB)": 77.59,
      "step": 8780,
      "train_speed(iter/s)": 1.636388
    },
    {
      "acc": 0.64721255,
      "epoch": 0.22285641806189752,
      "grad_norm": 6.375,
      "learning_rate": 9.897190254441653e-06,
      "loss": 1.67772217,
      "memory(GiB)": 77.59,
      "step": 8785,
      "train_speed(iter/s)": 1.636495
    },
    {
      "acc": 0.63037624,
      "epoch": 0.22298325722983256,
      "grad_norm": 5.375,
      "learning_rate": 9.896978592004363e-06,
      "loss": 1.70336437,
      "memory(GiB)": 77.59,
      "step": 8790,
      "train_speed(iter/s)": 1.636599
    },
    {
      "acc": 0.63616052,
      "epoch": 0.22311009639776763,
      "grad_norm": 7.0,
      "learning_rate": 9.896766714176089e-06,
      "loss": 1.7008009,
      "memory(GiB)": 77.59,
      "step": 8795,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.64856014,
      "epoch": 0.22323693556570268,
      "grad_norm": 6.0625,
      "learning_rate": 9.896554620966152e-06,
      "loss": 1.67676983,
      "memory(GiB)": 77.59,
      "step": 8800,
      "train_speed(iter/s)": 1.636819
    },
    {
      "acc": 0.65209675,
      "epoch": 0.22336377473363775,
      "grad_norm": 6.5,
      "learning_rate": 9.896342312383883e-06,
      "loss": 1.64270687,
      "memory(GiB)": 77.59,
      "step": 8805,
      "train_speed(iter/s)": 1.636929
    },
    {
      "acc": 0.65134401,
      "epoch": 0.2234906139015728,
      "grad_norm": 6.65625,
      "learning_rate": 9.896129788438617e-06,
      "loss": 1.65449104,
      "memory(GiB)": 77.59,
      "step": 8810,
      "train_speed(iter/s)": 1.637029
    },
    {
      "acc": 0.65437455,
      "epoch": 0.22361745306950787,
      "grad_norm": 4.96875,
      "learning_rate": 9.895917049139704e-06,
      "loss": 1.62094345,
      "memory(GiB)": 77.59,
      "step": 8815,
      "train_speed(iter/s)": 1.637136
    },
    {
      "acc": 0.6425766,
      "epoch": 0.2237442922374429,
      "grad_norm": 7.0,
      "learning_rate": 9.895704094496502e-06,
      "loss": 1.70247765,
      "memory(GiB)": 77.59,
      "step": 8820,
      "train_speed(iter/s)": 1.637252
    },
    {
      "acc": 0.666113,
      "epoch": 0.22387113140537798,
      "grad_norm": 4.84375,
      "learning_rate": 9.895490924518372e-06,
      "loss": 1.58082705,
      "memory(GiB)": 77.59,
      "step": 8825,
      "train_speed(iter/s)": 1.637358
    },
    {
      "acc": 0.65249109,
      "epoch": 0.22399797057331303,
      "grad_norm": 5.03125,
      "learning_rate": 9.895277539214698e-06,
      "loss": 1.57726479,
      "memory(GiB)": 77.59,
      "step": 8830,
      "train_speed(iter/s)": 1.637458
    },
    {
      "acc": 0.64011312,
      "epoch": 0.2241248097412481,
      "grad_norm": 5.46875,
      "learning_rate": 9.895063938594859e-06,
      "loss": 1.71025105,
      "memory(GiB)": 77.59,
      "step": 8835,
      "train_speed(iter/s)": 1.637557
    },
    {
      "acc": 0.6358242,
      "epoch": 0.22425164890918314,
      "grad_norm": 4.6875,
      "learning_rate": 9.894850122668256e-06,
      "loss": 1.69120293,
      "memory(GiB)": 77.59,
      "step": 8840,
      "train_speed(iter/s)": 1.63766
    },
    {
      "acc": 0.65193229,
      "epoch": 0.22437848807711822,
      "grad_norm": 4.78125,
      "learning_rate": 9.89463609144429e-06,
      "loss": 1.61133499,
      "memory(GiB)": 77.59,
      "step": 8845,
      "train_speed(iter/s)": 1.637764
    },
    {
      "acc": 0.65598588,
      "epoch": 0.22450532724505326,
      "grad_norm": 6.3125,
      "learning_rate": 9.894421844932375e-06,
      "loss": 1.65222321,
      "memory(GiB)": 77.59,
      "step": 8850,
      "train_speed(iter/s)": 1.637875
    },
    {
      "acc": 0.67114768,
      "epoch": 0.22463216641298833,
      "grad_norm": 6.40625,
      "learning_rate": 9.894207383141937e-06,
      "loss": 1.58363075,
      "memory(GiB)": 77.59,
      "step": 8855,
      "train_speed(iter/s)": 1.637981
    },
    {
      "acc": 0.64559302,
      "epoch": 0.22475900558092338,
      "grad_norm": 6.46875,
      "learning_rate": 9.893992706082405e-06,
      "loss": 1.60230484,
      "memory(GiB)": 77.59,
      "step": 8860,
      "train_speed(iter/s)": 1.638089
    },
    {
      "acc": 0.64186382,
      "epoch": 0.22488584474885845,
      "grad_norm": 5.125,
      "learning_rate": 9.893777813763223e-06,
      "loss": 1.65982876,
      "memory(GiB)": 77.59,
      "step": 8865,
      "train_speed(iter/s)": 1.638192
    },
    {
      "acc": 0.64359713,
      "epoch": 0.2250126839167935,
      "grad_norm": 7.6875,
      "learning_rate": 9.893562706193847e-06,
      "loss": 1.74115181,
      "memory(GiB)": 77.59,
      "step": 8870,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.6513104,
      "epoch": 0.22513952308472857,
      "grad_norm": 4.96875,
      "learning_rate": 9.893347383383732e-06,
      "loss": 1.6545517,
      "memory(GiB)": 77.59,
      "step": 8875,
      "train_speed(iter/s)": 1.638421
    },
    {
      "acc": 0.66517544,
      "epoch": 0.2252663622526636,
      "grad_norm": 5.71875,
      "learning_rate": 9.893131845342352e-06,
      "loss": 1.6524456,
      "memory(GiB)": 77.59,
      "step": 8880,
      "train_speed(iter/s)": 1.638526
    },
    {
      "acc": 0.65319042,
      "epoch": 0.22539320142059868,
      "grad_norm": 5.28125,
      "learning_rate": 9.892916092079188e-06,
      "loss": 1.67773113,
      "memory(GiB)": 77.59,
      "step": 8885,
      "train_speed(iter/s)": 1.638626
    },
    {
      "acc": 0.64998131,
      "epoch": 0.22552004058853373,
      "grad_norm": 5.15625,
      "learning_rate": 9.89270012360373e-06,
      "loss": 1.67427521,
      "memory(GiB)": 77.59,
      "step": 8890,
      "train_speed(iter/s)": 1.638731
    },
    {
      "acc": 0.65385656,
      "epoch": 0.2256468797564688,
      "grad_norm": 5.28125,
      "learning_rate": 9.892483939925476e-06,
      "loss": 1.63581619,
      "memory(GiB)": 77.59,
      "step": 8895,
      "train_speed(iter/s)": 1.638832
    },
    {
      "acc": 0.63542638,
      "epoch": 0.22577371892440384,
      "grad_norm": 6.40625,
      "learning_rate": 9.892267541053933e-06,
      "loss": 1.67770729,
      "memory(GiB)": 77.59,
      "step": 8900,
      "train_speed(iter/s)": 1.638932
    },
    {
      "acc": 0.6464417,
      "epoch": 0.22590055809233892,
      "grad_norm": 6.0,
      "learning_rate": 9.892050926998624e-06,
      "loss": 1.6212553,
      "memory(GiB)": 77.59,
      "step": 8905,
      "train_speed(iter/s)": 1.639027
    },
    {
      "acc": 0.64251628,
      "epoch": 0.22602739726027396,
      "grad_norm": 5.1875,
      "learning_rate": 9.891834097769071e-06,
      "loss": 1.71904182,
      "memory(GiB)": 77.59,
      "step": 8910,
      "train_speed(iter/s)": 1.639131
    },
    {
      "acc": 0.63100786,
      "epoch": 0.22615423642820903,
      "grad_norm": 4.96875,
      "learning_rate": 9.891617053374816e-06,
      "loss": 1.74044991,
      "memory(GiB)": 77.59,
      "step": 8915,
      "train_speed(iter/s)": 1.639227
    },
    {
      "acc": 0.64144549,
      "epoch": 0.22628107559614408,
      "grad_norm": 6.1875,
      "learning_rate": 9.891399793825403e-06,
      "loss": 1.69350529,
      "memory(GiB)": 77.59,
      "step": 8920,
      "train_speed(iter/s)": 1.639323
    },
    {
      "acc": 0.64767647,
      "epoch": 0.22640791476407915,
      "grad_norm": 5.25,
      "learning_rate": 9.891182319130387e-06,
      "loss": 1.72190552,
      "memory(GiB)": 77.59,
      "step": 8925,
      "train_speed(iter/s)": 1.639426
    },
    {
      "acc": 0.64755979,
      "epoch": 0.2265347539320142,
      "grad_norm": 5.90625,
      "learning_rate": 9.890964629299336e-06,
      "loss": 1.70096512,
      "memory(GiB)": 77.59,
      "step": 8930,
      "train_speed(iter/s)": 1.639526
    },
    {
      "acc": 0.64155512,
      "epoch": 0.22666159309994927,
      "grad_norm": 5.15625,
      "learning_rate": 9.890746724341825e-06,
      "loss": 1.73209248,
      "memory(GiB)": 77.59,
      "step": 8935,
      "train_speed(iter/s)": 1.639627
    },
    {
      "acc": 0.64422998,
      "epoch": 0.2267884322678843,
      "grad_norm": 5.59375,
      "learning_rate": 9.890528604267436e-06,
      "loss": 1.58922834,
      "memory(GiB)": 77.59,
      "step": 8940,
      "train_speed(iter/s)": 1.639722
    },
    {
      "acc": 0.65478997,
      "epoch": 0.22691527143581938,
      "grad_norm": 6.71875,
      "learning_rate": 9.890310269085765e-06,
      "loss": 1.63509693,
      "memory(GiB)": 77.59,
      "step": 8945,
      "train_speed(iter/s)": 1.639821
    },
    {
      "acc": 0.63246875,
      "epoch": 0.22704211060375443,
      "grad_norm": 5.40625,
      "learning_rate": 9.890091718806414e-06,
      "loss": 1.70619316,
      "memory(GiB)": 77.59,
      "step": 8950,
      "train_speed(iter/s)": 1.639924
    },
    {
      "acc": 0.65057049,
      "epoch": 0.2271689497716895,
      "grad_norm": 6.09375,
      "learning_rate": 9.889872953438996e-06,
      "loss": 1.65867653,
      "memory(GiB)": 77.59,
      "step": 8955,
      "train_speed(iter/s)": 1.640024
    },
    {
      "acc": 0.64201136,
      "epoch": 0.22729578893962454,
      "grad_norm": 8.25,
      "learning_rate": 9.889653972993136e-06,
      "loss": 1.68201637,
      "memory(GiB)": 77.59,
      "step": 8960,
      "train_speed(iter/s)": 1.640134
    },
    {
      "acc": 0.64295058,
      "epoch": 0.22742262810755962,
      "grad_norm": 5.21875,
      "learning_rate": 9.889434777478464e-06,
      "loss": 1.67074699,
      "memory(GiB)": 77.59,
      "step": 8965,
      "train_speed(iter/s)": 1.640239
    },
    {
      "acc": 0.65191102,
      "epoch": 0.22754946727549466,
      "grad_norm": 5.3125,
      "learning_rate": 9.88921536690462e-06,
      "loss": 1.68216286,
      "memory(GiB)": 77.59,
      "step": 8970,
      "train_speed(iter/s)": 1.640342
    },
    {
      "acc": 0.63716831,
      "epoch": 0.22767630644342973,
      "grad_norm": 5.21875,
      "learning_rate": 9.888995741281252e-06,
      "loss": 1.73848877,
      "memory(GiB)": 77.59,
      "step": 8975,
      "train_speed(iter/s)": 1.640437
    },
    {
      "acc": 0.64836941,
      "epoch": 0.22780314561136478,
      "grad_norm": 5.8125,
      "learning_rate": 9.888775900618028e-06,
      "loss": 1.605476,
      "memory(GiB)": 77.59,
      "step": 8980,
      "train_speed(iter/s)": 1.640539
    },
    {
      "acc": 0.644102,
      "epoch": 0.22792998477929985,
      "grad_norm": 6.0,
      "learning_rate": 9.88855584492461e-06,
      "loss": 1.67533607,
      "memory(GiB)": 77.59,
      "step": 8985,
      "train_speed(iter/s)": 1.640647
    },
    {
      "acc": 0.64010086,
      "epoch": 0.2280568239472349,
      "grad_norm": 6.71875,
      "learning_rate": 9.888335574210681e-06,
      "loss": 1.68984547,
      "memory(GiB)": 77.59,
      "step": 8990,
      "train_speed(iter/s)": 1.640745
    },
    {
      "acc": 0.64795132,
      "epoch": 0.22818366311516997,
      "grad_norm": 6.34375,
      "learning_rate": 9.888115088485931e-06,
      "loss": 1.55505962,
      "memory(GiB)": 77.59,
      "step": 8995,
      "train_speed(iter/s)": 1.640843
    },
    {
      "acc": 0.6501296,
      "epoch": 0.228310502283105,
      "grad_norm": 5.40625,
      "learning_rate": 9.887894387760053e-06,
      "loss": 1.64787827,
      "memory(GiB)": 77.59,
      "step": 9000,
      "train_speed(iter/s)": 1.640949
    },
    {
      "epoch": 0.228310502283105,
      "eval_acc": 0.6373861090054025,
      "eval_loss": 1.6237483024597168,
      "eval_runtime": 58.6039,
      "eval_samples_per_second": 108.696,
      "eval_steps_per_second": 27.182,
      "step": 9000
    },
    {
      "acc": 0.64845643,
      "epoch": 0.22843734145104008,
      "grad_norm": 7.375,
      "learning_rate": 9.887673472042757e-06,
      "loss": 1.68617744,
      "memory(GiB)": 77.59,
      "step": 9005,
      "train_speed(iter/s)": 1.622451
    },
    {
      "acc": 0.65458899,
      "epoch": 0.22856418061897513,
      "grad_norm": 5.71875,
      "learning_rate": 9.88745234134376e-06,
      "loss": 1.66633224,
      "memory(GiB)": 77.59,
      "step": 9010,
      "train_speed(iter/s)": 1.622546
    },
    {
      "acc": 0.63907366,
      "epoch": 0.2286910197869102,
      "grad_norm": 5.21875,
      "learning_rate": 9.887230995672789e-06,
      "loss": 1.66835785,
      "memory(GiB)": 77.59,
      "step": 9015,
      "train_speed(iter/s)": 1.622638
    },
    {
      "acc": 0.64539042,
      "epoch": 0.22881785895484524,
      "grad_norm": 6.09375,
      "learning_rate": 9.887009435039578e-06,
      "loss": 1.59948788,
      "memory(GiB)": 77.59,
      "step": 9020,
      "train_speed(iter/s)": 1.622714
    },
    {
      "acc": 0.63725467,
      "epoch": 0.22894469812278032,
      "grad_norm": 5.96875,
      "learning_rate": 9.886787659453873e-06,
      "loss": 1.65513573,
      "memory(GiB)": 77.59,
      "step": 9025,
      "train_speed(iter/s)": 1.622807
    },
    {
      "acc": 0.64495363,
      "epoch": 0.22907153729071536,
      "grad_norm": 5.46875,
      "learning_rate": 9.886565668925429e-06,
      "loss": 1.64530029,
      "memory(GiB)": 77.59,
      "step": 9030,
      "train_speed(iter/s)": 1.622904
    },
    {
      "acc": 0.64777966,
      "epoch": 0.22919837645865043,
      "grad_norm": 10.0625,
      "learning_rate": 9.88634346346401e-06,
      "loss": 1.64854698,
      "memory(GiB)": 77.59,
      "step": 9035,
      "train_speed(iter/s)": 1.623001
    },
    {
      "acc": 0.64281402,
      "epoch": 0.22932521562658548,
      "grad_norm": 5.21875,
      "learning_rate": 9.88612104307939e-06,
      "loss": 1.61526184,
      "memory(GiB)": 77.59,
      "step": 9040,
      "train_speed(iter/s)": 1.623108
    },
    {
      "acc": 0.66017203,
      "epoch": 0.22945205479452055,
      "grad_norm": 5.21875,
      "learning_rate": 9.885898407781352e-06,
      "loss": 1.6112999,
      "memory(GiB)": 77.59,
      "step": 9045,
      "train_speed(iter/s)": 1.623206
    },
    {
      "acc": 0.65367327,
      "epoch": 0.2295788939624556,
      "grad_norm": 5.75,
      "learning_rate": 9.885675557579686e-06,
      "loss": 1.72248211,
      "memory(GiB)": 77.59,
      "step": 9050,
      "train_speed(iter/s)": 1.62331
    },
    {
      "acc": 0.65201654,
      "epoch": 0.22970573313039067,
      "grad_norm": 6.25,
      "learning_rate": 9.885452492484198e-06,
      "loss": 1.57434149,
      "memory(GiB)": 77.59,
      "step": 9055,
      "train_speed(iter/s)": 1.623417
    },
    {
      "acc": 0.64312749,
      "epoch": 0.2298325722983257,
      "grad_norm": 5.8125,
      "learning_rate": 9.885229212504697e-06,
      "loss": 1.66974411,
      "memory(GiB)": 77.59,
      "step": 9060,
      "train_speed(iter/s)": 1.623519
    },
    {
      "acc": 0.64915266,
      "epoch": 0.22995941146626078,
      "grad_norm": 4.75,
      "learning_rate": 9.885005717651002e-06,
      "loss": 1.62329941,
      "memory(GiB)": 77.59,
      "step": 9065,
      "train_speed(iter/s)": 1.623617
    },
    {
      "acc": 0.65288315,
      "epoch": 0.23008625063419583,
      "grad_norm": 6.125,
      "learning_rate": 9.88478200793295e-06,
      "loss": 1.63091946,
      "memory(GiB)": 77.59,
      "step": 9070,
      "train_speed(iter/s)": 1.623699
    },
    {
      "acc": 0.64600859,
      "epoch": 0.2302130898021309,
      "grad_norm": 5.59375,
      "learning_rate": 9.884558083360372e-06,
      "loss": 1.62570229,
      "memory(GiB)": 77.59,
      "step": 9075,
      "train_speed(iter/s)": 1.6238
    },
    {
      "acc": 0.6642405,
      "epoch": 0.23033992897006594,
      "grad_norm": 5.28125,
      "learning_rate": 9.884333943943123e-06,
      "loss": 1.55125132,
      "memory(GiB)": 77.59,
      "step": 9080,
      "train_speed(iter/s)": 1.623893
    },
    {
      "acc": 0.64701395,
      "epoch": 0.23046676813800102,
      "grad_norm": 5.34375,
      "learning_rate": 9.884109589691062e-06,
      "loss": 1.66710243,
      "memory(GiB)": 77.59,
      "step": 9085,
      "train_speed(iter/s)": 1.623996
    },
    {
      "acc": 0.64406352,
      "epoch": 0.23059360730593606,
      "grad_norm": 5.3125,
      "learning_rate": 9.883885020614052e-06,
      "loss": 1.68455009,
      "memory(GiB)": 77.59,
      "step": 9090,
      "train_speed(iter/s)": 1.624103
    },
    {
      "acc": 0.66169415,
      "epoch": 0.23072044647387113,
      "grad_norm": 4.9375,
      "learning_rate": 9.883660236721977e-06,
      "loss": 1.5562604,
      "memory(GiB)": 77.59,
      "step": 9095,
      "train_speed(iter/s)": 1.624204
    },
    {
      "acc": 0.6408246,
      "epoch": 0.23084728564180618,
      "grad_norm": 5.28125,
      "learning_rate": 9.883435238024718e-06,
      "loss": 1.67281418,
      "memory(GiB)": 77.59,
      "step": 9100,
      "train_speed(iter/s)": 1.624303
    },
    {
      "acc": 0.66595106,
      "epoch": 0.23097412480974125,
      "grad_norm": 5.34375,
      "learning_rate": 9.883210024532176e-06,
      "loss": 1.57440062,
      "memory(GiB)": 77.59,
      "step": 9105,
      "train_speed(iter/s)": 1.624397
    },
    {
      "acc": 0.65892649,
      "epoch": 0.2311009639776763,
      "grad_norm": 5.75,
      "learning_rate": 9.882984596254255e-06,
      "loss": 1.65532417,
      "memory(GiB)": 77.59,
      "step": 9110,
      "train_speed(iter/s)": 1.6245
    },
    {
      "acc": 0.64969063,
      "epoch": 0.23122780314561137,
      "grad_norm": 5.625,
      "learning_rate": 9.88275895320087e-06,
      "loss": 1.68483906,
      "memory(GiB)": 77.59,
      "step": 9115,
      "train_speed(iter/s)": 1.624599
    },
    {
      "acc": 0.64919991,
      "epoch": 0.2313546423135464,
      "grad_norm": 6.03125,
      "learning_rate": 9.882533095381947e-06,
      "loss": 1.67741852,
      "memory(GiB)": 77.59,
      "step": 9120,
      "train_speed(iter/s)": 1.624699
    },
    {
      "acc": 0.63494534,
      "epoch": 0.23148148148148148,
      "grad_norm": 7.625,
      "learning_rate": 9.882307022807419e-06,
      "loss": 1.64587288,
      "memory(GiB)": 77.59,
      "step": 9125,
      "train_speed(iter/s)": 1.624799
    },
    {
      "acc": 0.6446043,
      "epoch": 0.23160832064941653,
      "grad_norm": 4.90625,
      "learning_rate": 9.88208073548723e-06,
      "loss": 1.73853683,
      "memory(GiB)": 77.59,
      "step": 9130,
      "train_speed(iter/s)": 1.624897
    },
    {
      "acc": 0.64634643,
      "epoch": 0.2317351598173516,
      "grad_norm": 5.28125,
      "learning_rate": 9.881854233431333e-06,
      "loss": 1.68168411,
      "memory(GiB)": 77.59,
      "step": 9135,
      "train_speed(iter/s)": 1.624992
    },
    {
      "acc": 0.6489604,
      "epoch": 0.23186199898528664,
      "grad_norm": 4.96875,
      "learning_rate": 9.881627516649692e-06,
      "loss": 1.65658894,
      "memory(GiB)": 77.59,
      "step": 9140,
      "train_speed(iter/s)": 1.625089
    },
    {
      "acc": 0.63461447,
      "epoch": 0.23198883815322172,
      "grad_norm": 4.9375,
      "learning_rate": 9.881400585152278e-06,
      "loss": 1.73371582,
      "memory(GiB)": 77.59,
      "step": 9145,
      "train_speed(iter/s)": 1.62519
    },
    {
      "acc": 0.66043954,
      "epoch": 0.23211567732115676,
      "grad_norm": 4.84375,
      "learning_rate": 9.881173438949072e-06,
      "loss": 1.67182083,
      "memory(GiB)": 77.59,
      "step": 9150,
      "train_speed(iter/s)": 1.625293
    },
    {
      "acc": 0.65812225,
      "epoch": 0.23224251648909183,
      "grad_norm": 5.0,
      "learning_rate": 9.880946078050064e-06,
      "loss": 1.62039642,
      "memory(GiB)": 77.59,
      "step": 9155,
      "train_speed(iter/s)": 1.625395
    },
    {
      "acc": 0.63304405,
      "epoch": 0.23236935565702688,
      "grad_norm": 5.40625,
      "learning_rate": 9.880718502465258e-06,
      "loss": 1.70877666,
      "memory(GiB)": 77.59,
      "step": 9160,
      "train_speed(iter/s)": 1.625498
    },
    {
      "acc": 0.6330555,
      "epoch": 0.23249619482496195,
      "grad_norm": 5.125,
      "learning_rate": 9.88049071220466e-06,
      "loss": 1.71098824,
      "memory(GiB)": 77.59,
      "step": 9165,
      "train_speed(iter/s)": 1.625597
    },
    {
      "acc": 0.64827271,
      "epoch": 0.232623033992897,
      "grad_norm": 5.3125,
      "learning_rate": 9.88026270727829e-06,
      "loss": 1.67315445,
      "memory(GiB)": 77.59,
      "step": 9170,
      "train_speed(iter/s)": 1.625693
    },
    {
      "acc": 0.64434195,
      "epoch": 0.23274987316083207,
      "grad_norm": 5.03125,
      "learning_rate": 9.880034487696179e-06,
      "loss": 1.64116821,
      "memory(GiB)": 77.59,
      "step": 9175,
      "train_speed(iter/s)": 1.625793
    },
    {
      "acc": 0.64183378,
      "epoch": 0.2328767123287671,
      "grad_norm": 9.3125,
      "learning_rate": 9.879806053468361e-06,
      "loss": 1.68195076,
      "memory(GiB)": 77.59,
      "step": 9180,
      "train_speed(iter/s)": 1.625894
    },
    {
      "acc": 0.64677091,
      "epoch": 0.23300355149670218,
      "grad_norm": 5.15625,
      "learning_rate": 9.879577404604889e-06,
      "loss": 1.65706902,
      "memory(GiB)": 77.59,
      "step": 9185,
      "train_speed(iter/s)": 1.625998
    },
    {
      "acc": 0.63322067,
      "epoch": 0.23313039066463723,
      "grad_norm": 5.1875,
      "learning_rate": 9.879348541115816e-06,
      "loss": 1.67011566,
      "memory(GiB)": 77.59,
      "step": 9190,
      "train_speed(iter/s)": 1.626102
    },
    {
      "acc": 0.64351492,
      "epoch": 0.2332572298325723,
      "grad_norm": 4.71875,
      "learning_rate": 9.879119463011208e-06,
      "loss": 1.65603466,
      "memory(GiB)": 77.59,
      "step": 9195,
      "train_speed(iter/s)": 1.626206
    },
    {
      "acc": 0.65794177,
      "epoch": 0.23338406900050734,
      "grad_norm": 5.875,
      "learning_rate": 9.878890170301143e-06,
      "loss": 1.61584663,
      "memory(GiB)": 77.59,
      "step": 9200,
      "train_speed(iter/s)": 1.62631
    },
    {
      "acc": 0.66397114,
      "epoch": 0.23351090816844242,
      "grad_norm": 4.71875,
      "learning_rate": 9.878660662995706e-06,
      "loss": 1.60551891,
      "memory(GiB)": 77.59,
      "step": 9205,
      "train_speed(iter/s)": 1.626408
    },
    {
      "acc": 0.64669518,
      "epoch": 0.23363774733637746,
      "grad_norm": 6.9375,
      "learning_rate": 9.878430941104991e-06,
      "loss": 1.67409706,
      "memory(GiB)": 77.59,
      "step": 9210,
      "train_speed(iter/s)": 1.626516
    },
    {
      "acc": 0.64231701,
      "epoch": 0.23376458650431253,
      "grad_norm": 6.1875,
      "learning_rate": 9.878201004639104e-06,
      "loss": 1.73436699,
      "memory(GiB)": 77.59,
      "step": 9215,
      "train_speed(iter/s)": 1.626623
    },
    {
      "acc": 0.6697144,
      "epoch": 0.23389142567224758,
      "grad_norm": 5.65625,
      "learning_rate": 9.877970853608156e-06,
      "loss": 1.61370983,
      "memory(GiB)": 77.59,
      "step": 9220,
      "train_speed(iter/s)": 1.626735
    },
    {
      "acc": 0.64825511,
      "epoch": 0.23401826484018265,
      "grad_norm": 6.5,
      "learning_rate": 9.87774048802227e-06,
      "loss": 1.66075726,
      "memory(GiB)": 77.59,
      "step": 9225,
      "train_speed(iter/s)": 1.626832
    },
    {
      "acc": 0.6384212,
      "epoch": 0.2341451040081177,
      "grad_norm": 5.71875,
      "learning_rate": 9.877509907891583e-06,
      "loss": 1.57854786,
      "memory(GiB)": 77.59,
      "step": 9230,
      "train_speed(iter/s)": 1.626941
    },
    {
      "acc": 0.65710521,
      "epoch": 0.23427194317605277,
      "grad_norm": 5.65625,
      "learning_rate": 9.877279113226232e-06,
      "loss": 1.56658764,
      "memory(GiB)": 77.59,
      "step": 9235,
      "train_speed(iter/s)": 1.627041
    },
    {
      "acc": 0.65100546,
      "epoch": 0.2343987823439878,
      "grad_norm": 5.34375,
      "learning_rate": 9.87704810403637e-06,
      "loss": 1.61526814,
      "memory(GiB)": 77.59,
      "step": 9240,
      "train_speed(iter/s)": 1.627141
    },
    {
      "acc": 0.64556274,
      "epoch": 0.23452562151192288,
      "grad_norm": 5.40625,
      "learning_rate": 9.876816880332157e-06,
      "loss": 1.68021069,
      "memory(GiB)": 77.59,
      "step": 9245,
      "train_speed(iter/s)": 1.627245
    },
    {
      "acc": 0.65085897,
      "epoch": 0.23465246067985793,
      "grad_norm": 4.25,
      "learning_rate": 9.876585442123765e-06,
      "loss": 1.64631863,
      "memory(GiB)": 77.59,
      "step": 9250,
      "train_speed(iter/s)": 1.627347
    },
    {
      "acc": 0.64963069,
      "epoch": 0.234779299847793,
      "grad_norm": 5.96875,
      "learning_rate": 9.876353789421373e-06,
      "loss": 1.67465096,
      "memory(GiB)": 77.59,
      "step": 9255,
      "train_speed(iter/s)": 1.627448
    },
    {
      "acc": 0.63810825,
      "epoch": 0.23490613901572804,
      "grad_norm": 7.3125,
      "learning_rate": 9.876121922235171e-06,
      "loss": 1.72027836,
      "memory(GiB)": 77.59,
      "step": 9260,
      "train_speed(iter/s)": 1.627557
    },
    {
      "acc": 0.64873571,
      "epoch": 0.23503297818366312,
      "grad_norm": 5.3125,
      "learning_rate": 9.875889840575356e-06,
      "loss": 1.71790314,
      "memory(GiB)": 77.59,
      "step": 9265,
      "train_speed(iter/s)": 1.627658
    },
    {
      "acc": 0.64158063,
      "epoch": 0.23515981735159816,
      "grad_norm": 4.8125,
      "learning_rate": 9.875657544452135e-06,
      "loss": 1.65621204,
      "memory(GiB)": 77.59,
      "step": 9270,
      "train_speed(iter/s)": 1.627759
    },
    {
      "acc": 0.63539329,
      "epoch": 0.23528665651953323,
      "grad_norm": 5.78125,
      "learning_rate": 9.875425033875728e-06,
      "loss": 1.67092934,
      "memory(GiB)": 77.59,
      "step": 9275,
      "train_speed(iter/s)": 1.627861
    },
    {
      "acc": 0.66319242,
      "epoch": 0.23541349568746828,
      "grad_norm": 7.21875,
      "learning_rate": 9.875192308856363e-06,
      "loss": 1.65409985,
      "memory(GiB)": 77.59,
      "step": 9280,
      "train_speed(iter/s)": 1.627961
    },
    {
      "acc": 0.66097026,
      "epoch": 0.23554033485540335,
      "grad_norm": 6.03125,
      "learning_rate": 9.87495936940427e-06,
      "loss": 1.6012291,
      "memory(GiB)": 77.59,
      "step": 9285,
      "train_speed(iter/s)": 1.628068
    },
    {
      "acc": 0.63480844,
      "epoch": 0.2356671740233384,
      "grad_norm": 5.59375,
      "learning_rate": 9.874726215529702e-06,
      "loss": 1.70720577,
      "memory(GiB)": 77.59,
      "step": 9290,
      "train_speed(iter/s)": 1.628177
    },
    {
      "acc": 0.64903154,
      "epoch": 0.23579401319127347,
      "grad_norm": 6.25,
      "learning_rate": 9.87449284724291e-06,
      "loss": 1.61668167,
      "memory(GiB)": 77.59,
      "step": 9295,
      "train_speed(iter/s)": 1.628279
    },
    {
      "acc": 0.65418267,
      "epoch": 0.2359208523592085,
      "grad_norm": 5.59375,
      "learning_rate": 9.874259264554159e-06,
      "loss": 1.68040714,
      "memory(GiB)": 77.59,
      "step": 9300,
      "train_speed(iter/s)": 1.628383
    },
    {
      "acc": 0.66054163,
      "epoch": 0.23604769152714358,
      "grad_norm": 5.21875,
      "learning_rate": 9.874025467473722e-06,
      "loss": 1.5883213,
      "memory(GiB)": 77.59,
      "step": 9305,
      "train_speed(iter/s)": 1.628484
    },
    {
      "acc": 0.62926149,
      "epoch": 0.23617453069507863,
      "grad_norm": 5.8125,
      "learning_rate": 9.873791456011887e-06,
      "loss": 1.70828228,
      "memory(GiB)": 77.59,
      "step": 9310,
      "train_speed(iter/s)": 1.628584
    },
    {
      "acc": 0.65225201,
      "epoch": 0.2363013698630137,
      "grad_norm": 6.25,
      "learning_rate": 9.873557230178942e-06,
      "loss": 1.64824638,
      "memory(GiB)": 77.59,
      "step": 9315,
      "train_speed(iter/s)": 1.628683
    },
    {
      "acc": 0.64674082,
      "epoch": 0.23642820903094874,
      "grad_norm": 5.375,
      "learning_rate": 9.873322789985191e-06,
      "loss": 1.62088299,
      "memory(GiB)": 77.59,
      "step": 9320,
      "train_speed(iter/s)": 1.628784
    },
    {
      "acc": 0.63966265,
      "epoch": 0.23655504819888382,
      "grad_norm": 6.0625,
      "learning_rate": 9.873088135440949e-06,
      "loss": 1.69843178,
      "memory(GiB)": 77.59,
      "step": 9325,
      "train_speed(iter/s)": 1.628886
    },
    {
      "acc": 0.6216156,
      "epoch": 0.23668188736681886,
      "grad_norm": 5.5625,
      "learning_rate": 9.87285326655653e-06,
      "loss": 1.73059177,
      "memory(GiB)": 77.59,
      "step": 9330,
      "train_speed(iter/s)": 1.628988
    },
    {
      "acc": 0.64705725,
      "epoch": 0.23680872653475393,
      "grad_norm": 5.28125,
      "learning_rate": 9.87261818334227e-06,
      "loss": 1.62914925,
      "memory(GiB)": 77.59,
      "step": 9335,
      "train_speed(iter/s)": 1.62909
    },
    {
      "acc": 0.62429399,
      "epoch": 0.23693556570268898,
      "grad_norm": 5.59375,
      "learning_rate": 9.87238288580851e-06,
      "loss": 1.7366457,
      "memory(GiB)": 77.59,
      "step": 9340,
      "train_speed(iter/s)": 1.629192
    },
    {
      "acc": 0.64672909,
      "epoch": 0.23706240487062405,
      "grad_norm": 5.21875,
      "learning_rate": 9.872147373965594e-06,
      "loss": 1.65063057,
      "memory(GiB)": 77.59,
      "step": 9345,
      "train_speed(iter/s)": 1.629295
    },
    {
      "acc": 0.6331954,
      "epoch": 0.2371892440385591,
      "grad_norm": 8.3125,
      "learning_rate": 9.871911647823884e-06,
      "loss": 1.77910805,
      "memory(GiB)": 77.59,
      "step": 9350,
      "train_speed(iter/s)": 1.629405
    },
    {
      "acc": 0.65076671,
      "epoch": 0.23731608320649417,
      "grad_norm": 5.0,
      "learning_rate": 9.871675707393749e-06,
      "loss": 1.64002972,
      "memory(GiB)": 77.59,
      "step": 9355,
      "train_speed(iter/s)": 1.629505
    },
    {
      "acc": 0.6542901,
      "epoch": 0.2374429223744292,
      "grad_norm": 5.84375,
      "learning_rate": 9.871439552685566e-06,
      "loss": 1.66089725,
      "memory(GiB)": 77.59,
      "step": 9360,
      "train_speed(iter/s)": 1.629606
    },
    {
      "acc": 0.64613533,
      "epoch": 0.23756976154236428,
      "grad_norm": 5.03125,
      "learning_rate": 9.871203183709723e-06,
      "loss": 1.66794357,
      "memory(GiB)": 77.59,
      "step": 9365,
      "train_speed(iter/s)": 1.629705
    },
    {
      "acc": 0.65607357,
      "epoch": 0.23769660071029933,
      "grad_norm": 6.59375,
      "learning_rate": 9.870966600476614e-06,
      "loss": 1.6301157,
      "memory(GiB)": 77.59,
      "step": 9370,
      "train_speed(iter/s)": 1.629808
    },
    {
      "acc": 0.66108446,
      "epoch": 0.2378234398782344,
      "grad_norm": 6.71875,
      "learning_rate": 9.870729802996647e-06,
      "loss": 1.64777222,
      "memory(GiB)": 77.59,
      "step": 9375,
      "train_speed(iter/s)": 1.629912
    },
    {
      "acc": 0.6409246,
      "epoch": 0.23795027904616944,
      "grad_norm": 6.5,
      "learning_rate": 9.870492791280239e-06,
      "loss": 1.63623047,
      "memory(GiB)": 77.59,
      "step": 9380,
      "train_speed(iter/s)": 1.630012
    },
    {
      "acc": 0.64289203,
      "epoch": 0.23807711821410452,
      "grad_norm": 9.4375,
      "learning_rate": 9.87025556533781e-06,
      "loss": 1.63623619,
      "memory(GiB)": 77.59,
      "step": 9385,
      "train_speed(iter/s)": 1.630111
    },
    {
      "acc": 0.672299,
      "epoch": 0.23820395738203956,
      "grad_norm": 6.5625,
      "learning_rate": 9.870018125179799e-06,
      "loss": 1.53563261,
      "memory(GiB)": 77.59,
      "step": 9390,
      "train_speed(iter/s)": 1.630213
    },
    {
      "acc": 0.66684465,
      "epoch": 0.23833079654997463,
      "grad_norm": 6.96875,
      "learning_rate": 9.869780470816647e-06,
      "loss": 1.53617125,
      "memory(GiB)": 77.59,
      "step": 9395,
      "train_speed(iter/s)": 1.630318
    },
    {
      "acc": 0.66304359,
      "epoch": 0.23845763571790968,
      "grad_norm": 4.59375,
      "learning_rate": 9.869542602258809e-06,
      "loss": 1.57127762,
      "memory(GiB)": 77.59,
      "step": 9400,
      "train_speed(iter/s)": 1.630418
    },
    {
      "acc": 0.64068289,
      "epoch": 0.23858447488584475,
      "grad_norm": 7.375,
      "learning_rate": 9.869304519516745e-06,
      "loss": 1.75879402,
      "memory(GiB)": 77.59,
      "step": 9405,
      "train_speed(iter/s)": 1.630517
    },
    {
      "acc": 0.65158987,
      "epoch": 0.2387113140537798,
      "grad_norm": 5.15625,
      "learning_rate": 9.869066222600928e-06,
      "loss": 1.68611336,
      "memory(GiB)": 77.59,
      "step": 9410,
      "train_speed(iter/s)": 1.630614
    },
    {
      "acc": 0.64545851,
      "epoch": 0.23883815322171487,
      "grad_norm": 4.8125,
      "learning_rate": 9.86882771152184e-06,
      "loss": 1.65368767,
      "memory(GiB)": 77.59,
      "step": 9415,
      "train_speed(iter/s)": 1.630707
    },
    {
      "acc": 0.63275785,
      "epoch": 0.2389649923896499,
      "grad_norm": 5.28125,
      "learning_rate": 9.868588986289973e-06,
      "loss": 1.79201832,
      "memory(GiB)": 77.59,
      "step": 9420,
      "train_speed(iter/s)": 1.630804
    },
    {
      "acc": 0.63314781,
      "epoch": 0.23909183155758498,
      "grad_norm": 5.40625,
      "learning_rate": 9.868350046915825e-06,
      "loss": 1.73071785,
      "memory(GiB)": 77.59,
      "step": 9425,
      "train_speed(iter/s)": 1.630902
    },
    {
      "acc": 0.63926659,
      "epoch": 0.23921867072552003,
      "grad_norm": 6.15625,
      "learning_rate": 9.868110893409906e-06,
      "loss": 1.72271461,
      "memory(GiB)": 77.59,
      "step": 9430,
      "train_speed(iter/s)": 1.631011
    },
    {
      "acc": 0.67679691,
      "epoch": 0.2393455098934551,
      "grad_norm": 6.8125,
      "learning_rate": 9.867871525782735e-06,
      "loss": 1.58200169,
      "memory(GiB)": 77.59,
      "step": 9435,
      "train_speed(iter/s)": 1.631114
    },
    {
      "acc": 0.65555964,
      "epoch": 0.23947234906139014,
      "grad_norm": 5.125,
      "learning_rate": 9.86763194404484e-06,
      "loss": 1.65161896,
      "memory(GiB)": 77.59,
      "step": 9440,
      "train_speed(iter/s)": 1.631221
    },
    {
      "acc": 0.63352251,
      "epoch": 0.23959918822932522,
      "grad_norm": 5.59375,
      "learning_rate": 9.867392148206762e-06,
      "loss": 1.70929241,
      "memory(GiB)": 77.59,
      "step": 9445,
      "train_speed(iter/s)": 1.63132
    },
    {
      "acc": 0.64765925,
      "epoch": 0.23972602739726026,
      "grad_norm": 6.375,
      "learning_rate": 9.867152138279043e-06,
      "loss": 1.66059628,
      "memory(GiB)": 77.59,
      "step": 9450,
      "train_speed(iter/s)": 1.631419
    },
    {
      "acc": 0.64454942,
      "epoch": 0.23985286656519533,
      "grad_norm": 4.84375,
      "learning_rate": 9.866911914272246e-06,
      "loss": 1.63672409,
      "memory(GiB)": 77.59,
      "step": 9455,
      "train_speed(iter/s)": 1.631519
    },
    {
      "acc": 0.65048313,
      "epoch": 0.23997970573313038,
      "grad_norm": 5.28125,
      "learning_rate": 9.866671476196931e-06,
      "loss": 1.64810219,
      "memory(GiB)": 77.59,
      "step": 9460,
      "train_speed(iter/s)": 1.631628
    },
    {
      "acc": 0.66130323,
      "epoch": 0.24010654490106545,
      "grad_norm": 5.0,
      "learning_rate": 9.866430824063678e-06,
      "loss": 1.60275383,
      "memory(GiB)": 77.59,
      "step": 9465,
      "train_speed(iter/s)": 1.631727
    },
    {
      "acc": 0.65562611,
      "epoch": 0.2402333840690005,
      "grad_norm": 6.46875,
      "learning_rate": 9.86618995788307e-06,
      "loss": 1.60493183,
      "memory(GiB)": 87.33,
      "step": 9470,
      "train_speed(iter/s)": 1.631826
    },
    {
      "acc": 0.62938862,
      "epoch": 0.24036022323693557,
      "grad_norm": 6.1875,
      "learning_rate": 9.865948877665702e-06,
      "loss": 1.6870388,
      "memory(GiB)": 87.33,
      "step": 9475,
      "train_speed(iter/s)": 1.631931
    },
    {
      "acc": 0.64937038,
      "epoch": 0.2404870624048706,
      "grad_norm": 6.71875,
      "learning_rate": 9.865707583422178e-06,
      "loss": 1.69814262,
      "memory(GiB)": 87.33,
      "step": 9480,
      "train_speed(iter/s)": 1.632035
    },
    {
      "acc": 0.64474349,
      "epoch": 0.24061390157280568,
      "grad_norm": 5.5625,
      "learning_rate": 9.865466075163108e-06,
      "loss": 1.66721516,
      "memory(GiB)": 87.33,
      "step": 9485,
      "train_speed(iter/s)": 1.632134
    },
    {
      "acc": 0.65736027,
      "epoch": 0.24074074074074073,
      "grad_norm": 5.0,
      "learning_rate": 9.86522435289912e-06,
      "loss": 1.60760689,
      "memory(GiB)": 87.33,
      "step": 9490,
      "train_speed(iter/s)": 1.632235
    },
    {
      "acc": 0.64089384,
      "epoch": 0.2408675799086758,
      "grad_norm": 5.40625,
      "learning_rate": 9.864982416640843e-06,
      "loss": 1.72783279,
      "memory(GiB)": 87.33,
      "step": 9495,
      "train_speed(iter/s)": 1.632331
    },
    {
      "acc": 0.64851279,
      "epoch": 0.24099441907661084,
      "grad_norm": 6.5625,
      "learning_rate": 9.864740266398918e-06,
      "loss": 1.66511307,
      "memory(GiB)": 87.33,
      "step": 9500,
      "train_speed(iter/s)": 1.632435
    },
    {
      "acc": 0.65543299,
      "epoch": 0.24112125824454592,
      "grad_norm": 5.5625,
      "learning_rate": 9.864497902183996e-06,
      "loss": 1.59703064,
      "memory(GiB)": 87.33,
      "step": 9505,
      "train_speed(iter/s)": 1.632534
    },
    {
      "acc": 0.65599937,
      "epoch": 0.24124809741248096,
      "grad_norm": 6.75,
      "learning_rate": 9.864255324006738e-06,
      "loss": 1.62961349,
      "memory(GiB)": 87.33,
      "step": 9510,
      "train_speed(iter/s)": 1.632635
    },
    {
      "acc": 0.63429327,
      "epoch": 0.24137493658041603,
      "grad_norm": 5.96875,
      "learning_rate": 9.864012531877814e-06,
      "loss": 1.73694382,
      "memory(GiB)": 87.33,
      "step": 9515,
      "train_speed(iter/s)": 1.632737
    },
    {
      "acc": 0.63781829,
      "epoch": 0.24150177574835108,
      "grad_norm": 5.25,
      "learning_rate": 9.863769525807903e-06,
      "loss": 1.72291622,
      "memory(GiB)": 87.33,
      "step": 9520,
      "train_speed(iter/s)": 1.632841
    },
    {
      "acc": 0.66068192,
      "epoch": 0.24162861491628615,
      "grad_norm": 6.0625,
      "learning_rate": 9.863526305807694e-06,
      "loss": 1.56554413,
      "memory(GiB)": 87.33,
      "step": 9525,
      "train_speed(iter/s)": 1.632947
    },
    {
      "acc": 0.64753299,
      "epoch": 0.2417554540842212,
      "grad_norm": 8.1875,
      "learning_rate": 9.863282871887882e-06,
      "loss": 1.73369293,
      "memory(GiB)": 87.33,
      "step": 9530,
      "train_speed(iter/s)": 1.633048
    },
    {
      "acc": 0.65124426,
      "epoch": 0.24188229325215627,
      "grad_norm": 5.1875,
      "learning_rate": 9.863039224059177e-06,
      "loss": 1.64500675,
      "memory(GiB)": 87.33,
      "step": 9535,
      "train_speed(iter/s)": 1.633144
    },
    {
      "acc": 0.65118494,
      "epoch": 0.2420091324200913,
      "grad_norm": 4.90625,
      "learning_rate": 9.862795362332293e-06,
      "loss": 1.60982132,
      "memory(GiB)": 87.33,
      "step": 9540,
      "train_speed(iter/s)": 1.633235
    },
    {
      "acc": 0.66644897,
      "epoch": 0.24213597158802638,
      "grad_norm": 5.59375,
      "learning_rate": 9.862551286717961e-06,
      "loss": 1.59954548,
      "memory(GiB)": 87.33,
      "step": 9545,
      "train_speed(iter/s)": 1.63333
    },
    {
      "acc": 0.65461111,
      "epoch": 0.24226281075596143,
      "grad_norm": 8.1875,
      "learning_rate": 9.862306997226914e-06,
      "loss": 1.62875366,
      "memory(GiB)": 87.33,
      "step": 9550,
      "train_speed(iter/s)": 1.633421
    },
    {
      "acc": 0.64794388,
      "epoch": 0.2423896499238965,
      "grad_norm": 6.65625,
      "learning_rate": 9.862062493869895e-06,
      "loss": 1.68707466,
      "memory(GiB)": 87.33,
      "step": 9555,
      "train_speed(iter/s)": 1.633522
    },
    {
      "acc": 0.64705133,
      "epoch": 0.24251648909183154,
      "grad_norm": 5.90625,
      "learning_rate": 9.861817776657661e-06,
      "loss": 1.60952816,
      "memory(GiB)": 87.33,
      "step": 9560,
      "train_speed(iter/s)": 1.633619
    },
    {
      "acc": 0.62973495,
      "epoch": 0.24264332825976662,
      "grad_norm": 7.40625,
      "learning_rate": 9.861572845600973e-06,
      "loss": 1.68779659,
      "memory(GiB)": 87.33,
      "step": 9565,
      "train_speed(iter/s)": 1.633721
    },
    {
      "acc": 0.64386315,
      "epoch": 0.24277016742770166,
      "grad_norm": 4.84375,
      "learning_rate": 9.861327700710608e-06,
      "loss": 1.61891365,
      "memory(GiB)": 87.33,
      "step": 9570,
      "train_speed(iter/s)": 1.633817
    },
    {
      "acc": 0.64163079,
      "epoch": 0.24289700659563673,
      "grad_norm": 5.59375,
      "learning_rate": 9.861082341997345e-06,
      "loss": 1.70166588,
      "memory(GiB)": 87.33,
      "step": 9575,
      "train_speed(iter/s)": 1.633921
    },
    {
      "acc": 0.65208187,
      "epoch": 0.24302384576357178,
      "grad_norm": 4.75,
      "learning_rate": 9.860836769471977e-06,
      "loss": 1.63839931,
      "memory(GiB)": 87.33,
      "step": 9580,
      "train_speed(iter/s)": 1.634017
    },
    {
      "acc": 0.63927927,
      "epoch": 0.24315068493150685,
      "grad_norm": 5.3125,
      "learning_rate": 9.860590983145307e-06,
      "loss": 1.72652149,
      "memory(GiB)": 87.33,
      "step": 9585,
      "train_speed(iter/s)": 1.634115
    },
    {
      "acc": 0.65039825,
      "epoch": 0.2432775240994419,
      "grad_norm": 6.34375,
      "learning_rate": 9.860344983028146e-06,
      "loss": 1.68072662,
      "memory(GiB)": 87.33,
      "step": 9590,
      "train_speed(iter/s)": 1.634214
    },
    {
      "acc": 0.64354124,
      "epoch": 0.24340436326737697,
      "grad_norm": 5.21875,
      "learning_rate": 9.86009876913131e-06,
      "loss": 1.68386497,
      "memory(GiB)": 87.33,
      "step": 9595,
      "train_speed(iter/s)": 1.634308
    },
    {
      "acc": 0.64121289,
      "epoch": 0.243531202435312,
      "grad_norm": 4.1875,
      "learning_rate": 9.859852341465633e-06,
      "loss": 1.70940952,
      "memory(GiB)": 87.33,
      "step": 9600,
      "train_speed(iter/s)": 1.634407
    },
    {
      "acc": 0.64821434,
      "epoch": 0.24365804160324708,
      "grad_norm": 5.28125,
      "learning_rate": 9.859605700041951e-06,
      "loss": 1.67825699,
      "memory(GiB)": 87.33,
      "step": 9605,
      "train_speed(iter/s)": 1.634502
    },
    {
      "acc": 0.66226764,
      "epoch": 0.24378488077118213,
      "grad_norm": 5.90625,
      "learning_rate": 9.859358844871113e-06,
      "loss": 1.58269835,
      "memory(GiB)": 87.33,
      "step": 9610,
      "train_speed(iter/s)": 1.634599
    },
    {
      "acc": 0.65524387,
      "epoch": 0.2439117199391172,
      "grad_norm": 5.40625,
      "learning_rate": 9.859111775963981e-06,
      "loss": 1.65098114,
      "memory(GiB)": 87.33,
      "step": 9615,
      "train_speed(iter/s)": 1.634698
    },
    {
      "acc": 0.66100526,
      "epoch": 0.24403855910705224,
      "grad_norm": 5.28125,
      "learning_rate": 9.858864493331417e-06,
      "loss": 1.61300449,
      "memory(GiB)": 87.33,
      "step": 9620,
      "train_speed(iter/s)": 1.634794
    },
    {
      "acc": 0.63697333,
      "epoch": 0.24416539827498732,
      "grad_norm": 7.875,
      "learning_rate": 9.858616996984297e-06,
      "loss": 1.72463856,
      "memory(GiB)": 87.33,
      "step": 9625,
      "train_speed(iter/s)": 1.634896
    },
    {
      "acc": 0.63661032,
      "epoch": 0.24429223744292236,
      "grad_norm": 5.90625,
      "learning_rate": 9.858369286933513e-06,
      "loss": 1.68524361,
      "memory(GiB)": 87.33,
      "step": 9630,
      "train_speed(iter/s)": 1.634988
    },
    {
      "acc": 0.65131092,
      "epoch": 0.24441907661085743,
      "grad_norm": 5.59375,
      "learning_rate": 9.858121363189954e-06,
      "loss": 1.66075993,
      "memory(GiB)": 87.33,
      "step": 9635,
      "train_speed(iter/s)": 1.635081
    },
    {
      "acc": 0.64018021,
      "epoch": 0.24454591577879248,
      "grad_norm": 5.90625,
      "learning_rate": 9.85787322576453e-06,
      "loss": 1.73212318,
      "memory(GiB)": 87.33,
      "step": 9640,
      "train_speed(iter/s)": 1.635185
    },
    {
      "acc": 0.65091796,
      "epoch": 0.24467275494672755,
      "grad_norm": 4.9375,
      "learning_rate": 9.85762487466815e-06,
      "loss": 1.65248299,
      "memory(GiB)": 87.33,
      "step": 9645,
      "train_speed(iter/s)": 1.635277
    },
    {
      "acc": 0.66229463,
      "epoch": 0.2447995941146626,
      "grad_norm": 6.0625,
      "learning_rate": 9.857376309911741e-06,
      "loss": 1.53893776,
      "memory(GiB)": 87.33,
      "step": 9650,
      "train_speed(iter/s)": 1.635371
    },
    {
      "acc": 0.64031124,
      "epoch": 0.24492643328259767,
      "grad_norm": 4.90625,
      "learning_rate": 9.857127531506237e-06,
      "loss": 1.72968102,
      "memory(GiB)": 87.33,
      "step": 9655,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.64228048,
      "epoch": 0.2450532724505327,
      "grad_norm": 5.9375,
      "learning_rate": 9.856878539462577e-06,
      "loss": 1.66113605,
      "memory(GiB)": 87.33,
      "step": 9660,
      "train_speed(iter/s)": 1.635554
    },
    {
      "acc": 0.66556864,
      "epoch": 0.24518011161846778,
      "grad_norm": 5.5,
      "learning_rate": 9.856629333791716e-06,
      "loss": 1.58413582,
      "memory(GiB)": 87.33,
      "step": 9665,
      "train_speed(iter/s)": 1.635645
    },
    {
      "acc": 0.64543605,
      "epoch": 0.24530695078640283,
      "grad_norm": 5.125,
      "learning_rate": 9.856379914504612e-06,
      "loss": 1.68589516,
      "memory(GiB)": 87.33,
      "step": 9670,
      "train_speed(iter/s)": 1.63574
    },
    {
      "acc": 0.65295658,
      "epoch": 0.2454337899543379,
      "grad_norm": 4.8125,
      "learning_rate": 9.856130281612237e-06,
      "loss": 1.69418297,
      "memory(GiB)": 87.33,
      "step": 9675,
      "train_speed(iter/s)": 1.635839
    },
    {
      "acc": 0.63720455,
      "epoch": 0.24556062912227294,
      "grad_norm": 5.25,
      "learning_rate": 9.855880435125572e-06,
      "loss": 1.61891174,
      "memory(GiB)": 87.33,
      "step": 9680,
      "train_speed(iter/s)": 1.635939
    },
    {
      "acc": 0.64129586,
      "epoch": 0.24568746829020802,
      "grad_norm": 5.90625,
      "learning_rate": 9.855630375055604e-06,
      "loss": 1.63581123,
      "memory(GiB)": 87.33,
      "step": 9685,
      "train_speed(iter/s)": 1.636032
    },
    {
      "acc": 0.64883432,
      "epoch": 0.24581430745814306,
      "grad_norm": 5.28125,
      "learning_rate": 9.855380101413336e-06,
      "loss": 1.65804214,
      "memory(GiB)": 87.33,
      "step": 9690,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.64615369,
      "epoch": 0.24594114662607813,
      "grad_norm": 5.4375,
      "learning_rate": 9.855129614209771e-06,
      "loss": 1.6511528,
      "memory(GiB)": 87.33,
      "step": 9695,
      "train_speed(iter/s)": 1.636212
    },
    {
      "acc": 0.64521065,
      "epoch": 0.24606798579401318,
      "grad_norm": 6.125,
      "learning_rate": 9.85487891345593e-06,
      "loss": 1.68964539,
      "memory(GiB)": 87.33,
      "step": 9700,
      "train_speed(iter/s)": 1.636309
    },
    {
      "acc": 0.63514032,
      "epoch": 0.24619482496194825,
      "grad_norm": 7.03125,
      "learning_rate": 9.85462799916284e-06,
      "loss": 1.70887966,
      "memory(GiB)": 87.33,
      "step": 9705,
      "train_speed(iter/s)": 1.636403
    },
    {
      "acc": 0.63588052,
      "epoch": 0.2463216641298833,
      "grad_norm": 5.625,
      "learning_rate": 9.854376871341535e-06,
      "loss": 1.70032406,
      "memory(GiB)": 87.33,
      "step": 9710,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.63490038,
      "epoch": 0.24644850329781837,
      "grad_norm": 6.875,
      "learning_rate": 9.854125530003063e-06,
      "loss": 1.70630455,
      "memory(GiB)": 87.33,
      "step": 9715,
      "train_speed(iter/s)": 1.63659
    },
    {
      "acc": 0.65200148,
      "epoch": 0.2465753424657534,
      "grad_norm": 6.34375,
      "learning_rate": 9.853873975158476e-06,
      "loss": 1.60985165,
      "memory(GiB)": 87.33,
      "step": 9720,
      "train_speed(iter/s)": 1.636691
    },
    {
      "acc": 0.63832979,
      "epoch": 0.24670218163368848,
      "grad_norm": 5.65625,
      "learning_rate": 9.853622206818842e-06,
      "loss": 1.72486095,
      "memory(GiB)": 87.33,
      "step": 9725,
      "train_speed(iter/s)": 1.636782
    },
    {
      "acc": 0.64781246,
      "epoch": 0.24682902080162353,
      "grad_norm": 4.875,
      "learning_rate": 9.853370224995233e-06,
      "loss": 1.65095253,
      "memory(GiB)": 87.33,
      "step": 9730,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.66477175,
      "epoch": 0.2469558599695586,
      "grad_norm": 5.46875,
      "learning_rate": 9.853118029698733e-06,
      "loss": 1.58915806,
      "memory(GiB)": 87.33,
      "step": 9735,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.6595706,
      "epoch": 0.24708269913749364,
      "grad_norm": 5.96875,
      "learning_rate": 9.852865620940436e-06,
      "loss": 1.66983242,
      "memory(GiB)": 87.33,
      "step": 9740,
      "train_speed(iter/s)": 1.637039
    },
    {
      "acc": 0.6453341,
      "epoch": 0.24720953830542872,
      "grad_norm": 4.8125,
      "learning_rate": 9.85261299873144e-06,
      "loss": 1.65589428,
      "memory(GiB)": 87.33,
      "step": 9745,
      "train_speed(iter/s)": 1.637135
    },
    {
      "acc": 0.6633359,
      "epoch": 0.24733637747336376,
      "grad_norm": 5.25,
      "learning_rate": 9.85236016308286e-06,
      "loss": 1.61111488,
      "memory(GiB)": 87.33,
      "step": 9750,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.6529355,
      "epoch": 0.24746321664129883,
      "grad_norm": 5.4375,
      "learning_rate": 9.852107114005816e-06,
      "loss": 1.66232491,
      "memory(GiB)": 87.33,
      "step": 9755,
      "train_speed(iter/s)": 1.637313
    },
    {
      "acc": 0.66931586,
      "epoch": 0.24759005580923388,
      "grad_norm": 5.34375,
      "learning_rate": 9.851853851511437e-06,
      "loss": 1.60879478,
      "memory(GiB)": 87.33,
      "step": 9760,
      "train_speed(iter/s)": 1.637408
    },
    {
      "acc": 0.64560871,
      "epoch": 0.24771689497716895,
      "grad_norm": 6.34375,
      "learning_rate": 9.851600375610864e-06,
      "loss": 1.62980614,
      "memory(GiB)": 87.33,
      "step": 9765,
      "train_speed(iter/s)": 1.6375
    },
    {
      "acc": 0.64487877,
      "epoch": 0.247843734145104,
      "grad_norm": 6.125,
      "learning_rate": 9.851346686315246e-06,
      "loss": 1.67241478,
      "memory(GiB)": 87.33,
      "step": 9770,
      "train_speed(iter/s)": 1.637592
    },
    {
      "acc": 0.64298697,
      "epoch": 0.24797057331303907,
      "grad_norm": 5.0,
      "learning_rate": 9.851092783635742e-06,
      "loss": 1.72505569,
      "memory(GiB)": 87.33,
      "step": 9775,
      "train_speed(iter/s)": 1.637682
    },
    {
      "acc": 0.64806018,
      "epoch": 0.2480974124809741,
      "grad_norm": 6.28125,
      "learning_rate": 9.850838667583518e-06,
      "loss": 1.62771339,
      "memory(GiB)": 87.33,
      "step": 9780,
      "train_speed(iter/s)": 1.637775
    },
    {
      "acc": 0.65341506,
      "epoch": 0.24822425164890918,
      "grad_norm": 5.59375,
      "learning_rate": 9.850584338169752e-06,
      "loss": 1.61090584,
      "memory(GiB)": 87.33,
      "step": 9785,
      "train_speed(iter/s)": 1.637866
    },
    {
      "acc": 0.64384122,
      "epoch": 0.24835109081684423,
      "grad_norm": 5.8125,
      "learning_rate": 9.85032979540563e-06,
      "loss": 1.66910725,
      "memory(GiB)": 87.33,
      "step": 9790,
      "train_speed(iter/s)": 1.637957
    },
    {
      "acc": 0.63692422,
      "epoch": 0.2484779299847793,
      "grad_norm": 5.1875,
      "learning_rate": 9.85007503930235e-06,
      "loss": 1.69653721,
      "memory(GiB)": 87.33,
      "step": 9795,
      "train_speed(iter/s)": 1.63805
    },
    {
      "acc": 0.65492153,
      "epoch": 0.24860476915271434,
      "grad_norm": 5.34375,
      "learning_rate": 9.849820069871114e-06,
      "loss": 1.6327713,
      "memory(GiB)": 87.33,
      "step": 9800,
      "train_speed(iter/s)": 1.638138
    },
    {
      "acc": 0.64292245,
      "epoch": 0.24873160832064942,
      "grad_norm": 6.625,
      "learning_rate": 9.849564887123138e-06,
      "loss": 1.67443733,
      "memory(GiB)": 87.33,
      "step": 9805,
      "train_speed(iter/s)": 1.638226
    },
    {
      "acc": 0.63440418,
      "epoch": 0.24885844748858446,
      "grad_norm": 5.21875,
      "learning_rate": 9.849309491069647e-06,
      "loss": 1.674436,
      "memory(GiB)": 87.33,
      "step": 9810,
      "train_speed(iter/s)": 1.638317
    },
    {
      "acc": 0.64905891,
      "epoch": 0.24898528665651953,
      "grad_norm": 7.15625,
      "learning_rate": 9.849053881721876e-06,
      "loss": 1.59621582,
      "memory(GiB)": 87.33,
      "step": 9815,
      "train_speed(iter/s)": 1.638409
    },
    {
      "acc": 0.65379725,
      "epoch": 0.24911212582445458,
      "grad_norm": 5.75,
      "learning_rate": 9.848798059091064e-06,
      "loss": 1.69068108,
      "memory(GiB)": 87.33,
      "step": 9820,
      "train_speed(iter/s)": 1.638499
    },
    {
      "acc": 0.64928665,
      "epoch": 0.24923896499238965,
      "grad_norm": 5.90625,
      "learning_rate": 9.848542023188466e-06,
      "loss": 1.69615974,
      "memory(GiB)": 87.33,
      "step": 9825,
      "train_speed(iter/s)": 1.63858
    },
    {
      "acc": 0.66876736,
      "epoch": 0.2493658041603247,
      "grad_norm": 6.34375,
      "learning_rate": 9.848285774025342e-06,
      "loss": 1.52942619,
      "memory(GiB)": 87.33,
      "step": 9830,
      "train_speed(iter/s)": 1.638666
    },
    {
      "acc": 0.64703846,
      "epoch": 0.24949264332825977,
      "grad_norm": 8.125,
      "learning_rate": 9.848029311612963e-06,
      "loss": 1.64497204,
      "memory(GiB)": 87.33,
      "step": 9835,
      "train_speed(iter/s)": 1.638753
    },
    {
      "acc": 0.65347133,
      "epoch": 0.2496194824961948,
      "grad_norm": 6.9375,
      "learning_rate": 9.84777263596261e-06,
      "loss": 1.6713623,
      "memory(GiB)": 87.33,
      "step": 9840,
      "train_speed(iter/s)": 1.638848
    },
    {
      "acc": 0.643645,
      "epoch": 0.24974632166412988,
      "grad_norm": 5.21875,
      "learning_rate": 9.847515747085573e-06,
      "loss": 1.6845932,
      "memory(GiB)": 87.33,
      "step": 9845,
      "train_speed(iter/s)": 1.638938
    },
    {
      "acc": 0.64831305,
      "epoch": 0.24987316083206493,
      "grad_norm": 4.90625,
      "learning_rate": 9.847258644993151e-06,
      "loss": 1.67672653,
      "memory(GiB)": 87.33,
      "step": 9850,
      "train_speed(iter/s)": 1.639026
    },
    {
      "acc": 0.63253489,
      "epoch": 0.25,
      "grad_norm": 4.875,
      "learning_rate": 9.847001329696653e-06,
      "loss": 1.71245689,
      "memory(GiB)": 87.33,
      "step": 9855,
      "train_speed(iter/s)": 1.639113
    },
    {
      "acc": 0.64879866,
      "epoch": 0.25012683916793504,
      "grad_norm": 5.125,
      "learning_rate": 9.846743801207395e-06,
      "loss": 1.64888535,
      "memory(GiB)": 87.33,
      "step": 9860,
      "train_speed(iter/s)": 1.6392
    },
    {
      "acc": 0.62758493,
      "epoch": 0.25025367833587014,
      "grad_norm": 5.65625,
      "learning_rate": 9.846486059536706e-06,
      "loss": 1.70128193,
      "memory(GiB)": 97.12,
      "step": 9865,
      "train_speed(iter/s)": 1.639281
    },
    {
      "acc": 0.64468579,
      "epoch": 0.2503805175038052,
      "grad_norm": 5.0,
      "learning_rate": 9.846228104695922e-06,
      "loss": 1.6634449,
      "memory(GiB)": 97.12,
      "step": 9870,
      "train_speed(iter/s)": 1.639375
    },
    {
      "acc": 0.63298979,
      "epoch": 0.25050735667174023,
      "grad_norm": 5.71875,
      "learning_rate": 9.84596993669639e-06,
      "loss": 1.76489449,
      "memory(GiB)": 97.12,
      "step": 9875,
      "train_speed(iter/s)": 1.639467
    },
    {
      "acc": 0.65223188,
      "epoch": 0.2506341958396753,
      "grad_norm": 11.8125,
      "learning_rate": 9.845711555549464e-06,
      "loss": 1.61567135,
      "memory(GiB)": 97.12,
      "step": 9880,
      "train_speed(iter/s)": 1.639558
    },
    {
      "acc": 0.647159,
      "epoch": 0.2507610350076104,
      "grad_norm": 5.21875,
      "learning_rate": 9.845452961266509e-06,
      "loss": 1.63127213,
      "memory(GiB)": 97.12,
      "step": 9885,
      "train_speed(iter/s)": 1.639645
    },
    {
      "acc": 0.65376697,
      "epoch": 0.2508878741755454,
      "grad_norm": 6.0,
      "learning_rate": 9.845194153858899e-06,
      "loss": 1.63588867,
      "memory(GiB)": 97.12,
      "step": 9890,
      "train_speed(iter/s)": 1.639732
    },
    {
      "acc": 0.64734588,
      "epoch": 0.25101471334348047,
      "grad_norm": 5.0,
      "learning_rate": 9.844935133338018e-06,
      "loss": 1.59693937,
      "memory(GiB)": 97.12,
      "step": 9895,
      "train_speed(iter/s)": 1.639813
    },
    {
      "acc": 0.63036313,
      "epoch": 0.2511415525114155,
      "grad_norm": 5.125,
      "learning_rate": 9.84467589971526e-06,
      "loss": 1.7552578,
      "memory(GiB)": 97.12,
      "step": 9900,
      "train_speed(iter/s)": 1.639896
    },
    {
      "acc": 0.64947052,
      "epoch": 0.2512683916793506,
      "grad_norm": 6.53125,
      "learning_rate": 9.844416453002027e-06,
      "loss": 1.67681274,
      "memory(GiB)": 97.12,
      "step": 9905,
      "train_speed(iter/s)": 1.639984
    },
    {
      "acc": 0.64688048,
      "epoch": 0.25139523084728566,
      "grad_norm": 5.15625,
      "learning_rate": 9.844156793209725e-06,
      "loss": 1.63420868,
      "memory(GiB)": 97.12,
      "step": 9910,
      "train_speed(iter/s)": 1.640075
    },
    {
      "acc": 0.64966488,
      "epoch": 0.2515220700152207,
      "grad_norm": 6.125,
      "learning_rate": 9.843896920349783e-06,
      "loss": 1.65382786,
      "memory(GiB)": 97.12,
      "step": 9915,
      "train_speed(iter/s)": 1.640165
    },
    {
      "acc": 0.64519625,
      "epoch": 0.25164890918315574,
      "grad_norm": 4.6875,
      "learning_rate": 9.843636834433627e-06,
      "loss": 1.68276272,
      "memory(GiB)": 97.12,
      "step": 9920,
      "train_speed(iter/s)": 1.640259
    },
    {
      "acc": 0.64749675,
      "epoch": 0.25177574835109084,
      "grad_norm": 5.65625,
      "learning_rate": 9.843376535472698e-06,
      "loss": 1.64375229,
      "memory(GiB)": 97.12,
      "step": 9925,
      "train_speed(iter/s)": 1.640357
    },
    {
      "acc": 0.65989466,
      "epoch": 0.2519025875190259,
      "grad_norm": 4.4375,
      "learning_rate": 9.843116023478445e-06,
      "loss": 1.57091999,
      "memory(GiB)": 97.12,
      "step": 9930,
      "train_speed(iter/s)": 1.640444
    },
    {
      "acc": 0.6505558,
      "epoch": 0.25202942668696093,
      "grad_norm": 5.09375,
      "learning_rate": 9.842855298462327e-06,
      "loss": 1.68581715,
      "memory(GiB)": 97.12,
      "step": 9935,
      "train_speed(iter/s)": 1.640529
    },
    {
      "acc": 0.64149489,
      "epoch": 0.252156265854896,
      "grad_norm": 6.28125,
      "learning_rate": 9.84259436043581e-06,
      "loss": 1.64889679,
      "memory(GiB)": 97.12,
      "step": 9940,
      "train_speed(iter/s)": 1.64062
    },
    {
      "acc": 0.64940042,
      "epoch": 0.2522831050228311,
      "grad_norm": 5.90625,
      "learning_rate": 9.842333209410372e-06,
      "loss": 1.62770901,
      "memory(GiB)": 97.12,
      "step": 9945,
      "train_speed(iter/s)": 1.640706
    },
    {
      "acc": 0.65872822,
      "epoch": 0.2524099441907661,
      "grad_norm": 5.28125,
      "learning_rate": 9.842071845397502e-06,
      "loss": 1.5567337,
      "memory(GiB)": 97.12,
      "step": 9950,
      "train_speed(iter/s)": 1.640796
    },
    {
      "acc": 0.63890805,
      "epoch": 0.25253678335870117,
      "grad_norm": 5.5,
      "learning_rate": 9.841810268408692e-06,
      "loss": 1.6995182,
      "memory(GiB)": 97.12,
      "step": 9955,
      "train_speed(iter/s)": 1.640886
    },
    {
      "acc": 0.65421972,
      "epoch": 0.2526636225266362,
      "grad_norm": 5.0625,
      "learning_rate": 9.841548478455451e-06,
      "loss": 1.71245689,
      "memory(GiB)": 97.12,
      "step": 9960,
      "train_speed(iter/s)": 1.640977
    },
    {
      "acc": 0.6484355,
      "epoch": 0.2527904616945713,
      "grad_norm": 5.96875,
      "learning_rate": 9.841286475549291e-06,
      "loss": 1.6949028,
      "memory(GiB)": 97.12,
      "step": 9965,
      "train_speed(iter/s)": 1.641062
    },
    {
      "acc": 0.64724045,
      "epoch": 0.25291730086250636,
      "grad_norm": 5.75,
      "learning_rate": 9.841024259701737e-06,
      "loss": 1.62486916,
      "memory(GiB)": 97.12,
      "step": 9970,
      "train_speed(iter/s)": 1.641149
    },
    {
      "acc": 0.64365444,
      "epoch": 0.2530441400304414,
      "grad_norm": 5.15625,
      "learning_rate": 9.840761830924323e-06,
      "loss": 1.66178799,
      "memory(GiB)": 97.12,
      "step": 9975,
      "train_speed(iter/s)": 1.641243
    },
    {
      "acc": 0.64171257,
      "epoch": 0.25317097919837644,
      "grad_norm": 7.15625,
      "learning_rate": 9.84049918922859e-06,
      "loss": 1.68366013,
      "memory(GiB)": 97.12,
      "step": 9980,
      "train_speed(iter/s)": 1.641335
    },
    {
      "acc": 0.63384504,
      "epoch": 0.25329781836631154,
      "grad_norm": 5.625,
      "learning_rate": 9.840236334626091e-06,
      "loss": 1.66978188,
      "memory(GiB)": 97.12,
      "step": 9985,
      "train_speed(iter/s)": 1.64142
    },
    {
      "acc": 0.64788871,
      "epoch": 0.2534246575342466,
      "grad_norm": 4.46875,
      "learning_rate": 9.83997326712839e-06,
      "loss": 1.63131371,
      "memory(GiB)": 97.12,
      "step": 9990,
      "train_speed(iter/s)": 1.64151
    },
    {
      "acc": 0.66368647,
      "epoch": 0.25355149670218163,
      "grad_norm": 4.84375,
      "learning_rate": 9.839709986747054e-06,
      "loss": 1.56834793,
      "memory(GiB)": 97.12,
      "step": 9995,
      "train_speed(iter/s)": 1.641598
    },
    {
      "acc": 0.65299573,
      "epoch": 0.2536783358701167,
      "grad_norm": 5.03125,
      "learning_rate": 9.839446493493667e-06,
      "loss": 1.57435932,
      "memory(GiB)": 97.12,
      "step": 10000,
      "train_speed(iter/s)": 1.641683
    },
    {
      "epoch": 0.2536783358701167,
      "eval_acc": 0.6385112883386899,
      "eval_loss": 1.6177022457122803,
      "eval_runtime": 58.776,
      "eval_samples_per_second": 108.378,
      "eval_steps_per_second": 27.103,
      "step": 10000
    },
    {
      "acc": 0.64907846,
      "epoch": 0.2538051750380518,
      "grad_norm": 5.0,
      "learning_rate": 9.839182787379815e-06,
      "loss": 1.58981676,
      "memory(GiB)": 97.12,
      "step": 10005,
      "train_speed(iter/s)": 1.624967
    },
    {
      "acc": 0.64923172,
      "epoch": 0.2539320142059868,
      "grad_norm": 5.53125,
      "learning_rate": 9.8389188684171e-06,
      "loss": 1.60591602,
      "memory(GiB)": 97.12,
      "step": 10010,
      "train_speed(iter/s)": 1.625059
    },
    {
      "acc": 0.65199113,
      "epoch": 0.25405885337392187,
      "grad_norm": 5.46875,
      "learning_rate": 9.838654736617128e-06,
      "loss": 1.55235291,
      "memory(GiB)": 97.12,
      "step": 10015,
      "train_speed(iter/s)": 1.625155
    },
    {
      "acc": 0.64934359,
      "epoch": 0.2541856925418569,
      "grad_norm": 5.65625,
      "learning_rate": 9.838390391991517e-06,
      "loss": 1.62928314,
      "memory(GiB)": 97.12,
      "step": 10020,
      "train_speed(iter/s)": 1.625253
    },
    {
      "acc": 0.64461603,
      "epoch": 0.254312531709792,
      "grad_norm": 5.65625,
      "learning_rate": 9.838125834551895e-06,
      "loss": 1.67571144,
      "memory(GiB)": 97.12,
      "step": 10025,
      "train_speed(iter/s)": 1.625348
    },
    {
      "acc": 0.6496933,
      "epoch": 0.25443937087772706,
      "grad_norm": 5.4375,
      "learning_rate": 9.837861064309899e-06,
      "loss": 1.67061977,
      "memory(GiB)": 97.12,
      "step": 10030,
      "train_speed(iter/s)": 1.625444
    },
    {
      "acc": 0.64946065,
      "epoch": 0.2545662100456621,
      "grad_norm": 5.46875,
      "learning_rate": 9.837596081277173e-06,
      "loss": 1.58180618,
      "memory(GiB)": 97.12,
      "step": 10035,
      "train_speed(iter/s)": 1.625539
    },
    {
      "acc": 0.65233727,
      "epoch": 0.25469304921359714,
      "grad_norm": 5.875,
      "learning_rate": 9.837330885465373e-06,
      "loss": 1.65300846,
      "memory(GiB)": 97.12,
      "step": 10040,
      "train_speed(iter/s)": 1.625636
    },
    {
      "acc": 0.64740419,
      "epoch": 0.25481988838153224,
      "grad_norm": 5.09375,
      "learning_rate": 9.837065476886163e-06,
      "loss": 1.66153011,
      "memory(GiB)": 97.12,
      "step": 10045,
      "train_speed(iter/s)": 1.625731
    },
    {
      "acc": 0.644137,
      "epoch": 0.2549467275494673,
      "grad_norm": 7.15625,
      "learning_rate": 9.83679985555122e-06,
      "loss": 1.64038887,
      "memory(GiB)": 97.12,
      "step": 10050,
      "train_speed(iter/s)": 1.625823
    },
    {
      "acc": 0.64094353,
      "epoch": 0.25507356671740233,
      "grad_norm": 5.75,
      "learning_rate": 9.836534021472222e-06,
      "loss": 1.67950783,
      "memory(GiB)": 97.12,
      "step": 10055,
      "train_speed(iter/s)": 1.625919
    },
    {
      "acc": 0.65049376,
      "epoch": 0.2552004058853374,
      "grad_norm": 5.96875,
      "learning_rate": 9.836267974660866e-06,
      "loss": 1.67010841,
      "memory(GiB)": 97.12,
      "step": 10060,
      "train_speed(iter/s)": 1.626009
    },
    {
      "acc": 0.65028048,
      "epoch": 0.2553272450532725,
      "grad_norm": 5.46875,
      "learning_rate": 9.836001715128851e-06,
      "loss": 1.66651993,
      "memory(GiB)": 97.12,
      "step": 10065,
      "train_speed(iter/s)": 1.6261
    },
    {
      "acc": 0.64809399,
      "epoch": 0.2554540842212075,
      "grad_norm": 6.28125,
      "learning_rate": 9.835735242887889e-06,
      "loss": 1.61924095,
      "memory(GiB)": 97.12,
      "step": 10070,
      "train_speed(iter/s)": 1.626198
    },
    {
      "acc": 0.65004787,
      "epoch": 0.25558092338914257,
      "grad_norm": 5.34375,
      "learning_rate": 9.835468557949701e-06,
      "loss": 1.62342682,
      "memory(GiB)": 97.12,
      "step": 10075,
      "train_speed(iter/s)": 1.626294
    },
    {
      "acc": 0.64583135,
      "epoch": 0.2557077625570776,
      "grad_norm": 5.0,
      "learning_rate": 9.83520166032602e-06,
      "loss": 1.65300865,
      "memory(GiB)": 97.12,
      "step": 10080,
      "train_speed(iter/s)": 1.626388
    },
    {
      "acc": 0.64032602,
      "epoch": 0.2558346017250127,
      "grad_norm": 6.34375,
      "learning_rate": 9.834934550028579e-06,
      "loss": 1.63845844,
      "memory(GiB)": 97.12,
      "step": 10085,
      "train_speed(iter/s)": 1.626482
    },
    {
      "acc": 0.64456515,
      "epoch": 0.25596144089294776,
      "grad_norm": 6.34375,
      "learning_rate": 9.83466722706913e-06,
      "loss": 1.68282528,
      "memory(GiB)": 97.12,
      "step": 10090,
      "train_speed(iter/s)": 1.626573
    },
    {
      "acc": 0.64471092,
      "epoch": 0.2560882800608828,
      "grad_norm": 5.71875,
      "learning_rate": 9.834399691459433e-06,
      "loss": 1.67646332,
      "memory(GiB)": 97.12,
      "step": 10095,
      "train_speed(iter/s)": 1.626662
    },
    {
      "acc": 0.64106884,
      "epoch": 0.25621511922881784,
      "grad_norm": 5.5625,
      "learning_rate": 9.83413194321125e-06,
      "loss": 1.62911358,
      "memory(GiB)": 97.12,
      "step": 10100,
      "train_speed(iter/s)": 1.626753
    },
    {
      "acc": 0.65749855,
      "epoch": 0.25634195839675294,
      "grad_norm": 6.78125,
      "learning_rate": 9.833863982336365e-06,
      "loss": 1.60307426,
      "memory(GiB)": 97.12,
      "step": 10105,
      "train_speed(iter/s)": 1.626844
    },
    {
      "acc": 0.66185999,
      "epoch": 0.256468797564688,
      "grad_norm": 4.90625,
      "learning_rate": 9.83359580884656e-06,
      "loss": 1.64014359,
      "memory(GiB)": 97.12,
      "step": 10110,
      "train_speed(iter/s)": 1.626938
    },
    {
      "acc": 0.66246119,
      "epoch": 0.25659563673262303,
      "grad_norm": 5.34375,
      "learning_rate": 9.83332742275363e-06,
      "loss": 1.63745556,
      "memory(GiB)": 97.12,
      "step": 10115,
      "train_speed(iter/s)": 1.627026
    },
    {
      "acc": 0.63125472,
      "epoch": 0.2567224759005581,
      "grad_norm": 7.3125,
      "learning_rate": 9.833058824069382e-06,
      "loss": 1.69879913,
      "memory(GiB)": 97.12,
      "step": 10120,
      "train_speed(iter/s)": 1.627117
    },
    {
      "acc": 0.65348196,
      "epoch": 0.2568493150684932,
      "grad_norm": 6.375,
      "learning_rate": 9.832790012805626e-06,
      "loss": 1.6033205,
      "memory(GiB)": 97.12,
      "step": 10125,
      "train_speed(iter/s)": 1.627209
    },
    {
      "acc": 0.65726557,
      "epoch": 0.2569761542364282,
      "grad_norm": 6.90625,
      "learning_rate": 9.832520988974191e-06,
      "loss": 1.57655811,
      "memory(GiB)": 97.12,
      "step": 10130,
      "train_speed(iter/s)": 1.627302
    },
    {
      "acc": 0.65688319,
      "epoch": 0.25710299340436327,
      "grad_norm": 5.3125,
      "learning_rate": 9.832251752586907e-06,
      "loss": 1.58754587,
      "memory(GiB)": 97.12,
      "step": 10135,
      "train_speed(iter/s)": 1.627391
    },
    {
      "acc": 0.65311823,
      "epoch": 0.2572298325722983,
      "grad_norm": 5.0,
      "learning_rate": 9.831982303655617e-06,
      "loss": 1.59830036,
      "memory(GiB)": 97.12,
      "step": 10140,
      "train_speed(iter/s)": 1.627481
    },
    {
      "acc": 0.65430326,
      "epoch": 0.2573566717402334,
      "grad_norm": 5.65625,
      "learning_rate": 9.83171264219217e-06,
      "loss": 1.62602406,
      "memory(GiB)": 97.12,
      "step": 10145,
      "train_speed(iter/s)": 1.627566
    },
    {
      "acc": 0.6770484,
      "epoch": 0.25748351090816846,
      "grad_norm": 5.71875,
      "learning_rate": 9.831442768208429e-06,
      "loss": 1.4724946,
      "memory(GiB)": 97.12,
      "step": 10150,
      "train_speed(iter/s)": 1.627653
    },
    {
      "acc": 0.67030296,
      "epoch": 0.2576103500761035,
      "grad_norm": 4.75,
      "learning_rate": 9.831172681716265e-06,
      "loss": 1.60321083,
      "memory(GiB)": 97.12,
      "step": 10155,
      "train_speed(iter/s)": 1.627739
    },
    {
      "acc": 0.6637229,
      "epoch": 0.25773718924403854,
      "grad_norm": 5.0,
      "learning_rate": 9.830902382727556e-06,
      "loss": 1.60972404,
      "memory(GiB)": 97.12,
      "step": 10160,
      "train_speed(iter/s)": 1.627829
    },
    {
      "acc": 0.66137056,
      "epoch": 0.25786402841197364,
      "grad_norm": 5.0625,
      "learning_rate": 9.830631871254193e-06,
      "loss": 1.59377689,
      "memory(GiB)": 97.12,
      "step": 10165,
      "train_speed(iter/s)": 1.62792
    },
    {
      "acc": 0.65381718,
      "epoch": 0.2579908675799087,
      "grad_norm": 5.28125,
      "learning_rate": 9.830361147308074e-06,
      "loss": 1.6218132,
      "memory(GiB)": 97.12,
      "step": 10170,
      "train_speed(iter/s)": 1.628017
    },
    {
      "acc": 0.64129176,
      "epoch": 0.25811770674784373,
      "grad_norm": 4.6875,
      "learning_rate": 9.830090210901104e-06,
      "loss": 1.6643589,
      "memory(GiB)": 97.12,
      "step": 10175,
      "train_speed(iter/s)": 1.628111
    },
    {
      "acc": 0.64200764,
      "epoch": 0.2582445459157788,
      "grad_norm": 6.4375,
      "learning_rate": 9.829819062045203e-06,
      "loss": 1.67048416,
      "memory(GiB)": 97.12,
      "step": 10180,
      "train_speed(iter/s)": 1.628199
    },
    {
      "acc": 0.66616635,
      "epoch": 0.2583713850837139,
      "grad_norm": 5.65625,
      "learning_rate": 9.829547700752295e-06,
      "loss": 1.54576893,
      "memory(GiB)": 97.12,
      "step": 10185,
      "train_speed(iter/s)": 1.628293
    },
    {
      "acc": 0.63299341,
      "epoch": 0.2584982242516489,
      "grad_norm": 5.28125,
      "learning_rate": 9.829276127034315e-06,
      "loss": 1.71097012,
      "memory(GiB)": 97.12,
      "step": 10190,
      "train_speed(iter/s)": 1.628384
    },
    {
      "acc": 0.63835087,
      "epoch": 0.25862506341958397,
      "grad_norm": 6.125,
      "learning_rate": 9.829004340903214e-06,
      "loss": 1.67719135,
      "memory(GiB)": 97.12,
      "step": 10195,
      "train_speed(iter/s)": 1.628477
    },
    {
      "acc": 0.64494486,
      "epoch": 0.258751902587519,
      "grad_norm": 4.4375,
      "learning_rate": 9.82873234237094e-06,
      "loss": 1.6942091,
      "memory(GiB)": 97.12,
      "step": 10200,
      "train_speed(iter/s)": 1.628546
    },
    {
      "acc": 0.63703461,
      "epoch": 0.2588787417554541,
      "grad_norm": 7.3125,
      "learning_rate": 9.828460131449457e-06,
      "loss": 1.72451572,
      "memory(GiB)": 97.12,
      "step": 10205,
      "train_speed(iter/s)": 1.628641
    },
    {
      "acc": 0.63502483,
      "epoch": 0.25900558092338916,
      "grad_norm": 5.65625,
      "learning_rate": 9.828187708150743e-06,
      "loss": 1.67805843,
      "memory(GiB)": 97.12,
      "step": 10210,
      "train_speed(iter/s)": 1.628727
    },
    {
      "acc": 0.66034803,
      "epoch": 0.2591324200913242,
      "grad_norm": 5.28125,
      "learning_rate": 9.827915072486776e-06,
      "loss": 1.59761457,
      "memory(GiB)": 97.12,
      "step": 10215,
      "train_speed(iter/s)": 1.628817
    },
    {
      "acc": 0.65768232,
      "epoch": 0.25925925925925924,
      "grad_norm": 6.0625,
      "learning_rate": 9.827642224469547e-06,
      "loss": 1.6083252,
      "memory(GiB)": 97.12,
      "step": 10220,
      "train_speed(iter/s)": 1.62891
    },
    {
      "acc": 0.64734974,
      "epoch": 0.25938609842719434,
      "grad_norm": 6.6875,
      "learning_rate": 9.827369164111062e-06,
      "loss": 1.69996796,
      "memory(GiB)": 97.12,
      "step": 10225,
      "train_speed(iter/s)": 1.629003
    },
    {
      "acc": 0.64079723,
      "epoch": 0.2595129375951294,
      "grad_norm": 6.28125,
      "learning_rate": 9.827095891423328e-06,
      "loss": 1.73891068,
      "memory(GiB)": 97.12,
      "step": 10230,
      "train_speed(iter/s)": 1.629102
    },
    {
      "acc": 0.66326094,
      "epoch": 0.25963977676306443,
      "grad_norm": 6.28125,
      "learning_rate": 9.826822406418366e-06,
      "loss": 1.53361692,
      "memory(GiB)": 97.12,
      "step": 10235,
      "train_speed(iter/s)": 1.629191
    },
    {
      "acc": 0.6562459,
      "epoch": 0.2597666159309995,
      "grad_norm": 4.875,
      "learning_rate": 9.826548709108202e-06,
      "loss": 1.55373507,
      "memory(GiB)": 97.12,
      "step": 10240,
      "train_speed(iter/s)": 1.62928
    },
    {
      "acc": 0.64401865,
      "epoch": 0.2598934550989346,
      "grad_norm": 5.8125,
      "learning_rate": 9.826274799504878e-06,
      "loss": 1.61399059,
      "memory(GiB)": 97.12,
      "step": 10245,
      "train_speed(iter/s)": 1.629372
    },
    {
      "acc": 0.65200453,
      "epoch": 0.2600202942668696,
      "grad_norm": 4.96875,
      "learning_rate": 9.82600067762044e-06,
      "loss": 1.63602333,
      "memory(GiB)": 97.12,
      "step": 10250,
      "train_speed(iter/s)": 1.629468
    },
    {
      "acc": 0.66021328,
      "epoch": 0.26014713343480467,
      "grad_norm": 5.90625,
      "learning_rate": 9.825726343466947e-06,
      "loss": 1.6139225,
      "memory(GiB)": 97.12,
      "step": 10255,
      "train_speed(iter/s)": 1.629559
    },
    {
      "acc": 0.64531975,
      "epoch": 0.2602739726027397,
      "grad_norm": 5.5,
      "learning_rate": 9.825451797056462e-06,
      "loss": 1.66134338,
      "memory(GiB)": 97.12,
      "step": 10260,
      "train_speed(iter/s)": 1.629651
    },
    {
      "acc": 0.64689102,
      "epoch": 0.2604008117706748,
      "grad_norm": 6.40625,
      "learning_rate": 9.825177038401064e-06,
      "loss": 1.64150772,
      "memory(GiB)": 97.12,
      "step": 10265,
      "train_speed(iter/s)": 1.629742
    },
    {
      "acc": 0.64313393,
      "epoch": 0.26052765093860986,
      "grad_norm": 5.03125,
      "learning_rate": 9.824902067512838e-06,
      "loss": 1.67537346,
      "memory(GiB)": 97.12,
      "step": 10270,
      "train_speed(iter/s)": 1.629834
    },
    {
      "acc": 0.65759106,
      "epoch": 0.2606544901065449,
      "grad_norm": 5.15625,
      "learning_rate": 9.824626884403877e-06,
      "loss": 1.62237453,
      "memory(GiB)": 97.12,
      "step": 10275,
      "train_speed(iter/s)": 1.629926
    },
    {
      "acc": 0.64620781,
      "epoch": 0.26078132927447994,
      "grad_norm": 5.125,
      "learning_rate": 9.824351489086283e-06,
      "loss": 1.62887878,
      "memory(GiB)": 97.12,
      "step": 10280,
      "train_speed(iter/s)": 1.630012
    },
    {
      "acc": 0.65930338,
      "epoch": 0.26090816844241504,
      "grad_norm": 7.5625,
      "learning_rate": 9.824075881572176e-06,
      "loss": 1.63321438,
      "memory(GiB)": 97.12,
      "step": 10285,
      "train_speed(iter/s)": 1.630105
    },
    {
      "acc": 0.66241989,
      "epoch": 0.2610350076103501,
      "grad_norm": 5.71875,
      "learning_rate": 9.823800061873669e-06,
      "loss": 1.63174801,
      "memory(GiB)": 97.12,
      "step": 10290,
      "train_speed(iter/s)": 1.630197
    },
    {
      "acc": 0.655688,
      "epoch": 0.26116184677828513,
      "grad_norm": 6.78125,
      "learning_rate": 9.8235240300029e-06,
      "loss": 1.68890839,
      "memory(GiB)": 97.12,
      "step": 10295,
      "train_speed(iter/s)": 1.630291
    },
    {
      "acc": 0.65024319,
      "epoch": 0.2612886859462202,
      "grad_norm": 5.09375,
      "learning_rate": 9.82324778597201e-06,
      "loss": 1.62787571,
      "memory(GiB)": 97.12,
      "step": 10300,
      "train_speed(iter/s)": 1.630383
    },
    {
      "acc": 0.64887404,
      "epoch": 0.2614155251141553,
      "grad_norm": 5.46875,
      "learning_rate": 9.822971329793147e-06,
      "loss": 1.6628849,
      "memory(GiB)": 97.12,
      "step": 10305,
      "train_speed(iter/s)": 1.630478
    },
    {
      "acc": 0.66143398,
      "epoch": 0.2615423642820903,
      "grad_norm": 5.0625,
      "learning_rate": 9.822694661478471e-06,
      "loss": 1.55211201,
      "memory(GiB)": 97.12,
      "step": 10310,
      "train_speed(iter/s)": 1.630566
    },
    {
      "acc": 0.64220591,
      "epoch": 0.26166920345002537,
      "grad_norm": 6.28125,
      "learning_rate": 9.822417781040154e-06,
      "loss": 1.66868706,
      "memory(GiB)": 97.12,
      "step": 10315,
      "train_speed(iter/s)": 1.630655
    },
    {
      "acc": 0.64102364,
      "epoch": 0.2617960426179604,
      "grad_norm": 4.8125,
      "learning_rate": 9.822140688490372e-06,
      "loss": 1.63731155,
      "memory(GiB)": 97.12,
      "step": 10320,
      "train_speed(iter/s)": 1.630744
    },
    {
      "acc": 0.65031204,
      "epoch": 0.2619228817858955,
      "grad_norm": 5.875,
      "learning_rate": 9.821863383841312e-06,
      "loss": 1.62481346,
      "memory(GiB)": 97.12,
      "step": 10325,
      "train_speed(iter/s)": 1.630833
    },
    {
      "acc": 0.66170759,
      "epoch": 0.26204972095383056,
      "grad_norm": 5.75,
      "learning_rate": 9.821585867105173e-06,
      "loss": 1.62501106,
      "memory(GiB)": 97.12,
      "step": 10330,
      "train_speed(iter/s)": 1.630925
    },
    {
      "acc": 0.63578653,
      "epoch": 0.2621765601217656,
      "grad_norm": 5.8125,
      "learning_rate": 9.821308138294162e-06,
      "loss": 1.6882782,
      "memory(GiB)": 97.12,
      "step": 10335,
      "train_speed(iter/s)": 1.631012
    },
    {
      "acc": 0.64705033,
      "epoch": 0.26230339928970065,
      "grad_norm": 6.03125,
      "learning_rate": 9.821030197420492e-06,
      "loss": 1.6899147,
      "memory(GiB)": 97.12,
      "step": 10340,
      "train_speed(iter/s)": 1.631102
    },
    {
      "acc": 0.64936504,
      "epoch": 0.26243023845763574,
      "grad_norm": 5.65625,
      "learning_rate": 9.820752044496389e-06,
      "loss": 1.63202305,
      "memory(GiB)": 97.12,
      "step": 10345,
      "train_speed(iter/s)": 1.63119
    },
    {
      "acc": 0.63620114,
      "epoch": 0.2625570776255708,
      "grad_norm": 5.90625,
      "learning_rate": 9.82047367953409e-06,
      "loss": 1.60717659,
      "memory(GiB)": 97.12,
      "step": 10350,
      "train_speed(iter/s)": 1.631271
    },
    {
      "acc": 0.66164856,
      "epoch": 0.26268391679350583,
      "grad_norm": 6.125,
      "learning_rate": 9.820195102545835e-06,
      "loss": 1.59535751,
      "memory(GiB)": 97.12,
      "step": 10355,
      "train_speed(iter/s)": 1.631362
    },
    {
      "acc": 0.63654194,
      "epoch": 0.2628107559614409,
      "grad_norm": 4.75,
      "learning_rate": 9.81991631354388e-06,
      "loss": 1.63763847,
      "memory(GiB)": 97.12,
      "step": 10360,
      "train_speed(iter/s)": 1.631455
    },
    {
      "acc": 0.64154792,
      "epoch": 0.262937595129376,
      "grad_norm": 5.75,
      "learning_rate": 9.819637312540485e-06,
      "loss": 1.71660652,
      "memory(GiB)": 97.12,
      "step": 10365,
      "train_speed(iter/s)": 1.631539
    },
    {
      "acc": 0.65010271,
      "epoch": 0.263064434297311,
      "grad_norm": 5.03125,
      "learning_rate": 9.819358099547923e-06,
      "loss": 1.62426643,
      "memory(GiB)": 97.12,
      "step": 10370,
      "train_speed(iter/s)": 1.631625
    },
    {
      "acc": 0.63507938,
      "epoch": 0.26319127346524607,
      "grad_norm": 5.34375,
      "learning_rate": 9.819078674578474e-06,
      "loss": 1.73412094,
      "memory(GiB)": 97.12,
      "step": 10375,
      "train_speed(iter/s)": 1.631717
    },
    {
      "acc": 0.63094172,
      "epoch": 0.2633181126331811,
      "grad_norm": 5.5625,
      "learning_rate": 9.818799037644432e-06,
      "loss": 1.71045914,
      "memory(GiB)": 97.12,
      "step": 10380,
      "train_speed(iter/s)": 1.631811
    },
    {
      "acc": 0.65303602,
      "epoch": 0.2634449518011162,
      "grad_norm": 4.9375,
      "learning_rate": 9.818519188758092e-06,
      "loss": 1.65938244,
      "memory(GiB)": 97.12,
      "step": 10385,
      "train_speed(iter/s)": 1.631901
    },
    {
      "acc": 0.65113139,
      "epoch": 0.26357179096905126,
      "grad_norm": 5.375,
      "learning_rate": 9.818239127931765e-06,
      "loss": 1.64128151,
      "memory(GiB)": 97.12,
      "step": 10390,
      "train_speed(iter/s)": 1.631993
    },
    {
      "acc": 0.63585367,
      "epoch": 0.2636986301369863,
      "grad_norm": 5.4375,
      "learning_rate": 9.81795885517777e-06,
      "loss": 1.7380724,
      "memory(GiB)": 97.12,
      "step": 10395,
      "train_speed(iter/s)": 1.63208
    },
    {
      "acc": 0.65230999,
      "epoch": 0.26382546930492135,
      "grad_norm": 4.625,
      "learning_rate": 9.817678370508434e-06,
      "loss": 1.67239571,
      "memory(GiB)": 97.12,
      "step": 10400,
      "train_speed(iter/s)": 1.632172
    },
    {
      "acc": 0.63712001,
      "epoch": 0.26395230847285645,
      "grad_norm": 5.71875,
      "learning_rate": 9.817397673936093e-06,
      "loss": 1.67624626,
      "memory(GiB)": 97.12,
      "step": 10405,
      "train_speed(iter/s)": 1.632264
    },
    {
      "acc": 0.6386425,
      "epoch": 0.2640791476407915,
      "grad_norm": 6.1875,
      "learning_rate": 9.817116765473095e-06,
      "loss": 1.71765099,
      "memory(GiB)": 97.12,
      "step": 10410,
      "train_speed(iter/s)": 1.63235
    },
    {
      "acc": 0.64774165,
      "epoch": 0.26420598680872653,
      "grad_norm": 5.28125,
      "learning_rate": 9.816835645131795e-06,
      "loss": 1.60468521,
      "memory(GiB)": 97.12,
      "step": 10415,
      "train_speed(iter/s)": 1.632442
    },
    {
      "acc": 0.64888167,
      "epoch": 0.2643328259766616,
      "grad_norm": 5.84375,
      "learning_rate": 9.816554312924555e-06,
      "loss": 1.61209583,
      "memory(GiB)": 97.12,
      "step": 10420,
      "train_speed(iter/s)": 1.632531
    },
    {
      "acc": 0.66007862,
      "epoch": 0.2644596651445967,
      "grad_norm": 5.53125,
      "learning_rate": 9.816272768863756e-06,
      "loss": 1.60714874,
      "memory(GiB)": 97.12,
      "step": 10425,
      "train_speed(iter/s)": 1.632613
    },
    {
      "acc": 0.65386915,
      "epoch": 0.2645865043125317,
      "grad_norm": 5.5625,
      "learning_rate": 9.815991012961773e-06,
      "loss": 1.60328808,
      "memory(GiB)": 97.12,
      "step": 10430,
      "train_speed(iter/s)": 1.632699
    },
    {
      "acc": 0.65707402,
      "epoch": 0.26471334348046677,
      "grad_norm": 5.09375,
      "learning_rate": 9.815709045231008e-06,
      "loss": 1.69138889,
      "memory(GiB)": 97.12,
      "step": 10435,
      "train_speed(iter/s)": 1.632783
    },
    {
      "acc": 0.64251747,
      "epoch": 0.2648401826484018,
      "grad_norm": 4.875,
      "learning_rate": 9.815426865683858e-06,
      "loss": 1.74192123,
      "memory(GiB)": 97.12,
      "step": 10440,
      "train_speed(iter/s)": 1.632875
    },
    {
      "acc": 0.65555048,
      "epoch": 0.2649670218163369,
      "grad_norm": 5.3125,
      "learning_rate": 9.815144474332732e-06,
      "loss": 1.60484638,
      "memory(GiB)": 97.12,
      "step": 10445,
      "train_speed(iter/s)": 1.632959
    },
    {
      "acc": 0.65817327,
      "epoch": 0.26509386098427196,
      "grad_norm": 5.1875,
      "learning_rate": 9.814861871190056e-06,
      "loss": 1.60427895,
      "memory(GiB)": 97.12,
      "step": 10450,
      "train_speed(iter/s)": 1.633043
    },
    {
      "acc": 0.6541358,
      "epoch": 0.265220700152207,
      "grad_norm": 6.46875,
      "learning_rate": 9.814579056268256e-06,
      "loss": 1.5771595,
      "memory(GiB)": 97.12,
      "step": 10455,
      "train_speed(iter/s)": 1.633126
    },
    {
      "acc": 0.65767775,
      "epoch": 0.26534753932014205,
      "grad_norm": 5.59375,
      "learning_rate": 9.814296029579776e-06,
      "loss": 1.61042709,
      "memory(GiB)": 97.12,
      "step": 10460,
      "train_speed(iter/s)": 1.633217
    },
    {
      "acc": 0.62915583,
      "epoch": 0.26547437848807715,
      "grad_norm": 5.21875,
      "learning_rate": 9.814012791137063e-06,
      "loss": 1.75139866,
      "memory(GiB)": 97.12,
      "step": 10465,
      "train_speed(iter/s)": 1.633303
    },
    {
      "acc": 0.64266024,
      "epoch": 0.2656012176560122,
      "grad_norm": 5.625,
      "learning_rate": 9.81372934095257e-06,
      "loss": 1.68595734,
      "memory(GiB)": 97.12,
      "step": 10470,
      "train_speed(iter/s)": 1.633381
    },
    {
      "acc": 0.62889385,
      "epoch": 0.26572805682394723,
      "grad_norm": 12.625,
      "learning_rate": 9.813445679038773e-06,
      "loss": 1.71039658,
      "memory(GiB)": 97.12,
      "step": 10475,
      "train_speed(iter/s)": 1.633468
    },
    {
      "acc": 0.63495979,
      "epoch": 0.2658548959918823,
      "grad_norm": 9.1875,
      "learning_rate": 9.813161805408145e-06,
      "loss": 1.70312405,
      "memory(GiB)": 97.12,
      "step": 10480,
      "train_speed(iter/s)": 1.633555
    },
    {
      "acc": 0.65271301,
      "epoch": 0.2659817351598174,
      "grad_norm": 5.6875,
      "learning_rate": 9.812877720073169e-06,
      "loss": 1.64671345,
      "memory(GiB)": 97.12,
      "step": 10485,
      "train_speed(iter/s)": 1.633641
    },
    {
      "acc": 0.64352012,
      "epoch": 0.2661085743277524,
      "grad_norm": 6.65625,
      "learning_rate": 9.812593423046344e-06,
      "loss": 1.69086628,
      "memory(GiB)": 97.12,
      "step": 10490,
      "train_speed(iter/s)": 1.633727
    },
    {
      "acc": 0.63739977,
      "epoch": 0.26623541349568747,
      "grad_norm": 4.9375,
      "learning_rate": 9.812308914340174e-06,
      "loss": 1.76656399,
      "memory(GiB)": 97.12,
      "step": 10495,
      "train_speed(iter/s)": 1.633812
    },
    {
      "acc": 0.64444456,
      "epoch": 0.2663622526636225,
      "grad_norm": 6.59375,
      "learning_rate": 9.812024193967171e-06,
      "loss": 1.70287209,
      "memory(GiB)": 97.12,
      "step": 10500,
      "train_speed(iter/s)": 1.633903
    },
    {
      "acc": 0.6491879,
      "epoch": 0.2664890918315576,
      "grad_norm": 5.46875,
      "learning_rate": 9.811739261939861e-06,
      "loss": 1.69029388,
      "memory(GiB)": 97.12,
      "step": 10505,
      "train_speed(iter/s)": 1.63399
    },
    {
      "acc": 0.65156522,
      "epoch": 0.26661593099949266,
      "grad_norm": 4.53125,
      "learning_rate": 9.811454118270775e-06,
      "loss": 1.67729492,
      "memory(GiB)": 97.12,
      "step": 10510,
      "train_speed(iter/s)": 1.634068
    },
    {
      "acc": 0.63503504,
      "epoch": 0.2667427701674277,
      "grad_norm": 5.59375,
      "learning_rate": 9.811168762972457e-06,
      "loss": 1.6670002,
      "memory(GiB)": 97.12,
      "step": 10515,
      "train_speed(iter/s)": 1.634155
    },
    {
      "acc": 0.64255152,
      "epoch": 0.26686960933536275,
      "grad_norm": 6.90625,
      "learning_rate": 9.810883196057454e-06,
      "loss": 1.66788483,
      "memory(GiB)": 97.12,
      "step": 10520,
      "train_speed(iter/s)": 1.63424
    },
    {
      "acc": 0.65614009,
      "epoch": 0.26699644850329785,
      "grad_norm": 5.40625,
      "learning_rate": 9.81059741753833e-06,
      "loss": 1.61547508,
      "memory(GiB)": 97.12,
      "step": 10525,
      "train_speed(iter/s)": 1.634326
    },
    {
      "acc": 0.64814548,
      "epoch": 0.2671232876712329,
      "grad_norm": 5.5625,
      "learning_rate": 9.810311427427653e-06,
      "loss": 1.64961472,
      "memory(GiB)": 97.12,
      "step": 10530,
      "train_speed(iter/s)": 1.634414
    },
    {
      "acc": 0.65829902,
      "epoch": 0.26725012683916793,
      "grad_norm": 5.03125,
      "learning_rate": 9.810025225738005e-06,
      "loss": 1.64391594,
      "memory(GiB)": 97.12,
      "step": 10535,
      "train_speed(iter/s)": 1.634502
    },
    {
      "acc": 0.61984053,
      "epoch": 0.267376966007103,
      "grad_norm": 7.28125,
      "learning_rate": 9.809738812481971e-06,
      "loss": 1.7640873,
      "memory(GiB)": 97.12,
      "step": 10540,
      "train_speed(iter/s)": 1.634585
    },
    {
      "acc": 0.64768834,
      "epoch": 0.2675038051750381,
      "grad_norm": 5.40625,
      "learning_rate": 9.809452187672149e-06,
      "loss": 1.63494072,
      "memory(GiB)": 97.12,
      "step": 10545,
      "train_speed(iter/s)": 1.634673
    },
    {
      "acc": 0.63901339,
      "epoch": 0.2676306443429731,
      "grad_norm": 4.9375,
      "learning_rate": 9.809165351321149e-06,
      "loss": 1.66816654,
      "memory(GiB)": 97.12,
      "step": 10550,
      "train_speed(iter/s)": 1.634759
    },
    {
      "acc": 0.66669645,
      "epoch": 0.26775748351090817,
      "grad_norm": 6.5625,
      "learning_rate": 9.808878303441585e-06,
      "loss": 1.59799433,
      "memory(GiB)": 97.12,
      "step": 10555,
      "train_speed(iter/s)": 1.634845
    },
    {
      "acc": 0.65254521,
      "epoch": 0.2678843226788432,
      "grad_norm": 6.9375,
      "learning_rate": 9.808591044046083e-06,
      "loss": 1.68372135,
      "memory(GiB)": 97.12,
      "step": 10560,
      "train_speed(iter/s)": 1.634929
    },
    {
      "acc": 0.65123339,
      "epoch": 0.2680111618467783,
      "grad_norm": 5.96875,
      "learning_rate": 9.808303573147277e-06,
      "loss": 1.66323242,
      "memory(GiB)": 97.17,
      "step": 10565,
      "train_speed(iter/s)": 1.635007
    },
    {
      "acc": 0.65460091,
      "epoch": 0.26813800101471336,
      "grad_norm": 6.46875,
      "learning_rate": 9.808015890757812e-06,
      "loss": 1.63129311,
      "memory(GiB)": 97.17,
      "step": 10570,
      "train_speed(iter/s)": 1.635097
    },
    {
      "acc": 0.65706034,
      "epoch": 0.2682648401826484,
      "grad_norm": 6.0,
      "learning_rate": 9.807727996890343e-06,
      "loss": 1.60857773,
      "memory(GiB)": 97.17,
      "step": 10575,
      "train_speed(iter/s)": 1.635179
    },
    {
      "acc": 0.64872966,
      "epoch": 0.26839167935058345,
      "grad_norm": 5.84375,
      "learning_rate": 9.807439891557533e-06,
      "loss": 1.64379444,
      "memory(GiB)": 97.17,
      "step": 10580,
      "train_speed(iter/s)": 1.635264
    },
    {
      "acc": 0.63933663,
      "epoch": 0.26851851851851855,
      "grad_norm": 6.1875,
      "learning_rate": 9.80715157477205e-06,
      "loss": 1.70501518,
      "memory(GiB)": 97.17,
      "step": 10585,
      "train_speed(iter/s)": 1.635347
    },
    {
      "acc": 0.64208035,
      "epoch": 0.2686453576864536,
      "grad_norm": 5.09375,
      "learning_rate": 9.806863046546581e-06,
      "loss": 1.69165344,
      "memory(GiB)": 97.17,
      "step": 10590,
      "train_speed(iter/s)": 1.635432
    },
    {
      "acc": 0.64665227,
      "epoch": 0.26877219685438863,
      "grad_norm": 4.6875,
      "learning_rate": 9.806574306893814e-06,
      "loss": 1.66041985,
      "memory(GiB)": 97.17,
      "step": 10595,
      "train_speed(iter/s)": 1.635518
    },
    {
      "acc": 0.66367922,
      "epoch": 0.2688990360223237,
      "grad_norm": 5.8125,
      "learning_rate": 9.806285355826447e-06,
      "loss": 1.60019341,
      "memory(GiB)": 97.17,
      "step": 10600,
      "train_speed(iter/s)": 1.635606
    },
    {
      "acc": 0.66562657,
      "epoch": 0.2690258751902588,
      "grad_norm": 6.0625,
      "learning_rate": 9.805996193357194e-06,
      "loss": 1.56760168,
      "memory(GiB)": 97.17,
      "step": 10605,
      "train_speed(iter/s)": 1.635693
    },
    {
      "acc": 0.62523212,
      "epoch": 0.2691527143581938,
      "grad_norm": 5.65625,
      "learning_rate": 9.80570681949877e-06,
      "loss": 1.68235588,
      "memory(GiB)": 97.17,
      "step": 10610,
      "train_speed(iter/s)": 1.635779
    },
    {
      "acc": 0.64439225,
      "epoch": 0.26927955352612887,
      "grad_norm": 5.21875,
      "learning_rate": 9.805417234263905e-06,
      "loss": 1.65805702,
      "memory(GiB)": 97.17,
      "step": 10615,
      "train_speed(iter/s)": 1.635869
    },
    {
      "acc": 0.63883467,
      "epoch": 0.2694063926940639,
      "grad_norm": 5.15625,
      "learning_rate": 9.805127437665333e-06,
      "loss": 1.75823822,
      "memory(GiB)": 97.17,
      "step": 10620,
      "train_speed(iter/s)": 1.635957
    },
    {
      "acc": 0.63620129,
      "epoch": 0.269533231861999,
      "grad_norm": 5.0625,
      "learning_rate": 9.804837429715805e-06,
      "loss": 1.70582237,
      "memory(GiB)": 97.17,
      "step": 10625,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.66414537,
      "epoch": 0.26966007102993406,
      "grad_norm": 5.09375,
      "learning_rate": 9.804547210428074e-06,
      "loss": 1.61211872,
      "memory(GiB)": 97.17,
      "step": 10630,
      "train_speed(iter/s)": 1.636122
    },
    {
      "acc": 0.63372812,
      "epoch": 0.2697869101978691,
      "grad_norm": 5.90625,
      "learning_rate": 9.804256779814906e-06,
      "loss": 1.69291878,
      "memory(GiB)": 97.17,
      "step": 10635,
      "train_speed(iter/s)": 1.636209
    },
    {
      "acc": 0.6433568,
      "epoch": 0.26991374936580415,
      "grad_norm": 5.375,
      "learning_rate": 9.803966137889076e-06,
      "loss": 1.6116333,
      "memory(GiB)": 97.17,
      "step": 10640,
      "train_speed(iter/s)": 1.636287
    },
    {
      "acc": 0.63128929,
      "epoch": 0.27004058853373925,
      "grad_norm": 5.75,
      "learning_rate": 9.803675284663368e-06,
      "loss": 1.72813683,
      "memory(GiB)": 97.17,
      "step": 10645,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.65008564,
      "epoch": 0.2701674277016743,
      "grad_norm": 6.28125,
      "learning_rate": 9.803384220150571e-06,
      "loss": 1.68646221,
      "memory(GiB)": 97.17,
      "step": 10650,
      "train_speed(iter/s)": 1.636463
    },
    {
      "acc": 0.63832312,
      "epoch": 0.27029426686960933,
      "grad_norm": 7.0,
      "learning_rate": 9.803092944363493e-06,
      "loss": 1.7142189,
      "memory(GiB)": 97.17,
      "step": 10655,
      "train_speed(iter/s)": 1.636553
    },
    {
      "acc": 0.66242409,
      "epoch": 0.2704211060375444,
      "grad_norm": 6.4375,
      "learning_rate": 9.802801457314943e-06,
      "loss": 1.61483803,
      "memory(GiB)": 97.17,
      "step": 10660,
      "train_speed(iter/s)": 1.636643
    },
    {
      "acc": 0.65640869,
      "epoch": 0.2705479452054795,
      "grad_norm": 7.3125,
      "learning_rate": 9.802509759017741e-06,
      "loss": 1.60437355,
      "memory(GiB)": 97.17,
      "step": 10665,
      "train_speed(iter/s)": 1.636727
    },
    {
      "acc": 0.63736796,
      "epoch": 0.2706747843734145,
      "grad_norm": 4.46875,
      "learning_rate": 9.802217849484719e-06,
      "loss": 1.72520905,
      "memory(GiB)": 97.17,
      "step": 10670,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.64779425,
      "epoch": 0.27080162354134957,
      "grad_norm": 6.1875,
      "learning_rate": 9.801925728728715e-06,
      "loss": 1.64617748,
      "memory(GiB)": 97.17,
      "step": 10675,
      "train_speed(iter/s)": 1.6369
    },
    {
      "acc": 0.6488193,
      "epoch": 0.2709284627092846,
      "grad_norm": 5.21875,
      "learning_rate": 9.801633396762577e-06,
      "loss": 1.66686363,
      "memory(GiB)": 97.17,
      "step": 10680,
      "train_speed(iter/s)": 1.636983
    },
    {
      "acc": 0.65254855,
      "epoch": 0.2710553018772197,
      "grad_norm": 6.1875,
      "learning_rate": 9.801340853599167e-06,
      "loss": 1.7029892,
      "memory(GiB)": 97.17,
      "step": 10685,
      "train_speed(iter/s)": 1.637065
    },
    {
      "acc": 0.65923319,
      "epoch": 0.27118214104515476,
      "grad_norm": 7.09375,
      "learning_rate": 9.801048099251348e-06,
      "loss": 1.60697441,
      "memory(GiB)": 97.17,
      "step": 10690,
      "train_speed(iter/s)": 1.637149
    },
    {
      "acc": 0.64430847,
      "epoch": 0.2713089802130898,
      "grad_norm": 5.125,
      "learning_rate": 9.800755133731999e-06,
      "loss": 1.6458149,
      "memory(GiB)": 97.17,
      "step": 10695,
      "train_speed(iter/s)": 1.637232
    },
    {
      "acc": 0.65353289,
      "epoch": 0.27143581938102485,
      "grad_norm": 5.5,
      "learning_rate": 9.800461957054006e-06,
      "loss": 1.63329773,
      "memory(GiB)": 97.17,
      "step": 10700,
      "train_speed(iter/s)": 1.637312
    },
    {
      "acc": 0.65184431,
      "epoch": 0.27156265854895995,
      "grad_norm": 5.9375,
      "learning_rate": 9.800168569230261e-06,
      "loss": 1.61327991,
      "memory(GiB)": 97.17,
      "step": 10705,
      "train_speed(iter/s)": 1.637394
    },
    {
      "acc": 0.65758886,
      "epoch": 0.271689497716895,
      "grad_norm": 5.125,
      "learning_rate": 9.799874970273674e-06,
      "loss": 1.59366589,
      "memory(GiB)": 97.17,
      "step": 10710,
      "train_speed(iter/s)": 1.637476
    },
    {
      "acc": 0.65175881,
      "epoch": 0.27181633688483003,
      "grad_norm": 5.0625,
      "learning_rate": 9.799581160197156e-06,
      "loss": 1.65314426,
      "memory(GiB)": 97.17,
      "step": 10715,
      "train_speed(iter/s)": 1.63756
    },
    {
      "acc": 0.65863752,
      "epoch": 0.2719431760527651,
      "grad_norm": 4.34375,
      "learning_rate": 9.799287139013628e-06,
      "loss": 1.59951286,
      "memory(GiB)": 97.17,
      "step": 10720,
      "train_speed(iter/s)": 1.637634
    },
    {
      "acc": 0.66159692,
      "epoch": 0.2720700152207002,
      "grad_norm": 5.53125,
      "learning_rate": 9.798992906736028e-06,
      "loss": 1.56686249,
      "memory(GiB)": 97.17,
      "step": 10725,
      "train_speed(iter/s)": 1.637716
    },
    {
      "acc": 0.64639587,
      "epoch": 0.2721968543886352,
      "grad_norm": 5.34375,
      "learning_rate": 9.79869846337729e-06,
      "loss": 1.67068996,
      "memory(GiB)": 97.17,
      "step": 10730,
      "train_speed(iter/s)": 1.637797
    },
    {
      "acc": 0.64133353,
      "epoch": 0.27232369355657027,
      "grad_norm": 6.40625,
      "learning_rate": 9.79840380895037e-06,
      "loss": 1.66402054,
      "memory(GiB)": 97.17,
      "step": 10735,
      "train_speed(iter/s)": 1.637874
    },
    {
      "acc": 0.65808792,
      "epoch": 0.2724505327245053,
      "grad_norm": 5.125,
      "learning_rate": 9.798108943468228e-06,
      "loss": 1.55847158,
      "memory(GiB)": 97.17,
      "step": 10740,
      "train_speed(iter/s)": 1.637956
    },
    {
      "acc": 0.645502,
      "epoch": 0.2725773718924404,
      "grad_norm": 4.96875,
      "learning_rate": 9.797813866943832e-06,
      "loss": 1.61261215,
      "memory(GiB)": 97.17,
      "step": 10745,
      "train_speed(iter/s)": 1.638035
    },
    {
      "acc": 0.6569036,
      "epoch": 0.27270421106037546,
      "grad_norm": 6.75,
      "learning_rate": 9.797518579390162e-06,
      "loss": 1.52561388,
      "memory(GiB)": 97.17,
      "step": 10750,
      "train_speed(iter/s)": 1.638118
    },
    {
      "acc": 0.65607042,
      "epoch": 0.2728310502283105,
      "grad_norm": 4.5625,
      "learning_rate": 9.797223080820204e-06,
      "loss": 1.63400593,
      "memory(GiB)": 97.17,
      "step": 10755,
      "train_speed(iter/s)": 1.638192
    },
    {
      "acc": 0.64432902,
      "epoch": 0.27295788939624555,
      "grad_norm": 5.5625,
      "learning_rate": 9.796927371246958e-06,
      "loss": 1.6504652,
      "memory(GiB)": 97.17,
      "step": 10760,
      "train_speed(iter/s)": 1.638276
    },
    {
      "acc": 0.64416261,
      "epoch": 0.27308472856418065,
      "grad_norm": 5.34375,
      "learning_rate": 9.796631450683431e-06,
      "loss": 1.63592777,
      "memory(GiB)": 97.17,
      "step": 10765,
      "train_speed(iter/s)": 1.638359
    },
    {
      "acc": 0.63731375,
      "epoch": 0.2732115677321157,
      "grad_norm": 5.28125,
      "learning_rate": 9.796335319142637e-06,
      "loss": 1.64401073,
      "memory(GiB)": 97.17,
      "step": 10770,
      "train_speed(iter/s)": 1.638439
    },
    {
      "acc": 0.64967432,
      "epoch": 0.27333840690005073,
      "grad_norm": 5.90625,
      "learning_rate": 9.796038976637599e-06,
      "loss": 1.63275318,
      "memory(GiB)": 97.17,
      "step": 10775,
      "train_speed(iter/s)": 1.63852
    },
    {
      "acc": 0.65172482,
      "epoch": 0.2734652460679858,
      "grad_norm": 5.3125,
      "learning_rate": 9.795742423181355e-06,
      "loss": 1.6356472,
      "memory(GiB)": 97.17,
      "step": 10780,
      "train_speed(iter/s)": 1.638603
    },
    {
      "acc": 0.63355989,
      "epoch": 0.2735920852359209,
      "grad_norm": 5.3125,
      "learning_rate": 9.795445658786948e-06,
      "loss": 1.74606743,
      "memory(GiB)": 97.17,
      "step": 10785,
      "train_speed(iter/s)": 1.638681
    },
    {
      "acc": 0.65085659,
      "epoch": 0.2737189244038559,
      "grad_norm": 5.09375,
      "learning_rate": 9.795148683467431e-06,
      "loss": 1.64198532,
      "memory(GiB)": 97.17,
      "step": 10790,
      "train_speed(iter/s)": 1.63876
    },
    {
      "acc": 0.64056921,
      "epoch": 0.27384576357179097,
      "grad_norm": 5.5,
      "learning_rate": 9.794851497235866e-06,
      "loss": 1.66883278,
      "memory(GiB)": 97.17,
      "step": 10795,
      "train_speed(iter/s)": 1.63884
    },
    {
      "acc": 0.63935785,
      "epoch": 0.273972602739726,
      "grad_norm": 6.21875,
      "learning_rate": 9.794554100105325e-06,
      "loss": 1.64258156,
      "memory(GiB)": 97.17,
      "step": 10800,
      "train_speed(iter/s)": 1.63892
    },
    {
      "acc": 0.6495461,
      "epoch": 0.2740994419076611,
      "grad_norm": 4.84375,
      "learning_rate": 9.794256492088888e-06,
      "loss": 1.6080349,
      "memory(GiB)": 97.17,
      "step": 10805,
      "train_speed(iter/s)": 1.639002
    },
    {
      "acc": 0.64967103,
      "epoch": 0.27422628107559616,
      "grad_norm": 6.09375,
      "learning_rate": 9.793958673199647e-06,
      "loss": 1.68081093,
      "memory(GiB)": 97.17,
      "step": 10810,
      "train_speed(iter/s)": 1.639082
    },
    {
      "acc": 0.6587369,
      "epoch": 0.2743531202435312,
      "grad_norm": 5.0,
      "learning_rate": 9.793660643450697e-06,
      "loss": 1.61782837,
      "memory(GiB)": 97.17,
      "step": 10815,
      "train_speed(iter/s)": 1.63916
    },
    {
      "acc": 0.64281974,
      "epoch": 0.27447995941146625,
      "grad_norm": 5.53125,
      "learning_rate": 9.793362402855152e-06,
      "loss": 1.64244423,
      "memory(GiB)": 97.17,
      "step": 10820,
      "train_speed(iter/s)": 1.639235
    },
    {
      "acc": 0.64013367,
      "epoch": 0.27460679857940135,
      "grad_norm": 5.5,
      "learning_rate": 9.79306395142613e-06,
      "loss": 1.63845024,
      "memory(GiB)": 97.17,
      "step": 10825,
      "train_speed(iter/s)": 1.639315
    },
    {
      "acc": 0.65530014,
      "epoch": 0.2747336377473364,
      "grad_norm": 4.625,
      "learning_rate": 9.792765289176751e-06,
      "loss": 1.60434837,
      "memory(GiB)": 97.17,
      "step": 10830,
      "train_speed(iter/s)": 1.639391
    },
    {
      "acc": 0.64279089,
      "epoch": 0.27486047691527143,
      "grad_norm": 5.34375,
      "learning_rate": 9.79246641612016e-06,
      "loss": 1.63921108,
      "memory(GiB)": 97.17,
      "step": 10835,
      "train_speed(iter/s)": 1.639471
    },
    {
      "acc": 0.64567795,
      "epoch": 0.2749873160832065,
      "grad_norm": 5.1875,
      "learning_rate": 9.792167332269498e-06,
      "loss": 1.62604542,
      "memory(GiB)": 97.17,
      "step": 10840,
      "train_speed(iter/s)": 1.63955
    },
    {
      "acc": 0.65950603,
      "epoch": 0.2751141552511416,
      "grad_norm": 4.84375,
      "learning_rate": 9.791868037637922e-06,
      "loss": 1.61371078,
      "memory(GiB)": 97.17,
      "step": 10845,
      "train_speed(iter/s)": 1.639627
    },
    {
      "acc": 0.64927797,
      "epoch": 0.2752409944190766,
      "grad_norm": 5.90625,
      "learning_rate": 9.791568532238594e-06,
      "loss": 1.66386395,
      "memory(GiB)": 97.17,
      "step": 10850,
      "train_speed(iter/s)": 1.639705
    },
    {
      "acc": 0.65427227,
      "epoch": 0.27536783358701167,
      "grad_norm": 5.75,
      "learning_rate": 9.79126881608469e-06,
      "loss": 1.66656418,
      "memory(GiB)": 97.17,
      "step": 10855,
      "train_speed(iter/s)": 1.63978
    },
    {
      "acc": 0.64983964,
      "epoch": 0.2754946727549467,
      "grad_norm": 6.15625,
      "learning_rate": 9.790968889189392e-06,
      "loss": 1.64085808,
      "memory(GiB)": 97.17,
      "step": 10860,
      "train_speed(iter/s)": 1.639858
    },
    {
      "acc": 0.65658627,
      "epoch": 0.2756215119228818,
      "grad_norm": 6.1875,
      "learning_rate": 9.790668751565893e-06,
      "loss": 1.65282135,
      "memory(GiB)": 97.17,
      "step": 10865,
      "train_speed(iter/s)": 1.639937
    },
    {
      "acc": 0.65698934,
      "epoch": 0.27574835109081686,
      "grad_norm": 4.875,
      "learning_rate": 9.790368403227391e-06,
      "loss": 1.67522392,
      "memory(GiB)": 97.17,
      "step": 10870,
      "train_speed(iter/s)": 1.640015
    },
    {
      "acc": 0.64279351,
      "epoch": 0.2758751902587519,
      "grad_norm": 4.65625,
      "learning_rate": 9.7900678441871e-06,
      "loss": 1.60776596,
      "memory(GiB)": 97.17,
      "step": 10875,
      "train_speed(iter/s)": 1.640092
    },
    {
      "acc": 0.64378433,
      "epoch": 0.27600202942668695,
      "grad_norm": 5.65625,
      "learning_rate": 9.78976707445824e-06,
      "loss": 1.625037,
      "memory(GiB)": 97.17,
      "step": 10880,
      "train_speed(iter/s)": 1.640171
    },
    {
      "acc": 0.65321908,
      "epoch": 0.27612886859462205,
      "grad_norm": 6.1875,
      "learning_rate": 9.78946609405404e-06,
      "loss": 1.65869293,
      "memory(GiB)": 97.17,
      "step": 10885,
      "train_speed(iter/s)": 1.640242
    },
    {
      "acc": 0.66312265,
      "epoch": 0.2762557077625571,
      "grad_norm": 8.875,
      "learning_rate": 9.789164902987738e-06,
      "loss": 1.61773643,
      "memory(GiB)": 97.17,
      "step": 10890,
      "train_speed(iter/s)": 1.640322
    },
    {
      "acc": 0.66834092,
      "epoch": 0.27638254693049213,
      "grad_norm": 5.03125,
      "learning_rate": 9.78886350127258e-06,
      "loss": 1.59298992,
      "memory(GiB)": 97.17,
      "step": 10895,
      "train_speed(iter/s)": 1.640399
    },
    {
      "acc": 0.64224973,
      "epoch": 0.2765093860984272,
      "grad_norm": 5.3125,
      "learning_rate": 9.788561888921825e-06,
      "loss": 1.69609795,
      "memory(GiB)": 97.17,
      "step": 10900,
      "train_speed(iter/s)": 1.64048
    },
    {
      "acc": 0.66138825,
      "epoch": 0.2766362252663623,
      "grad_norm": 6.625,
      "learning_rate": 9.788260065948738e-06,
      "loss": 1.65636482,
      "memory(GiB)": 97.17,
      "step": 10905,
      "train_speed(iter/s)": 1.640551
    },
    {
      "acc": 0.62622252,
      "epoch": 0.2767630644342973,
      "grad_norm": 6.03125,
      "learning_rate": 9.787958032366596e-06,
      "loss": 1.71186295,
      "memory(GiB)": 97.17,
      "step": 10910,
      "train_speed(iter/s)": 1.640623
    },
    {
      "acc": 0.6585865,
      "epoch": 0.27688990360223237,
      "grad_norm": 6.0,
      "learning_rate": 9.787655788188684e-06,
      "loss": 1.58614922,
      "memory(GiB)": 97.17,
      "step": 10915,
      "train_speed(iter/s)": 1.640697
    },
    {
      "acc": 0.66387935,
      "epoch": 0.2770167427701674,
      "grad_norm": 5.53125,
      "learning_rate": 9.787353333428293e-06,
      "loss": 1.54813852,
      "memory(GiB)": 97.17,
      "step": 10920,
      "train_speed(iter/s)": 1.640774
    },
    {
      "acc": 0.64432573,
      "epoch": 0.2771435819381025,
      "grad_norm": 6.09375,
      "learning_rate": 9.78705066809873e-06,
      "loss": 1.6542614,
      "memory(GiB)": 97.17,
      "step": 10925,
      "train_speed(iter/s)": 1.640848
    },
    {
      "acc": 0.63740487,
      "epoch": 0.27727042110603756,
      "grad_norm": 6.0,
      "learning_rate": 9.786747792213304e-06,
      "loss": 1.72560329,
      "memory(GiB)": 97.17,
      "step": 10930,
      "train_speed(iter/s)": 1.640921
    },
    {
      "acc": 0.65823984,
      "epoch": 0.2773972602739726,
      "grad_norm": 6.1875,
      "learning_rate": 9.78644470578534e-06,
      "loss": 1.62464523,
      "memory(GiB)": 97.17,
      "step": 10935,
      "train_speed(iter/s)": 1.640998
    },
    {
      "acc": 0.63928123,
      "epoch": 0.27752409944190765,
      "grad_norm": 5.09375,
      "learning_rate": 9.78614140882817e-06,
      "loss": 1.72192078,
      "memory(GiB)": 97.17,
      "step": 10940,
      "train_speed(iter/s)": 1.641074
    },
    {
      "acc": 0.66831856,
      "epoch": 0.27765093860984275,
      "grad_norm": 5.25,
      "learning_rate": 9.78583790135513e-06,
      "loss": 1.58727722,
      "memory(GiB)": 97.17,
      "step": 10945,
      "train_speed(iter/s)": 1.641145
    },
    {
      "acc": 0.65687213,
      "epoch": 0.2777777777777778,
      "grad_norm": 5.3125,
      "learning_rate": 9.785534183379571e-06,
      "loss": 1.6255949,
      "memory(GiB)": 97.17,
      "step": 10950,
      "train_speed(iter/s)": 1.641223
    },
    {
      "acc": 0.64482126,
      "epoch": 0.27790461694571283,
      "grad_norm": 5.25,
      "learning_rate": 9.785230254914855e-06,
      "loss": 1.68952808,
      "memory(GiB)": 97.17,
      "step": 10955,
      "train_speed(iter/s)": 1.641295
    },
    {
      "acc": 0.64171743,
      "epoch": 0.2780314561136479,
      "grad_norm": 5.0625,
      "learning_rate": 9.784926115974346e-06,
      "loss": 1.67399292,
      "memory(GiB)": 97.17,
      "step": 10960,
      "train_speed(iter/s)": 1.641376
    },
    {
      "acc": 0.65712643,
      "epoch": 0.278158295281583,
      "grad_norm": 4.9375,
      "learning_rate": 9.784621766571424e-06,
      "loss": 1.63215904,
      "memory(GiB)": 97.17,
      "step": 10965,
      "train_speed(iter/s)": 1.641448
    },
    {
      "acc": 0.64262872,
      "epoch": 0.278285134449518,
      "grad_norm": 5.5,
      "learning_rate": 9.784317206719475e-06,
      "loss": 1.73264351,
      "memory(GiB)": 97.17,
      "step": 10970,
      "train_speed(iter/s)": 1.64153
    },
    {
      "acc": 0.64218521,
      "epoch": 0.27841197361745307,
      "grad_norm": 4.59375,
      "learning_rate": 9.784012436431896e-06,
      "loss": 1.66874371,
      "memory(GiB)": 97.17,
      "step": 10975,
      "train_speed(iter/s)": 1.641607
    },
    {
      "acc": 0.65534229,
      "epoch": 0.2785388127853881,
      "grad_norm": 4.9375,
      "learning_rate": 9.78370745572209e-06,
      "loss": 1.66249256,
      "memory(GiB)": 97.17,
      "step": 10980,
      "train_speed(iter/s)": 1.641686
    },
    {
      "acc": 0.64304752,
      "epoch": 0.2786656519533232,
      "grad_norm": 6.4375,
      "learning_rate": 9.783402264603471e-06,
      "loss": 1.6038517,
      "memory(GiB)": 97.17,
      "step": 10985,
      "train_speed(iter/s)": 1.641764
    },
    {
      "acc": 0.64331083,
      "epoch": 0.27879249112125826,
      "grad_norm": 4.6875,
      "learning_rate": 9.783096863089465e-06,
      "loss": 1.6379612,
      "memory(GiB)": 97.17,
      "step": 10990,
      "train_speed(iter/s)": 1.641839
    },
    {
      "acc": 0.66518965,
      "epoch": 0.2789193302891933,
      "grad_norm": 5.875,
      "learning_rate": 9.782791251193505e-06,
      "loss": 1.51117487,
      "memory(GiB)": 97.17,
      "step": 10995,
      "train_speed(iter/s)": 1.641917
    },
    {
      "acc": 0.6459259,
      "epoch": 0.27904616945712835,
      "grad_norm": 4.59375,
      "learning_rate": 9.782485428929032e-06,
      "loss": 1.63293686,
      "memory(GiB)": 97.17,
      "step": 11000,
      "train_speed(iter/s)": 1.641992
    },
    {
      "epoch": 0.27904616945712835,
      "eval_acc": 0.6396231025128587,
      "eval_loss": 1.6118441820144653,
      "eval_runtime": 58.6012,
      "eval_samples_per_second": 108.701,
      "eval_steps_per_second": 27.184,
      "step": 11000
    },
    {
      "acc": 0.66373024,
      "epoch": 0.27917300862506345,
      "grad_norm": 5.65625,
      "learning_rate": 9.782179396309496e-06,
      "loss": 1.65504322,
      "memory(GiB)": 97.17,
      "step": 11005,
      "train_speed(iter/s)": 1.626816
    },
    {
      "acc": 0.63609104,
      "epoch": 0.2792998477929985,
      "grad_norm": 4.78125,
      "learning_rate": 9.78187315334836e-06,
      "loss": 1.66823139,
      "memory(GiB)": 97.17,
      "step": 11010,
      "train_speed(iter/s)": 1.626897
    },
    {
      "acc": 0.64733672,
      "epoch": 0.27942668696093353,
      "grad_norm": 5.875,
      "learning_rate": 9.781566700059094e-06,
      "loss": 1.65000801,
      "memory(GiB)": 97.17,
      "step": 11015,
      "train_speed(iter/s)": 1.626978
    },
    {
      "acc": 0.64460173,
      "epoch": 0.2795535261288686,
      "grad_norm": 7.34375,
      "learning_rate": 9.781260036455176e-06,
      "loss": 1.71171131,
      "memory(GiB)": 97.17,
      "step": 11020,
      "train_speed(iter/s)": 1.627058
    },
    {
      "acc": 0.66845303,
      "epoch": 0.2796803652968037,
      "grad_norm": 4.875,
      "learning_rate": 9.780953162550093e-06,
      "loss": 1.59873047,
      "memory(GiB)": 97.17,
      "step": 11025,
      "train_speed(iter/s)": 1.627134
    },
    {
      "acc": 0.64273496,
      "epoch": 0.2798072044647387,
      "grad_norm": 5.78125,
      "learning_rate": 9.780646078357346e-06,
      "loss": 1.65467281,
      "memory(GiB)": 97.17,
      "step": 11030,
      "train_speed(iter/s)": 1.627211
    },
    {
      "acc": 0.65384021,
      "epoch": 0.27993404363267377,
      "grad_norm": 5.5,
      "learning_rate": 9.78033878389044e-06,
      "loss": 1.6958107,
      "memory(GiB)": 97.17,
      "step": 11035,
      "train_speed(iter/s)": 1.627297
    },
    {
      "acc": 0.65496006,
      "epoch": 0.2800608828006088,
      "grad_norm": 6.34375,
      "learning_rate": 9.780031279162892e-06,
      "loss": 1.66777267,
      "memory(GiB)": 97.17,
      "step": 11040,
      "train_speed(iter/s)": 1.627372
    },
    {
      "acc": 0.66064839,
      "epoch": 0.2801877219685439,
      "grad_norm": 8.4375,
      "learning_rate": 9.779723564188228e-06,
      "loss": 1.67831955,
      "memory(GiB)": 97.17,
      "step": 11045,
      "train_speed(iter/s)": 1.627458
    },
    {
      "acc": 0.64634714,
      "epoch": 0.28031456113647896,
      "grad_norm": 6.59375,
      "learning_rate": 9.77941563897998e-06,
      "loss": 1.67265205,
      "memory(GiB)": 97.17,
      "step": 11050,
      "train_speed(iter/s)": 1.62754
    },
    {
      "acc": 0.65311151,
      "epoch": 0.280441400304414,
      "grad_norm": 5.125,
      "learning_rate": 9.779107503551695e-06,
      "loss": 1.66472244,
      "memory(GiB)": 97.17,
      "step": 11055,
      "train_speed(iter/s)": 1.627623
    },
    {
      "acc": 0.66278992,
      "epoch": 0.28056823947234905,
      "grad_norm": 6.46875,
      "learning_rate": 9.778799157916926e-06,
      "loss": 1.61254082,
      "memory(GiB)": 97.17,
      "step": 11060,
      "train_speed(iter/s)": 1.627701
    },
    {
      "acc": 0.64478445,
      "epoch": 0.28069507864028415,
      "grad_norm": 4.65625,
      "learning_rate": 9.77849060208923e-06,
      "loss": 1.64988365,
      "memory(GiB)": 97.17,
      "step": 11065,
      "train_speed(iter/s)": 1.627753
    },
    {
      "acc": 0.63668852,
      "epoch": 0.2808219178082192,
      "grad_norm": 4.90625,
      "learning_rate": 9.778181836082185e-06,
      "loss": 1.68410778,
      "memory(GiB)": 97.17,
      "step": 11070,
      "train_speed(iter/s)": 1.627832
    },
    {
      "acc": 0.65798497,
      "epoch": 0.28094875697615423,
      "grad_norm": 6.40625,
      "learning_rate": 9.777872859909373e-06,
      "loss": 1.63516426,
      "memory(GiB)": 97.17,
      "step": 11075,
      "train_speed(iter/s)": 1.62791
    },
    {
      "acc": 0.67312264,
      "epoch": 0.2810755961440893,
      "grad_norm": 6.0,
      "learning_rate": 9.777563673584376e-06,
      "loss": 1.5473176,
      "memory(GiB)": 97.17,
      "step": 11080,
      "train_speed(iter/s)": 1.62799
    },
    {
      "acc": 0.65361595,
      "epoch": 0.2812024353120244,
      "grad_norm": 5.9375,
      "learning_rate": 9.777254277120801e-06,
      "loss": 1.60058556,
      "memory(GiB)": 97.17,
      "step": 11085,
      "train_speed(iter/s)": 1.628067
    },
    {
      "acc": 0.66371036,
      "epoch": 0.2813292744799594,
      "grad_norm": 5.125,
      "learning_rate": 9.776944670532253e-06,
      "loss": 1.62192726,
      "memory(GiB)": 97.17,
      "step": 11090,
      "train_speed(iter/s)": 1.628151
    },
    {
      "acc": 0.65502048,
      "epoch": 0.28145611364789447,
      "grad_norm": 5.96875,
      "learning_rate": 9.776634853832352e-06,
      "loss": 1.6127655,
      "memory(GiB)": 97.17,
      "step": 11095,
      "train_speed(iter/s)": 1.628232
    },
    {
      "acc": 0.65877752,
      "epoch": 0.2815829528158295,
      "grad_norm": 6.21875,
      "learning_rate": 9.776324827034724e-06,
      "loss": 1.62296791,
      "memory(GiB)": 97.17,
      "step": 11100,
      "train_speed(iter/s)": 1.628309
    },
    {
      "acc": 0.64850435,
      "epoch": 0.2817097919837646,
      "grad_norm": 4.78125,
      "learning_rate": 9.776014590153005e-06,
      "loss": 1.64487839,
      "memory(GiB)": 97.17,
      "step": 11105,
      "train_speed(iter/s)": 1.628393
    },
    {
      "acc": 0.66627798,
      "epoch": 0.28183663115169966,
      "grad_norm": 6.1875,
      "learning_rate": 9.77570414320084e-06,
      "loss": 1.59494495,
      "memory(GiB)": 97.17,
      "step": 11110,
      "train_speed(iter/s)": 1.628473
    },
    {
      "acc": 0.65642114,
      "epoch": 0.2819634703196347,
      "grad_norm": 6.03125,
      "learning_rate": 9.775393486191884e-06,
      "loss": 1.61409416,
      "memory(GiB)": 97.17,
      "step": 11115,
      "train_speed(iter/s)": 1.628541
    },
    {
      "acc": 0.64382234,
      "epoch": 0.28209030948756975,
      "grad_norm": 6.5,
      "learning_rate": 9.775082619139805e-06,
      "loss": 1.67371445,
      "memory(GiB)": 97.17,
      "step": 11120,
      "train_speed(iter/s)": 1.628624
    },
    {
      "acc": 0.6633831,
      "epoch": 0.28221714865550485,
      "grad_norm": 6.21875,
      "learning_rate": 9.77477154205827e-06,
      "loss": 1.60559349,
      "memory(GiB)": 97.17,
      "step": 11125,
      "train_speed(iter/s)": 1.628711
    },
    {
      "acc": 0.64146528,
      "epoch": 0.2823439878234399,
      "grad_norm": 5.4375,
      "learning_rate": 9.774460254960968e-06,
      "loss": 1.66024208,
      "memory(GiB)": 97.17,
      "step": 11130,
      "train_speed(iter/s)": 1.628797
    },
    {
      "acc": 0.660462,
      "epoch": 0.28247082699137493,
      "grad_norm": 5.96875,
      "learning_rate": 9.774148757861584e-06,
      "loss": 1.57980824,
      "memory(GiB)": 97.17,
      "step": 11135,
      "train_speed(iter/s)": 1.628877
    },
    {
      "acc": 0.64353399,
      "epoch": 0.28259766615931,
      "grad_norm": 6.40625,
      "learning_rate": 9.773837050773824e-06,
      "loss": 1.6802948,
      "memory(GiB)": 97.17,
      "step": 11140,
      "train_speed(iter/s)": 1.628959
    },
    {
      "acc": 0.63185968,
      "epoch": 0.2827245053272451,
      "grad_norm": 5.8125,
      "learning_rate": 9.773525133711399e-06,
      "loss": 1.69713783,
      "memory(GiB)": 97.17,
      "step": 11145,
      "train_speed(iter/s)": 1.629041
    },
    {
      "acc": 0.65354457,
      "epoch": 0.2828513444951801,
      "grad_norm": 5.09375,
      "learning_rate": 9.773213006688024e-06,
      "loss": 1.68390293,
      "memory(GiB)": 97.17,
      "step": 11150,
      "train_speed(iter/s)": 1.62912
    },
    {
      "acc": 0.63266058,
      "epoch": 0.28297818366311517,
      "grad_norm": 5.9375,
      "learning_rate": 9.77290066971743e-06,
      "loss": 1.74093666,
      "memory(GiB)": 97.17,
      "step": 11155,
      "train_speed(iter/s)": 1.629202
    },
    {
      "acc": 0.64206438,
      "epoch": 0.2831050228310502,
      "grad_norm": 4.75,
      "learning_rate": 9.772588122813358e-06,
      "loss": 1.66956558,
      "memory(GiB)": 97.17,
      "step": 11160,
      "train_speed(iter/s)": 1.629282
    },
    {
      "acc": 0.66281791,
      "epoch": 0.2832318619989853,
      "grad_norm": 6.03125,
      "learning_rate": 9.772275365989548e-06,
      "loss": 1.65986023,
      "memory(GiB)": 97.17,
      "step": 11165,
      "train_speed(iter/s)": 1.629362
    },
    {
      "acc": 0.66160192,
      "epoch": 0.28335870116692036,
      "grad_norm": 5.625,
      "learning_rate": 9.771962399259764e-06,
      "loss": 1.6244503,
      "memory(GiB)": 97.17,
      "step": 11170,
      "train_speed(iter/s)": 1.629443
    },
    {
      "acc": 0.66339049,
      "epoch": 0.2834855403348554,
      "grad_norm": 5.1875,
      "learning_rate": 9.771649222637767e-06,
      "loss": 1.55549812,
      "memory(GiB)": 97.17,
      "step": 11175,
      "train_speed(iter/s)": 1.629523
    },
    {
      "acc": 0.62669697,
      "epoch": 0.28361237950279045,
      "grad_norm": 6.375,
      "learning_rate": 9.771335836137332e-06,
      "loss": 1.69616547,
      "memory(GiB)": 97.17,
      "step": 11180,
      "train_speed(iter/s)": 1.629605
    },
    {
      "acc": 0.65840244,
      "epoch": 0.28373921867072555,
      "grad_norm": 5.28125,
      "learning_rate": 9.771022239772248e-06,
      "loss": 1.5628808,
      "memory(GiB)": 97.17,
      "step": 11185,
      "train_speed(iter/s)": 1.629681
    },
    {
      "acc": 0.64685392,
      "epoch": 0.2838660578386606,
      "grad_norm": 5.84375,
      "learning_rate": 9.770708433556302e-06,
      "loss": 1.69844799,
      "memory(GiB)": 97.17,
      "step": 11190,
      "train_speed(iter/s)": 1.629762
    },
    {
      "acc": 0.64578285,
      "epoch": 0.28399289700659563,
      "grad_norm": 7.0,
      "learning_rate": 9.7703944175033e-06,
      "loss": 1.63361492,
      "memory(GiB)": 97.17,
      "step": 11195,
      "train_speed(iter/s)": 1.629846
    },
    {
      "acc": 0.65484905,
      "epoch": 0.2841197361745307,
      "grad_norm": 5.28125,
      "learning_rate": 9.770080191627054e-06,
      "loss": 1.60732327,
      "memory(GiB)": 97.17,
      "step": 11200,
      "train_speed(iter/s)": 1.629927
    },
    {
      "acc": 0.65470614,
      "epoch": 0.2842465753424658,
      "grad_norm": 4.71875,
      "learning_rate": 9.769765755941383e-06,
      "loss": 1.62965069,
      "memory(GiB)": 97.17,
      "step": 11205,
      "train_speed(iter/s)": 1.630006
    },
    {
      "acc": 0.64792995,
      "epoch": 0.2843734145104008,
      "grad_norm": 5.8125,
      "learning_rate": 9.76945111046012e-06,
      "loss": 1.63702583,
      "memory(GiB)": 97.17,
      "step": 11210,
      "train_speed(iter/s)": 1.630085
    },
    {
      "acc": 0.63567381,
      "epoch": 0.28450025367833587,
      "grad_norm": 5.6875,
      "learning_rate": 9.769136255197103e-06,
      "loss": 1.73173981,
      "memory(GiB)": 97.17,
      "step": 11215,
      "train_speed(iter/s)": 1.630161
    },
    {
      "acc": 0.64610329,
      "epoch": 0.2846270928462709,
      "grad_norm": 6.1875,
      "learning_rate": 9.768821190166179e-06,
      "loss": 1.63170204,
      "memory(GiB)": 97.17,
      "step": 11220,
      "train_speed(iter/s)": 1.63025
    },
    {
      "acc": 0.64216576,
      "epoch": 0.284753932014206,
      "grad_norm": 5.78125,
      "learning_rate": 9.76850591538121e-06,
      "loss": 1.5620615,
      "memory(GiB)": 97.17,
      "step": 11225,
      "train_speed(iter/s)": 1.630327
    },
    {
      "acc": 0.64817505,
      "epoch": 0.28488077118214106,
      "grad_norm": 5.59375,
      "learning_rate": 9.76819043085606e-06,
      "loss": 1.63430252,
      "memory(GiB)": 97.17,
      "step": 11230,
      "train_speed(iter/s)": 1.63041
    },
    {
      "acc": 0.65770745,
      "epoch": 0.2850076103500761,
      "grad_norm": 4.625,
      "learning_rate": 9.767874736604605e-06,
      "loss": 1.58754339,
      "memory(GiB)": 97.17,
      "step": 11235,
      "train_speed(iter/s)": 1.630487
    },
    {
      "acc": 0.66049709,
      "epoch": 0.28513444951801115,
      "grad_norm": 5.5625,
      "learning_rate": 9.767558832640734e-06,
      "loss": 1.55478992,
      "memory(GiB)": 97.17,
      "step": 11240,
      "train_speed(iter/s)": 1.630569
    },
    {
      "acc": 0.6559062,
      "epoch": 0.28526128868594625,
      "grad_norm": 5.53125,
      "learning_rate": 9.76724271897834e-06,
      "loss": 1.56323986,
      "memory(GiB)": 97.17,
      "step": 11245,
      "train_speed(iter/s)": 1.630647
    },
    {
      "acc": 0.6375628,
      "epoch": 0.2853881278538813,
      "grad_norm": 5.625,
      "learning_rate": 9.766926395631326e-06,
      "loss": 1.68773232,
      "memory(GiB)": 97.17,
      "step": 11250,
      "train_speed(iter/s)": 1.630729
    },
    {
      "acc": 0.65749111,
      "epoch": 0.28551496702181633,
      "grad_norm": 6.4375,
      "learning_rate": 9.766609862613607e-06,
      "loss": 1.59407263,
      "memory(GiB)": 97.17,
      "step": 11255,
      "train_speed(iter/s)": 1.630807
    },
    {
      "acc": 0.65684752,
      "epoch": 0.2856418061897514,
      "grad_norm": 5.84375,
      "learning_rate": 9.766293119939104e-06,
      "loss": 1.67577019,
      "memory(GiB)": 97.17,
      "step": 11260,
      "train_speed(iter/s)": 1.630889
    },
    {
      "acc": 0.65764813,
      "epoch": 0.2857686453576865,
      "grad_norm": 5.15625,
      "learning_rate": 9.76597616762175e-06,
      "loss": 1.61688423,
      "memory(GiB)": 97.17,
      "step": 11265,
      "train_speed(iter/s)": 1.63096
    },
    {
      "acc": 0.64011869,
      "epoch": 0.2858954845256215,
      "grad_norm": 4.71875,
      "learning_rate": 9.765659005675488e-06,
      "loss": 1.62728672,
      "memory(GiB)": 97.17,
      "step": 11270,
      "train_speed(iter/s)": 1.631034
    },
    {
      "acc": 0.6329874,
      "epoch": 0.28602232369355657,
      "grad_norm": 5.375,
      "learning_rate": 9.765341634114263e-06,
      "loss": 1.69099216,
      "memory(GiB)": 97.17,
      "step": 11275,
      "train_speed(iter/s)": 1.631112
    },
    {
      "acc": 0.64708767,
      "epoch": 0.2861491628614916,
      "grad_norm": 5.25,
      "learning_rate": 9.765024052952037e-06,
      "loss": 1.65065842,
      "memory(GiB)": 97.17,
      "step": 11280,
      "train_speed(iter/s)": 1.631191
    },
    {
      "acc": 0.64189854,
      "epoch": 0.2862760020294267,
      "grad_norm": 5.0625,
      "learning_rate": 9.76470626220278e-06,
      "loss": 1.69097328,
      "memory(GiB)": 97.17,
      "step": 11285,
      "train_speed(iter/s)": 1.63127
    },
    {
      "acc": 0.64078856,
      "epoch": 0.28640284119736176,
      "grad_norm": 5.5625,
      "learning_rate": 9.76438826188047e-06,
      "loss": 1.65014381,
      "memory(GiB)": 97.17,
      "step": 11290,
      "train_speed(iter/s)": 1.631341
    },
    {
      "acc": 0.6390677,
      "epoch": 0.2865296803652968,
      "grad_norm": 7.28125,
      "learning_rate": 9.76407005199909e-06,
      "loss": 1.68696251,
      "memory(GiB)": 97.17,
      "step": 11295,
      "train_speed(iter/s)": 1.631417
    },
    {
      "acc": 0.64552989,
      "epoch": 0.28665651953323185,
      "grad_norm": 4.4375,
      "learning_rate": 9.76375163257264e-06,
      "loss": 1.61886024,
      "memory(GiB)": 97.17,
      "step": 11300,
      "train_speed(iter/s)": 1.631493
    },
    {
      "acc": 0.65681577,
      "epoch": 0.28678335870116695,
      "grad_norm": 5.5,
      "learning_rate": 9.763433003615124e-06,
      "loss": 1.62201767,
      "memory(GiB)": 97.17,
      "step": 11305,
      "train_speed(iter/s)": 1.631578
    },
    {
      "acc": 0.64618597,
      "epoch": 0.286910197869102,
      "grad_norm": 6.25,
      "learning_rate": 9.763114165140559e-06,
      "loss": 1.72712708,
      "memory(GiB)": 97.17,
      "step": 11310,
      "train_speed(iter/s)": 1.631655
    },
    {
      "acc": 0.63099065,
      "epoch": 0.28703703703703703,
      "grad_norm": 5.28125,
      "learning_rate": 9.762795117162967e-06,
      "loss": 1.69247704,
      "memory(GiB)": 97.17,
      "step": 11315,
      "train_speed(iter/s)": 1.631743
    },
    {
      "acc": 0.66388273,
      "epoch": 0.2871638762049721,
      "grad_norm": 5.875,
      "learning_rate": 9.76247585969638e-06,
      "loss": 1.58248329,
      "memory(GiB)": 97.17,
      "step": 11320,
      "train_speed(iter/s)": 1.631819
    },
    {
      "acc": 0.65471501,
      "epoch": 0.2872907153729072,
      "grad_norm": 5.5625,
      "learning_rate": 9.762156392754842e-06,
      "loss": 1.58320112,
      "memory(GiB)": 97.17,
      "step": 11325,
      "train_speed(iter/s)": 1.631896
    },
    {
      "acc": 0.6289319,
      "epoch": 0.2874175545408422,
      "grad_norm": 5.6875,
      "learning_rate": 9.761836716352405e-06,
      "loss": 1.67280121,
      "memory(GiB)": 97.17,
      "step": 11330,
      "train_speed(iter/s)": 1.631969
    },
    {
      "acc": 0.63756342,
      "epoch": 0.28754439370877727,
      "grad_norm": 5.25,
      "learning_rate": 9.761516830503128e-06,
      "loss": 1.71608143,
      "memory(GiB)": 97.17,
      "step": 11335,
      "train_speed(iter/s)": 1.632053
    },
    {
      "acc": 0.65393553,
      "epoch": 0.2876712328767123,
      "grad_norm": 5.4375,
      "learning_rate": 9.761196735221083e-06,
      "loss": 1.66383781,
      "memory(GiB)": 97.17,
      "step": 11340,
      "train_speed(iter/s)": 1.63213
    },
    {
      "acc": 0.64664373,
      "epoch": 0.2877980720446474,
      "grad_norm": 5.71875,
      "learning_rate": 9.76087643052035e-06,
      "loss": 1.63629303,
      "memory(GiB)": 97.17,
      "step": 11345,
      "train_speed(iter/s)": 1.632205
    },
    {
      "acc": 0.64661798,
      "epoch": 0.28792491121258246,
      "grad_norm": 5.21875,
      "learning_rate": 9.760555916415015e-06,
      "loss": 1.66401596,
      "memory(GiB)": 97.17,
      "step": 11350,
      "train_speed(iter/s)": 1.632279
    },
    {
      "acc": 0.65716586,
      "epoch": 0.2880517503805175,
      "grad_norm": 5.78125,
      "learning_rate": 9.760235192919175e-06,
      "loss": 1.58799686,
      "memory(GiB)": 97.17,
      "step": 11355,
      "train_speed(iter/s)": 1.632355
    },
    {
      "acc": 0.64715643,
      "epoch": 0.28817858954845255,
      "grad_norm": 6.09375,
      "learning_rate": 9.75991426004694e-06,
      "loss": 1.65637703,
      "memory(GiB)": 97.17,
      "step": 11360,
      "train_speed(iter/s)": 1.632434
    },
    {
      "acc": 0.64976225,
      "epoch": 0.28830542871638765,
      "grad_norm": 5.1875,
      "learning_rate": 9.759593117812423e-06,
      "loss": 1.63441086,
      "memory(GiB)": 97.17,
      "step": 11365,
      "train_speed(iter/s)": 1.632505
    },
    {
      "acc": 0.67269187,
      "epoch": 0.2884322678843227,
      "grad_norm": 5.65625,
      "learning_rate": 9.75927176622975e-06,
      "loss": 1.56246796,
      "memory(GiB)": 97.17,
      "step": 11370,
      "train_speed(iter/s)": 1.632582
    },
    {
      "acc": 0.6465395,
      "epoch": 0.28855910705225774,
      "grad_norm": 5.09375,
      "learning_rate": 9.758950205313057e-06,
      "loss": 1.68576775,
      "memory(GiB)": 97.17,
      "step": 11375,
      "train_speed(iter/s)": 1.632654
    },
    {
      "acc": 0.66221781,
      "epoch": 0.2886859462201928,
      "grad_norm": 4.40625,
      "learning_rate": 9.758628435076488e-06,
      "loss": 1.59709435,
      "memory(GiB)": 97.17,
      "step": 11380,
      "train_speed(iter/s)": 1.632726
    },
    {
      "acc": 0.64337535,
      "epoch": 0.2888127853881279,
      "grad_norm": 5.3125,
      "learning_rate": 9.758306455534193e-06,
      "loss": 1.65946236,
      "memory(GiB)": 97.17,
      "step": 11385,
      "train_speed(iter/s)": 1.632798
    },
    {
      "acc": 0.63714933,
      "epoch": 0.2889396245560629,
      "grad_norm": 5.65625,
      "learning_rate": 9.757984266700336e-06,
      "loss": 1.68031082,
      "memory(GiB)": 97.17,
      "step": 11390,
      "train_speed(iter/s)": 1.632868
    },
    {
      "acc": 0.65039325,
      "epoch": 0.28906646372399797,
      "grad_norm": 7.21875,
      "learning_rate": 9.75766186858909e-06,
      "loss": 1.62771111,
      "memory(GiB)": 97.17,
      "step": 11395,
      "train_speed(iter/s)": 1.632943
    },
    {
      "acc": 0.6509644,
      "epoch": 0.289193302891933,
      "grad_norm": 5.125,
      "learning_rate": 9.757339261214631e-06,
      "loss": 1.69709454,
      "memory(GiB)": 97.17,
      "step": 11400,
      "train_speed(iter/s)": 1.633024
    },
    {
      "acc": 0.64038453,
      "epoch": 0.2893201420598681,
      "grad_norm": 5.4375,
      "learning_rate": 9.757016444591152e-06,
      "loss": 1.64617481,
      "memory(GiB)": 97.17,
      "step": 11405,
      "train_speed(iter/s)": 1.633097
    },
    {
      "acc": 0.65103588,
      "epoch": 0.28944698122780316,
      "grad_norm": 4.6875,
      "learning_rate": 9.756693418732852e-06,
      "loss": 1.64224205,
      "memory(GiB)": 97.17,
      "step": 11410,
      "train_speed(iter/s)": 1.633172
    },
    {
      "acc": 0.65074749,
      "epoch": 0.2895738203957382,
      "grad_norm": 5.65625,
      "learning_rate": 9.756370183653938e-06,
      "loss": 1.7003973,
      "memory(GiB)": 97.17,
      "step": 11415,
      "train_speed(iter/s)": 1.633249
    },
    {
      "acc": 0.64227662,
      "epoch": 0.28970065956367325,
      "grad_norm": 6.25,
      "learning_rate": 9.756046739368628e-06,
      "loss": 1.63268166,
      "memory(GiB)": 97.17,
      "step": 11420,
      "train_speed(iter/s)": 1.633334
    },
    {
      "acc": 0.63798075,
      "epoch": 0.28982749873160835,
      "grad_norm": 6.1875,
      "learning_rate": 9.755723085891147e-06,
      "loss": 1.68317146,
      "memory(GiB)": 97.17,
      "step": 11425,
      "train_speed(iter/s)": 1.633415
    },
    {
      "acc": 0.66065168,
      "epoch": 0.2899543378995434,
      "grad_norm": 5.8125,
      "learning_rate": 9.755399223235734e-06,
      "loss": 1.58074322,
      "memory(GiB)": 97.17,
      "step": 11430,
      "train_speed(iter/s)": 1.633486
    },
    {
      "acc": 0.64656153,
      "epoch": 0.29008117706747844,
      "grad_norm": 6.90625,
      "learning_rate": 9.75507515141663e-06,
      "loss": 1.64568062,
      "memory(GiB)": 97.17,
      "step": 11435,
      "train_speed(iter/s)": 1.633562
    },
    {
      "acc": 0.64046583,
      "epoch": 0.2902080162354135,
      "grad_norm": 6.28125,
      "learning_rate": 9.75475087044809e-06,
      "loss": 1.64064331,
      "memory(GiB)": 97.17,
      "step": 11440,
      "train_speed(iter/s)": 1.633635
    },
    {
      "acc": 0.64787884,
      "epoch": 0.2903348554033486,
      "grad_norm": 5.625,
      "learning_rate": 9.754426380344382e-06,
      "loss": 1.67387772,
      "memory(GiB)": 97.17,
      "step": 11445,
      "train_speed(iter/s)": 1.633712
    },
    {
      "acc": 0.64361849,
      "epoch": 0.2904616945712836,
      "grad_norm": 5.3125,
      "learning_rate": 9.754101681119772e-06,
      "loss": 1.71696663,
      "memory(GiB)": 97.17,
      "step": 11450,
      "train_speed(iter/s)": 1.633796
    },
    {
      "acc": 0.63113894,
      "epoch": 0.29058853373921867,
      "grad_norm": 6.28125,
      "learning_rate": 9.753776772788545e-06,
      "loss": 1.69180794,
      "memory(GiB)": 97.17,
      "step": 11455,
      "train_speed(iter/s)": 1.63388
    },
    {
      "acc": 0.64941807,
      "epoch": 0.2907153729071537,
      "grad_norm": 4.875,
      "learning_rate": 9.753451655364992e-06,
      "loss": 1.64529076,
      "memory(GiB)": 97.17,
      "step": 11460,
      "train_speed(iter/s)": 1.63396
    },
    {
      "acc": 0.6493063,
      "epoch": 0.2908422120750888,
      "grad_norm": 5.65625,
      "learning_rate": 9.75312632886341e-06,
      "loss": 1.71305027,
      "memory(GiB)": 97.17,
      "step": 11465,
      "train_speed(iter/s)": 1.634035
    },
    {
      "acc": 0.67065487,
      "epoch": 0.29096905124302386,
      "grad_norm": 5.0,
      "learning_rate": 9.752800793298113e-06,
      "loss": 1.56441851,
      "memory(GiB)": 97.17,
      "step": 11470,
      "train_speed(iter/s)": 1.634108
    },
    {
      "acc": 0.6549777,
      "epoch": 0.2910958904109589,
      "grad_norm": 5.59375,
      "learning_rate": 9.752475048683419e-06,
      "loss": 1.62848644,
      "memory(GiB)": 97.17,
      "step": 11475,
      "train_speed(iter/s)": 1.634183
    },
    {
      "acc": 0.65016327,
      "epoch": 0.29122272957889395,
      "grad_norm": 5.28125,
      "learning_rate": 9.752149095033651e-06,
      "loss": 1.61026001,
      "memory(GiB)": 97.17,
      "step": 11480,
      "train_speed(iter/s)": 1.634257
    },
    {
      "acc": 0.65454712,
      "epoch": 0.29134956874682905,
      "grad_norm": 6.375,
      "learning_rate": 9.75182293236315e-06,
      "loss": 1.61261635,
      "memory(GiB)": 97.17,
      "step": 11485,
      "train_speed(iter/s)": 1.634337
    },
    {
      "acc": 0.64424939,
      "epoch": 0.2914764079147641,
      "grad_norm": 6.34375,
      "learning_rate": 9.751496560686262e-06,
      "loss": 1.56798515,
      "memory(GiB)": 97.17,
      "step": 11490,
      "train_speed(iter/s)": 1.634417
    },
    {
      "acc": 0.65181317,
      "epoch": 0.29160324708269914,
      "grad_norm": 5.40625,
      "learning_rate": 9.751169980017341e-06,
      "loss": 1.61228981,
      "memory(GiB)": 97.17,
      "step": 11495,
      "train_speed(iter/s)": 1.634489
    },
    {
      "acc": 0.65896001,
      "epoch": 0.2917300862506342,
      "grad_norm": 4.9375,
      "learning_rate": 9.750843190370752e-06,
      "loss": 1.59504261,
      "memory(GiB)": 97.17,
      "step": 11500,
      "train_speed(iter/s)": 1.634557
    },
    {
      "acc": 0.67066407,
      "epoch": 0.2918569254185693,
      "grad_norm": 4.125,
      "learning_rate": 9.750516191760868e-06,
      "loss": 1.57905722,
      "memory(GiB)": 97.17,
      "step": 11505,
      "train_speed(iter/s)": 1.634633
    },
    {
      "acc": 0.64843054,
      "epoch": 0.2919837645865043,
      "grad_norm": 4.375,
      "learning_rate": 9.750188984202073e-06,
      "loss": 1.62277679,
      "memory(GiB)": 97.17,
      "step": 11510,
      "train_speed(iter/s)": 1.634711
    },
    {
      "acc": 0.6408927,
      "epoch": 0.29211060375443937,
      "grad_norm": 5.875,
      "learning_rate": 9.749861567708759e-06,
      "loss": 1.71921844,
      "memory(GiB)": 97.17,
      "step": 11515,
      "train_speed(iter/s)": 1.634789
    },
    {
      "acc": 0.6344408,
      "epoch": 0.2922374429223744,
      "grad_norm": 6.1875,
      "learning_rate": 9.749533942295323e-06,
      "loss": 1.68303127,
      "memory(GiB)": 97.17,
      "step": 11520,
      "train_speed(iter/s)": 1.634865
    },
    {
      "acc": 0.66041121,
      "epoch": 0.2923642820903095,
      "grad_norm": 6.78125,
      "learning_rate": 9.749206107976183e-06,
      "loss": 1.59183712,
      "memory(GiB)": 97.17,
      "step": 11525,
      "train_speed(iter/s)": 1.634947
    },
    {
      "acc": 0.64058123,
      "epoch": 0.29249112125824456,
      "grad_norm": 6.0,
      "learning_rate": 9.748878064765753e-06,
      "loss": 1.66036873,
      "memory(GiB)": 97.17,
      "step": 11530,
      "train_speed(iter/s)": 1.635021
    },
    {
      "acc": 0.64665055,
      "epoch": 0.2926179604261796,
      "grad_norm": 6.78125,
      "learning_rate": 9.748549812678466e-06,
      "loss": 1.68230362,
      "memory(GiB)": 97.17,
      "step": 11535,
      "train_speed(iter/s)": 1.635089
    },
    {
      "acc": 0.65084305,
      "epoch": 0.29274479959411465,
      "grad_norm": 5.125,
      "learning_rate": 9.748221351728754e-06,
      "loss": 1.68193531,
      "memory(GiB)": 97.17,
      "step": 11540,
      "train_speed(iter/s)": 1.635162
    },
    {
      "acc": 0.65032053,
      "epoch": 0.29287163876204975,
      "grad_norm": 6.0625,
      "learning_rate": 9.747892681931067e-06,
      "loss": 1.62254906,
      "memory(GiB)": 97.17,
      "step": 11545,
      "train_speed(iter/s)": 1.635233
    },
    {
      "acc": 0.64764028,
      "epoch": 0.2929984779299848,
      "grad_norm": 5.78125,
      "learning_rate": 9.747563803299865e-06,
      "loss": 1.63244286,
      "memory(GiB)": 97.17,
      "step": 11550,
      "train_speed(iter/s)": 1.635312
    },
    {
      "acc": 0.65060344,
      "epoch": 0.29312531709791984,
      "grad_norm": 5.5625,
      "learning_rate": 9.74723471584961e-06,
      "loss": 1.62937889,
      "memory(GiB)": 97.17,
      "step": 11555,
      "train_speed(iter/s)": 1.635386
    },
    {
      "acc": 0.65269713,
      "epoch": 0.2932521562658549,
      "grad_norm": 5.78125,
      "learning_rate": 9.746905419594777e-06,
      "loss": 1.59821739,
      "memory(GiB)": 97.17,
      "step": 11560,
      "train_speed(iter/s)": 1.635467
    },
    {
      "acc": 0.64498816,
      "epoch": 0.29337899543379,
      "grad_norm": 5.6875,
      "learning_rate": 9.746575914549851e-06,
      "loss": 1.63843441,
      "memory(GiB)": 97.17,
      "step": 11565,
      "train_speed(iter/s)": 1.635545
    },
    {
      "acc": 0.62795815,
      "epoch": 0.293505834601725,
      "grad_norm": 5.375,
      "learning_rate": 9.746246200729323e-06,
      "loss": 1.74897957,
      "memory(GiB)": 97.17,
      "step": 11570,
      "train_speed(iter/s)": 1.635617
    },
    {
      "acc": 0.65472836,
      "epoch": 0.29363267376966007,
      "grad_norm": 5.15625,
      "learning_rate": 9.745916278147696e-06,
      "loss": 1.61250858,
      "memory(GiB)": 97.17,
      "step": 11575,
      "train_speed(iter/s)": 1.635691
    },
    {
      "acc": 0.65050497,
      "epoch": 0.2937595129375951,
      "grad_norm": 7.46875,
      "learning_rate": 9.745586146819484e-06,
      "loss": 1.64993591,
      "memory(GiB)": 97.17,
      "step": 11580,
      "train_speed(iter/s)": 1.635762
    },
    {
      "acc": 0.64963841,
      "epoch": 0.2938863521055302,
      "grad_norm": 4.90625,
      "learning_rate": 9.745255806759205e-06,
      "loss": 1.72799568,
      "memory(GiB)": 97.17,
      "step": 11585,
      "train_speed(iter/s)": 1.635835
    },
    {
      "acc": 0.64927502,
      "epoch": 0.29401319127346526,
      "grad_norm": 6.125,
      "learning_rate": 9.74492525798139e-06,
      "loss": 1.67259102,
      "memory(GiB)": 97.17,
      "step": 11590,
      "train_speed(iter/s)": 1.635911
    },
    {
      "acc": 0.63580432,
      "epoch": 0.2941400304414003,
      "grad_norm": 5.15625,
      "learning_rate": 9.744594500500578e-06,
      "loss": 1.67914238,
      "memory(GiB)": 97.17,
      "step": 11595,
      "train_speed(iter/s)": 1.635988
    },
    {
      "acc": 0.64402461,
      "epoch": 0.29426686960933535,
      "grad_norm": 6.53125,
      "learning_rate": 9.744263534331315e-06,
      "loss": 1.61763115,
      "memory(GiB)": 97.17,
      "step": 11600,
      "train_speed(iter/s)": 1.636069
    },
    {
      "acc": 0.65339947,
      "epoch": 0.29439370877727045,
      "grad_norm": 5.6875,
      "learning_rate": 9.743932359488161e-06,
      "loss": 1.63013649,
      "memory(GiB)": 97.17,
      "step": 11605,
      "train_speed(iter/s)": 1.636144
    },
    {
      "acc": 0.64586911,
      "epoch": 0.2945205479452055,
      "grad_norm": 4.6875,
      "learning_rate": 9.743600975985681e-06,
      "loss": 1.6654829,
      "memory(GiB)": 97.17,
      "step": 11610,
      "train_speed(iter/s)": 1.636221
    },
    {
      "acc": 0.65215054,
      "epoch": 0.29464738711314054,
      "grad_norm": 5.53125,
      "learning_rate": 9.743269383838452e-06,
      "loss": 1.66870823,
      "memory(GiB)": 97.17,
      "step": 11615,
      "train_speed(iter/s)": 1.636295
    },
    {
      "acc": 0.65218196,
      "epoch": 0.2947742262810756,
      "grad_norm": 6.15625,
      "learning_rate": 9.74293758306106e-06,
      "loss": 1.630896,
      "memory(GiB)": 97.17,
      "step": 11620,
      "train_speed(iter/s)": 1.63637
    },
    {
      "acc": 0.627877,
      "epoch": 0.2949010654490107,
      "grad_norm": 8.0625,
      "learning_rate": 9.742605573668096e-06,
      "loss": 1.68542633,
      "memory(GiB)": 97.17,
      "step": 11625,
      "train_speed(iter/s)": 1.63645
    },
    {
      "acc": 0.64570208,
      "epoch": 0.2950279046169457,
      "grad_norm": 7.1875,
      "learning_rate": 9.742273355674164e-06,
      "loss": 1.68923283,
      "memory(GiB)": 97.17,
      "step": 11630,
      "train_speed(iter/s)": 1.636527
    },
    {
      "acc": 0.66054845,
      "epoch": 0.29515474378488077,
      "grad_norm": 5.1875,
      "learning_rate": 9.741940929093879e-06,
      "loss": 1.63594322,
      "memory(GiB)": 97.17,
      "step": 11635,
      "train_speed(iter/s)": 1.636598
    },
    {
      "acc": 0.65538964,
      "epoch": 0.2952815829528158,
      "grad_norm": 5.46875,
      "learning_rate": 9.741608293941858e-06,
      "loss": 1.60515289,
      "memory(GiB)": 97.17,
      "step": 11640,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.65831141,
      "epoch": 0.2954084221207509,
      "grad_norm": 5.625,
      "learning_rate": 9.741275450232736e-06,
      "loss": 1.63779564,
      "memory(GiB)": 97.17,
      "step": 11645,
      "train_speed(iter/s)": 1.636746
    },
    {
      "acc": 0.6579608,
      "epoch": 0.29553526128868596,
      "grad_norm": 5.53125,
      "learning_rate": 9.740942397981151e-06,
      "loss": 1.55092669,
      "memory(GiB)": 97.17,
      "step": 11650,
      "train_speed(iter/s)": 1.636819
    },
    {
      "acc": 0.649158,
      "epoch": 0.295662100456621,
      "grad_norm": 5.25,
      "learning_rate": 9.740609137201752e-06,
      "loss": 1.6361805,
      "memory(GiB)": 97.17,
      "step": 11655,
      "train_speed(iter/s)": 1.636895
    },
    {
      "acc": 0.64914727,
      "epoch": 0.29578893962455605,
      "grad_norm": 5.4375,
      "learning_rate": 9.7402756679092e-06,
      "loss": 1.64583282,
      "memory(GiB)": 97.17,
      "step": 11660,
      "train_speed(iter/s)": 1.636966
    },
    {
      "acc": 0.64234037,
      "epoch": 0.29591577879249115,
      "grad_norm": 5.71875,
      "learning_rate": 9.739941990118157e-06,
      "loss": 1.60633354,
      "memory(GiB)": 97.17,
      "step": 11665,
      "train_speed(iter/s)": 1.637037
    },
    {
      "acc": 0.65490937,
      "epoch": 0.2960426179604262,
      "grad_norm": 4.8125,
      "learning_rate": 9.739608103843306e-06,
      "loss": 1.63323059,
      "memory(GiB)": 97.17,
      "step": 11670,
      "train_speed(iter/s)": 1.637105
    },
    {
      "acc": 0.65202107,
      "epoch": 0.29616945712836124,
      "grad_norm": 4.875,
      "learning_rate": 9.739274009099328e-06,
      "loss": 1.64343796,
      "memory(GiB)": 97.17,
      "step": 11675,
      "train_speed(iter/s)": 1.637178
    },
    {
      "acc": 0.66397929,
      "epoch": 0.2962962962962963,
      "grad_norm": 5.53125,
      "learning_rate": 9.738939705900922e-06,
      "loss": 1.62341652,
      "memory(GiB)": 97.17,
      "step": 11680,
      "train_speed(iter/s)": 1.637258
    },
    {
      "acc": 0.6575644,
      "epoch": 0.2964231354642314,
      "grad_norm": 4.75,
      "learning_rate": 9.738605194262787e-06,
      "loss": 1.59835129,
      "memory(GiB)": 97.17,
      "step": 11685,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.649084,
      "epoch": 0.2965499746321664,
      "grad_norm": 4.78125,
      "learning_rate": 9.738270474199641e-06,
      "loss": 1.62513294,
      "memory(GiB)": 97.17,
      "step": 11690,
      "train_speed(iter/s)": 1.637407
    },
    {
      "acc": 0.63300428,
      "epoch": 0.29667681380010147,
      "grad_norm": 6.875,
      "learning_rate": 9.737935545726205e-06,
      "loss": 1.69159355,
      "memory(GiB)": 97.17,
      "step": 11695,
      "train_speed(iter/s)": 1.637488
    },
    {
      "acc": 0.65190229,
      "epoch": 0.2968036529680365,
      "grad_norm": 5.75,
      "learning_rate": 9.737600408857208e-06,
      "loss": 1.63336754,
      "memory(GiB)": 97.17,
      "step": 11700,
      "train_speed(iter/s)": 1.637569
    },
    {
      "acc": 0.63793454,
      "epoch": 0.2969304921359716,
      "grad_norm": 6.09375,
      "learning_rate": 9.737265063607395e-06,
      "loss": 1.6612709,
      "memory(GiB)": 97.17,
      "step": 11705,
      "train_speed(iter/s)": 1.637641
    },
    {
      "acc": 0.6532095,
      "epoch": 0.29705733130390666,
      "grad_norm": 5.875,
      "learning_rate": 9.736929509991515e-06,
      "loss": 1.59731693,
      "memory(GiB)": 97.17,
      "step": 11710,
      "train_speed(iter/s)": 1.637714
    },
    {
      "acc": 0.65627689,
      "epoch": 0.2971841704718417,
      "grad_norm": 6.0625,
      "learning_rate": 9.736593748024325e-06,
      "loss": 1.65486469,
      "memory(GiB)": 97.17,
      "step": 11715,
      "train_speed(iter/s)": 1.637797
    },
    {
      "acc": 0.65927405,
      "epoch": 0.29731100963977675,
      "grad_norm": 4.875,
      "learning_rate": 9.736257777720595e-06,
      "loss": 1.55941372,
      "memory(GiB)": 97.17,
      "step": 11720,
      "train_speed(iter/s)": 1.637872
    },
    {
      "acc": 0.6624898,
      "epoch": 0.29743784880771185,
      "grad_norm": 7.21875,
      "learning_rate": 9.735921599095101e-06,
      "loss": 1.57044315,
      "memory(GiB)": 97.17,
      "step": 11725,
      "train_speed(iter/s)": 1.637951
    },
    {
      "acc": 0.6622478,
      "epoch": 0.2975646879756469,
      "grad_norm": 5.09375,
      "learning_rate": 9.735585212162633e-06,
      "loss": 1.58110027,
      "memory(GiB)": 97.17,
      "step": 11730,
      "train_speed(iter/s)": 1.638026
    },
    {
      "acc": 0.64733858,
      "epoch": 0.29769152714358194,
      "grad_norm": 6.40625,
      "learning_rate": 9.735248616937983e-06,
      "loss": 1.66352272,
      "memory(GiB)": 97.17,
      "step": 11735,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.65417871,
      "epoch": 0.297818366311517,
      "grad_norm": 5.5625,
      "learning_rate": 9.734911813435957e-06,
      "loss": 1.59013805,
      "memory(GiB)": 97.17,
      "step": 11740,
      "train_speed(iter/s)": 1.638175
    },
    {
      "acc": 0.65507183,
      "epoch": 0.2979452054794521,
      "grad_norm": 4.90625,
      "learning_rate": 9.73457480167137e-06,
      "loss": 1.74874783,
      "memory(GiB)": 97.17,
      "step": 11745,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.66042709,
      "epoch": 0.2980720446473871,
      "grad_norm": 5.6875,
      "learning_rate": 9.734237581659045e-06,
      "loss": 1.63711433,
      "memory(GiB)": 97.17,
      "step": 11750,
      "train_speed(iter/s)": 1.63833
    },
    {
      "acc": 0.64553061,
      "epoch": 0.29819888381532217,
      "grad_norm": 5.53125,
      "learning_rate": 9.733900153413813e-06,
      "loss": 1.62274132,
      "memory(GiB)": 97.17,
      "step": 11755,
      "train_speed(iter/s)": 1.638408
    },
    {
      "acc": 0.65123997,
      "epoch": 0.2983257229832572,
      "grad_norm": 6.40625,
      "learning_rate": 9.733562516950519e-06,
      "loss": 1.61130638,
      "memory(GiB)": 97.17,
      "step": 11760,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.63914804,
      "epoch": 0.2984525621511923,
      "grad_norm": 6.96875,
      "learning_rate": 9.73322467228401e-06,
      "loss": 1.64715347,
      "memory(GiB)": 97.17,
      "step": 11765,
      "train_speed(iter/s)": 1.638558
    },
    {
      "acc": 0.64855762,
      "epoch": 0.29857940131912736,
      "grad_norm": 7.1875,
      "learning_rate": 9.73288661942915e-06,
      "loss": 1.62235718,
      "memory(GiB)": 97.17,
      "step": 11770,
      "train_speed(iter/s)": 1.638634
    },
    {
      "acc": 0.65780964,
      "epoch": 0.2987062404870624,
      "grad_norm": 6.15625,
      "learning_rate": 9.732548358400802e-06,
      "loss": 1.60995064,
      "memory(GiB)": 97.17,
      "step": 11775,
      "train_speed(iter/s)": 1.638707
    },
    {
      "acc": 0.652736,
      "epoch": 0.29883307965499745,
      "grad_norm": 4.84375,
      "learning_rate": 9.73220988921385e-06,
      "loss": 1.59062061,
      "memory(GiB)": 97.17,
      "step": 11780,
      "train_speed(iter/s)": 1.638777
    },
    {
      "acc": 0.65027757,
      "epoch": 0.29895991882293255,
      "grad_norm": 5.46875,
      "learning_rate": 9.73187121188318e-06,
      "loss": 1.60554485,
      "memory(GiB)": 97.17,
      "step": 11785,
      "train_speed(iter/s)": 1.638853
    },
    {
      "acc": 0.66155748,
      "epoch": 0.2990867579908676,
      "grad_norm": 7.03125,
      "learning_rate": 9.731532326423686e-06,
      "loss": 1.60177994,
      "memory(GiB)": 97.17,
      "step": 11790,
      "train_speed(iter/s)": 1.638928
    },
    {
      "acc": 0.66709805,
      "epoch": 0.29921359715880264,
      "grad_norm": 4.8125,
      "learning_rate": 9.731193232850277e-06,
      "loss": 1.56361628,
      "memory(GiB)": 97.17,
      "step": 11795,
      "train_speed(iter/s)": 1.639002
    },
    {
      "acc": 0.64233665,
      "epoch": 0.2993404363267377,
      "grad_norm": 5.03125,
      "learning_rate": 9.730853931177866e-06,
      "loss": 1.66935577,
      "memory(GiB)": 97.17,
      "step": 11800,
      "train_speed(iter/s)": 1.639074
    },
    {
      "acc": 0.65680666,
      "epoch": 0.2994672754946728,
      "grad_norm": 6.40625,
      "learning_rate": 9.730514421421378e-06,
      "loss": 1.60994625,
      "memory(GiB)": 97.17,
      "step": 11805,
      "train_speed(iter/s)": 1.63915
    },
    {
      "acc": 0.64663,
      "epoch": 0.2995941146626078,
      "grad_norm": 4.78125,
      "learning_rate": 9.730174703595745e-06,
      "loss": 1.64971447,
      "memory(GiB)": 97.17,
      "step": 11810,
      "train_speed(iter/s)": 1.639216
    },
    {
      "acc": 0.65689802,
      "epoch": 0.29972095383054287,
      "grad_norm": 4.9375,
      "learning_rate": 9.72983477771591e-06,
      "loss": 1.57845268,
      "memory(GiB)": 97.17,
      "step": 11815,
      "train_speed(iter/s)": 1.639285
    },
    {
      "acc": 0.6412138,
      "epoch": 0.2998477929984779,
      "grad_norm": 4.90625,
      "learning_rate": 9.729494643796823e-06,
      "loss": 1.68806343,
      "memory(GiB)": 97.17,
      "step": 11820,
      "train_speed(iter/s)": 1.639357
    },
    {
      "acc": 0.65105066,
      "epoch": 0.299974632166413,
      "grad_norm": 5.25,
      "learning_rate": 9.729154301853448e-06,
      "loss": 1.62693596,
      "memory(GiB)": 97.17,
      "step": 11825,
      "train_speed(iter/s)": 1.639429
    },
    {
      "acc": 0.6450696,
      "epoch": 0.30010147133434806,
      "grad_norm": 7.71875,
      "learning_rate": 9.72881375190075e-06,
      "loss": 1.68029881,
      "memory(GiB)": 97.17,
      "step": 11830,
      "train_speed(iter/s)": 1.639502
    },
    {
      "acc": 0.66626835,
      "epoch": 0.3002283105022831,
      "grad_norm": 4.65625,
      "learning_rate": 9.728472993953712e-06,
      "loss": 1.60636997,
      "memory(GiB)": 97.17,
      "step": 11835,
      "train_speed(iter/s)": 1.639571
    },
    {
      "acc": 0.64302607,
      "epoch": 0.30035514967021815,
      "grad_norm": 4.96875,
      "learning_rate": 9.728132028027323e-06,
      "loss": 1.58945427,
      "memory(GiB)": 97.17,
      "step": 11840,
      "train_speed(iter/s)": 1.639646
    },
    {
      "acc": 0.67251196,
      "epoch": 0.30048198883815325,
      "grad_norm": 4.875,
      "learning_rate": 9.727790854136573e-06,
      "loss": 1.62307892,
      "memory(GiB)": 97.17,
      "step": 11845,
      "train_speed(iter/s)": 1.639711
    },
    {
      "acc": 0.65551519,
      "epoch": 0.3006088280060883,
      "grad_norm": 4.6875,
      "learning_rate": 9.727449472296476e-06,
      "loss": 1.57311468,
      "memory(GiB)": 97.17,
      "step": 11850,
      "train_speed(iter/s)": 1.63978
    },
    {
      "acc": 0.64782877,
      "epoch": 0.30073566717402334,
      "grad_norm": 6.46875,
      "learning_rate": 9.727107882522045e-06,
      "loss": 1.67652302,
      "memory(GiB)": 97.17,
      "step": 11855,
      "train_speed(iter/s)": 1.639856
    },
    {
      "acc": 0.66182756,
      "epoch": 0.3008625063419584,
      "grad_norm": 5.65625,
      "learning_rate": 9.726766084828303e-06,
      "loss": 1.60665207,
      "memory(GiB)": 97.17,
      "step": 11860,
      "train_speed(iter/s)": 1.639926
    },
    {
      "acc": 0.64263477,
      "epoch": 0.3009893455098935,
      "grad_norm": 5.3125,
      "learning_rate": 9.726424079230286e-06,
      "loss": 1.62117882,
      "memory(GiB)": 97.17,
      "step": 11865,
      "train_speed(iter/s)": 1.639989
    },
    {
      "acc": 0.65106149,
      "epoch": 0.3011161846778285,
      "grad_norm": 5.46875,
      "learning_rate": 9.726081865743036e-06,
      "loss": 1.63567924,
      "memory(GiB)": 97.17,
      "step": 11870,
      "train_speed(iter/s)": 1.640061
    },
    {
      "acc": 0.67203779,
      "epoch": 0.30124302384576357,
      "grad_norm": 5.09375,
      "learning_rate": 9.725739444381603e-06,
      "loss": 1.60356159,
      "memory(GiB)": 97.17,
      "step": 11875,
      "train_speed(iter/s)": 1.640138
    },
    {
      "acc": 0.65460844,
      "epoch": 0.3013698630136986,
      "grad_norm": 5.125,
      "learning_rate": 9.725396815161053e-06,
      "loss": 1.65225525,
      "memory(GiB)": 97.17,
      "step": 11880,
      "train_speed(iter/s)": 1.640207
    },
    {
      "acc": 0.64531651,
      "epoch": 0.3014967021816337,
      "grad_norm": 5.40625,
      "learning_rate": 9.725053978096453e-06,
      "loss": 1.67289333,
      "memory(GiB)": 97.17,
      "step": 11885,
      "train_speed(iter/s)": 1.640271
    },
    {
      "acc": 0.6419682,
      "epoch": 0.30162354134956876,
      "grad_norm": 5.15625,
      "learning_rate": 9.724710933202884e-06,
      "loss": 1.71273708,
      "memory(GiB)": 97.17,
      "step": 11890,
      "train_speed(iter/s)": 1.640338
    },
    {
      "acc": 0.65387602,
      "epoch": 0.3017503805175038,
      "grad_norm": 5.71875,
      "learning_rate": 9.724367680495432e-06,
      "loss": 1.64997559,
      "memory(GiB)": 97.17,
      "step": 11895,
      "train_speed(iter/s)": 1.64041
    },
    {
      "acc": 0.662253,
      "epoch": 0.30187721968543885,
      "grad_norm": 5.0625,
      "learning_rate": 9.724024219989198e-06,
      "loss": 1.58682127,
      "memory(GiB)": 97.17,
      "step": 11900,
      "train_speed(iter/s)": 1.640479
    },
    {
      "acc": 0.64213529,
      "epoch": 0.30200405885337395,
      "grad_norm": 5.90625,
      "learning_rate": 9.723680551699286e-06,
      "loss": 1.63422241,
      "memory(GiB)": 97.17,
      "step": 11905,
      "train_speed(iter/s)": 1.640547
    },
    {
      "acc": 0.63804383,
      "epoch": 0.302130898021309,
      "grad_norm": 6.3125,
      "learning_rate": 9.723336675640815e-06,
      "loss": 1.740905,
      "memory(GiB)": 97.17,
      "step": 11910,
      "train_speed(iter/s)": 1.640617
    },
    {
      "acc": 0.64859419,
      "epoch": 0.30225773718924404,
      "grad_norm": 5.40625,
      "learning_rate": 9.722992591828908e-06,
      "loss": 1.62869473,
      "memory(GiB)": 97.17,
      "step": 11915,
      "train_speed(iter/s)": 1.640691
    },
    {
      "acc": 0.66343718,
      "epoch": 0.3023845763571791,
      "grad_norm": 5.8125,
      "learning_rate": 9.722648300278701e-06,
      "loss": 1.54543724,
      "memory(GiB)": 97.17,
      "step": 11920,
      "train_speed(iter/s)": 1.640763
    },
    {
      "acc": 0.66775374,
      "epoch": 0.3025114155251142,
      "grad_norm": 5.59375,
      "learning_rate": 9.722303801005338e-06,
      "loss": 1.62005329,
      "memory(GiB)": 97.17,
      "step": 11925,
      "train_speed(iter/s)": 1.640829
    },
    {
      "acc": 0.65400538,
      "epoch": 0.3026382546930492,
      "grad_norm": 6.28125,
      "learning_rate": 9.721959094023968e-06,
      "loss": 1.68701725,
      "memory(GiB)": 97.17,
      "step": 11930,
      "train_speed(iter/s)": 1.6409
    },
    {
      "acc": 0.67039242,
      "epoch": 0.30276509386098427,
      "grad_norm": 6.1875,
      "learning_rate": 9.721614179349754e-06,
      "loss": 1.54521427,
      "memory(GiB)": 97.17,
      "step": 11935,
      "train_speed(iter/s)": 1.640972
    },
    {
      "acc": 0.66024427,
      "epoch": 0.3028919330289193,
      "grad_norm": 6.03125,
      "learning_rate": 9.72126905699787e-06,
      "loss": 1.66078053,
      "memory(GiB)": 97.17,
      "step": 11940,
      "train_speed(iter/s)": 1.64105
    },
    {
      "acc": 0.64067478,
      "epoch": 0.3030187721968544,
      "grad_norm": 4.90625,
      "learning_rate": 9.720923726983493e-06,
      "loss": 1.62818756,
      "memory(GiB)": 97.17,
      "step": 11945,
      "train_speed(iter/s)": 1.641124
    },
    {
      "acc": 0.64188647,
      "epoch": 0.30314561136478946,
      "grad_norm": 5.0,
      "learning_rate": 9.720578189321814e-06,
      "loss": 1.63102531,
      "memory(GiB)": 97.17,
      "step": 11950,
      "train_speed(iter/s)": 1.641192
    },
    {
      "acc": 0.66136594,
      "epoch": 0.3032724505327245,
      "grad_norm": 6.28125,
      "learning_rate": 9.72023244402803e-06,
      "loss": 1.61931629,
      "memory(GiB)": 97.17,
      "step": 11955,
      "train_speed(iter/s)": 1.64126
    },
    {
      "acc": 0.6492928,
      "epoch": 0.30339928970065955,
      "grad_norm": 5.3125,
      "learning_rate": 9.719886491117348e-06,
      "loss": 1.62725468,
      "memory(GiB)": 97.17,
      "step": 11960,
      "train_speed(iter/s)": 1.641334
    },
    {
      "acc": 0.64850187,
      "epoch": 0.30352612886859465,
      "grad_norm": 6.25,
      "learning_rate": 9.719540330604986e-06,
      "loss": 1.58523388,
      "memory(GiB)": 97.17,
      "step": 11965,
      "train_speed(iter/s)": 1.641401
    },
    {
      "acc": 0.65585322,
      "epoch": 0.3036529680365297,
      "grad_norm": 5.90625,
      "learning_rate": 9.71919396250617e-06,
      "loss": 1.6394001,
      "memory(GiB)": 97.17,
      "step": 11970,
      "train_speed(iter/s)": 1.641473
    },
    {
      "acc": 0.62977915,
      "epoch": 0.30377980720446474,
      "grad_norm": 5.3125,
      "learning_rate": 9.718847386836131e-06,
      "loss": 1.70795479,
      "memory(GiB)": 97.17,
      "step": 11975,
      "train_speed(iter/s)": 1.641546
    },
    {
      "acc": 0.65073948,
      "epoch": 0.3039066463723998,
      "grad_norm": 5.34375,
      "learning_rate": 9.718500603610119e-06,
      "loss": 1.66167107,
      "memory(GiB)": 97.17,
      "step": 11980,
      "train_speed(iter/s)": 1.641623
    },
    {
      "acc": 0.6401463,
      "epoch": 0.3040334855403349,
      "grad_norm": 6.9375,
      "learning_rate": 9.718153612843382e-06,
      "loss": 1.69964352,
      "memory(GiB)": 97.17,
      "step": 11985,
      "train_speed(iter/s)": 1.641699
    },
    {
      "acc": 0.64130259,
      "epoch": 0.3041603247082699,
      "grad_norm": 4.75,
      "learning_rate": 9.717806414551186e-06,
      "loss": 1.68308411,
      "memory(GiB)": 97.17,
      "step": 11990,
      "train_speed(iter/s)": 1.641768
    },
    {
      "acc": 0.66283197,
      "epoch": 0.30428716387620497,
      "grad_norm": 6.96875,
      "learning_rate": 9.717459008748798e-06,
      "loss": 1.62097893,
      "memory(GiB)": 97.17,
      "step": 11995,
      "train_speed(iter/s)": 1.641843
    },
    {
      "acc": 0.65009527,
      "epoch": 0.30441400304414,
      "grad_norm": 6.5,
      "learning_rate": 9.717111395451501e-06,
      "loss": 1.62187538,
      "memory(GiB)": 97.17,
      "step": 12000,
      "train_speed(iter/s)": 1.64192
    },
    {
      "epoch": 0.30441400304414,
      "eval_acc": 0.6403051432891239,
      "eval_loss": 1.6072824001312256,
      "eval_runtime": 58.6693,
      "eval_samples_per_second": 108.575,
      "eval_steps_per_second": 27.152,
      "step": 12000
    },
    {
      "acc": 0.63487883,
      "epoch": 0.3045408422120751,
      "grad_norm": 5.0625,
      "learning_rate": 9.716763574674586e-06,
      "loss": 1.70920391,
      "memory(GiB)": 97.17,
      "step": 12005,
      "train_speed(iter/s)": 1.627986
    },
    {
      "acc": 0.66409378,
      "epoch": 0.30466768138001016,
      "grad_norm": 5.15625,
      "learning_rate": 9.71641554643335e-06,
      "loss": 1.53675585,
      "memory(GiB)": 97.17,
      "step": 12010,
      "train_speed(iter/s)": 1.628052
    },
    {
      "acc": 0.65317812,
      "epoch": 0.3047945205479452,
      "grad_norm": 5.21875,
      "learning_rate": 9.7160673107431e-06,
      "loss": 1.59409924,
      "memory(GiB)": 97.17,
      "step": 12015,
      "train_speed(iter/s)": 1.628124
    },
    {
      "acc": 0.62879562,
      "epoch": 0.30492135971588025,
      "grad_norm": 6.53125,
      "learning_rate": 9.715718867619155e-06,
      "loss": 1.72604752,
      "memory(GiB)": 97.17,
      "step": 12020,
      "train_speed(iter/s)": 1.628197
    },
    {
      "acc": 0.6527308,
      "epoch": 0.30504819888381535,
      "grad_norm": 5.0,
      "learning_rate": 9.715370217076838e-06,
      "loss": 1.59955444,
      "memory(GiB)": 97.17,
      "step": 12025,
      "train_speed(iter/s)": 1.62827
    },
    {
      "acc": 0.64126039,
      "epoch": 0.3051750380517504,
      "grad_norm": 5.59375,
      "learning_rate": 9.715021359131489e-06,
      "loss": 1.66279678,
      "memory(GiB)": 97.17,
      "step": 12030,
      "train_speed(iter/s)": 1.628327
    },
    {
      "acc": 0.63692665,
      "epoch": 0.30530187721968544,
      "grad_norm": 8.4375,
      "learning_rate": 9.714672293798449e-06,
      "loss": 1.69898586,
      "memory(GiB)": 97.17,
      "step": 12035,
      "train_speed(iter/s)": 1.628403
    },
    {
      "acc": 0.639188,
      "epoch": 0.3054287163876205,
      "grad_norm": 5.125,
      "learning_rate": 9.71432302109307e-06,
      "loss": 1.68704529,
      "memory(GiB)": 97.17,
      "step": 12040,
      "train_speed(iter/s)": 1.62847
    },
    {
      "acc": 0.6301796,
      "epoch": 0.3055555555555556,
      "grad_norm": 6.21875,
      "learning_rate": 9.713973541030716e-06,
      "loss": 1.65257874,
      "memory(GiB)": 97.17,
      "step": 12045,
      "train_speed(iter/s)": 1.628544
    },
    {
      "acc": 0.65060339,
      "epoch": 0.3056823947234906,
      "grad_norm": 5.6875,
      "learning_rate": 9.713623853626763e-06,
      "loss": 1.5629858,
      "memory(GiB)": 97.17,
      "step": 12050,
      "train_speed(iter/s)": 1.62862
    },
    {
      "acc": 0.65652709,
      "epoch": 0.30580923389142567,
      "grad_norm": 5.1875,
      "learning_rate": 9.713273958896586e-06,
      "loss": 1.63734894,
      "memory(GiB)": 97.17,
      "step": 12055,
      "train_speed(iter/s)": 1.628696
    },
    {
      "acc": 0.6524848,
      "epoch": 0.3059360730593607,
      "grad_norm": 4.59375,
      "learning_rate": 9.712923856855578e-06,
      "loss": 1.58321218,
      "memory(GiB)": 97.17,
      "step": 12060,
      "train_speed(iter/s)": 1.628771
    },
    {
      "acc": 0.63657403,
      "epoch": 0.3060629122272958,
      "grad_norm": 6.34375,
      "learning_rate": 9.712573547519134e-06,
      "loss": 1.66819191,
      "memory(GiB)": 97.17,
      "step": 12065,
      "train_speed(iter/s)": 1.628845
    },
    {
      "acc": 0.63482022,
      "epoch": 0.30618975139523086,
      "grad_norm": 7.3125,
      "learning_rate": 9.712223030902668e-06,
      "loss": 1.72340755,
      "memory(GiB)": 97.17,
      "step": 12070,
      "train_speed(iter/s)": 1.628919
    },
    {
      "acc": 0.65714412,
      "epoch": 0.3063165905631659,
      "grad_norm": 6.40625,
      "learning_rate": 9.711872307021594e-06,
      "loss": 1.62071381,
      "memory(GiB)": 97.17,
      "step": 12075,
      "train_speed(iter/s)": 1.628993
    },
    {
      "acc": 0.65527635,
      "epoch": 0.30644342973110095,
      "grad_norm": 6.875,
      "learning_rate": 9.711521375891339e-06,
      "loss": 1.67614136,
      "memory(GiB)": 97.17,
      "step": 12080,
      "train_speed(iter/s)": 1.629065
    },
    {
      "acc": 0.64842501,
      "epoch": 0.30657026889903605,
      "grad_norm": 5.875,
      "learning_rate": 9.71117023752734e-06,
      "loss": 1.66528587,
      "memory(GiB)": 97.17,
      "step": 12085,
      "train_speed(iter/s)": 1.629137
    },
    {
      "acc": 0.6477582,
      "epoch": 0.3066971080669711,
      "grad_norm": 5.5,
      "learning_rate": 9.710818891945037e-06,
      "loss": 1.61027756,
      "memory(GiB)": 97.17,
      "step": 12090,
      "train_speed(iter/s)": 1.629206
    },
    {
      "acc": 0.65924463,
      "epoch": 0.30682394723490614,
      "grad_norm": 5.78125,
      "learning_rate": 9.71046733915989e-06,
      "loss": 1.58891096,
      "memory(GiB)": 97.17,
      "step": 12095,
      "train_speed(iter/s)": 1.62928
    },
    {
      "acc": 0.64791431,
      "epoch": 0.3069507864028412,
      "grad_norm": 4.84375,
      "learning_rate": 9.710115579187356e-06,
      "loss": 1.64345722,
      "memory(GiB)": 97.17,
      "step": 12100,
      "train_speed(iter/s)": 1.629357
    },
    {
      "acc": 0.63740606,
      "epoch": 0.3070776255707763,
      "grad_norm": 6.09375,
      "learning_rate": 9.709763612042911e-06,
      "loss": 1.66559505,
      "memory(GiB)": 97.17,
      "step": 12105,
      "train_speed(iter/s)": 1.629435
    },
    {
      "acc": 0.65210972,
      "epoch": 0.3072044647387113,
      "grad_norm": 5.4375,
      "learning_rate": 9.709411437742035e-06,
      "loss": 1.56379681,
      "memory(GiB)": 97.17,
      "step": 12110,
      "train_speed(iter/s)": 1.629503
    },
    {
      "acc": 0.65660381,
      "epoch": 0.30733130390664637,
      "grad_norm": 6.40625,
      "learning_rate": 9.709059056300218e-06,
      "loss": 1.59955273,
      "memory(GiB)": 97.17,
      "step": 12115,
      "train_speed(iter/s)": 1.629565
    },
    {
      "acc": 0.65138302,
      "epoch": 0.3074581430745814,
      "grad_norm": 5.46875,
      "learning_rate": 9.708706467732958e-06,
      "loss": 1.59153452,
      "memory(GiB)": 97.17,
      "step": 12120,
      "train_speed(iter/s)": 1.629637
    },
    {
      "acc": 0.65156617,
      "epoch": 0.3075849822425165,
      "grad_norm": 5.53125,
      "learning_rate": 9.708353672055766e-06,
      "loss": 1.67782021,
      "memory(GiB)": 97.17,
      "step": 12125,
      "train_speed(iter/s)": 1.629714
    },
    {
      "acc": 0.65236835,
      "epoch": 0.30771182141045156,
      "grad_norm": 5.9375,
      "learning_rate": 9.708000669284158e-06,
      "loss": 1.66671524,
      "memory(GiB)": 97.17,
      "step": 12130,
      "train_speed(iter/s)": 1.629787
    },
    {
      "acc": 0.652806,
      "epoch": 0.3078386605783866,
      "grad_norm": 5.96875,
      "learning_rate": 9.707647459433661e-06,
      "loss": 1.64346886,
      "memory(GiB)": 97.17,
      "step": 12135,
      "train_speed(iter/s)": 1.629866
    },
    {
      "acc": 0.64888916,
      "epoch": 0.30796549974632165,
      "grad_norm": 6.40625,
      "learning_rate": 9.707294042519808e-06,
      "loss": 1.6143465,
      "memory(GiB)": 97.17,
      "step": 12140,
      "train_speed(iter/s)": 1.629939
    },
    {
      "acc": 0.65195875,
      "epoch": 0.30809233891425675,
      "grad_norm": 5.5,
      "learning_rate": 9.70694041855815e-06,
      "loss": 1.71236172,
      "memory(GiB)": 97.17,
      "step": 12145,
      "train_speed(iter/s)": 1.630014
    },
    {
      "acc": 0.65103683,
      "epoch": 0.3082191780821918,
      "grad_norm": 4.6875,
      "learning_rate": 9.706586587564236e-06,
      "loss": 1.68275299,
      "memory(GiB)": 97.17,
      "step": 12150,
      "train_speed(iter/s)": 1.630087
    },
    {
      "acc": 0.64341426,
      "epoch": 0.30834601725012684,
      "grad_norm": 6.28125,
      "learning_rate": 9.70623254955363e-06,
      "loss": 1.63932571,
      "memory(GiB)": 97.17,
      "step": 12155,
      "train_speed(iter/s)": 1.630159
    },
    {
      "acc": 0.66488438,
      "epoch": 0.3084728564180619,
      "grad_norm": 7.78125,
      "learning_rate": 9.705878304541905e-06,
      "loss": 1.55229368,
      "memory(GiB)": 97.17,
      "step": 12160,
      "train_speed(iter/s)": 1.630233
    },
    {
      "acc": 0.64901409,
      "epoch": 0.308599695585997,
      "grad_norm": 5.46875,
      "learning_rate": 9.705523852544643e-06,
      "loss": 1.61777954,
      "memory(GiB)": 97.17,
      "step": 12165,
      "train_speed(iter/s)": 1.630305
    },
    {
      "acc": 0.64030151,
      "epoch": 0.308726534753932,
      "grad_norm": 6.625,
      "learning_rate": 9.705169193577434e-06,
      "loss": 1.70685806,
      "memory(GiB)": 97.17,
      "step": 12170,
      "train_speed(iter/s)": 1.630374
    },
    {
      "acc": 0.65600266,
      "epoch": 0.30885337392186707,
      "grad_norm": 7.3125,
      "learning_rate": 9.704814327655874e-06,
      "loss": 1.63442345,
      "memory(GiB)": 97.17,
      "step": 12175,
      "train_speed(iter/s)": 1.630447
    },
    {
      "acc": 0.66261511,
      "epoch": 0.3089802130898021,
      "grad_norm": 5.5625,
      "learning_rate": 9.704459254795575e-06,
      "loss": 1.61930466,
      "memory(GiB)": 97.17,
      "step": 12180,
      "train_speed(iter/s)": 1.630516
    },
    {
      "acc": 0.62956772,
      "epoch": 0.3091070522577372,
      "grad_norm": 5.09375,
      "learning_rate": 9.704103975012155e-06,
      "loss": 1.68084412,
      "memory(GiB)": 97.17,
      "step": 12185,
      "train_speed(iter/s)": 1.630589
    },
    {
      "acc": 0.65555058,
      "epoch": 0.30923389142567226,
      "grad_norm": 5.5,
      "learning_rate": 9.703748488321238e-06,
      "loss": 1.58282928,
      "memory(GiB)": 97.17,
      "step": 12190,
      "train_speed(iter/s)": 1.630663
    },
    {
      "acc": 0.67006302,
      "epoch": 0.3093607305936073,
      "grad_norm": 5.96875,
      "learning_rate": 9.703392794738464e-06,
      "loss": 1.57716703,
      "memory(GiB)": 97.17,
      "step": 12195,
      "train_speed(iter/s)": 1.630732
    },
    {
      "acc": 0.64453869,
      "epoch": 0.30948756976154235,
      "grad_norm": 6.0625,
      "learning_rate": 9.703036894279476e-06,
      "loss": 1.6787096,
      "memory(GiB)": 97.17,
      "step": 12200,
      "train_speed(iter/s)": 1.630804
    },
    {
      "acc": 0.66845036,
      "epoch": 0.30961440892947745,
      "grad_norm": 6.375,
      "learning_rate": 9.702680786959925e-06,
      "loss": 1.54558868,
      "memory(GiB)": 97.17,
      "step": 12205,
      "train_speed(iter/s)": 1.630872
    },
    {
      "acc": 0.65537148,
      "epoch": 0.3097412480974125,
      "grad_norm": 5.21875,
      "learning_rate": 9.702324472795481e-06,
      "loss": 1.62108536,
      "memory(GiB)": 97.17,
      "step": 12210,
      "train_speed(iter/s)": 1.630943
    },
    {
      "acc": 0.64351082,
      "epoch": 0.30986808726534754,
      "grad_norm": 5.5625,
      "learning_rate": 9.70196795180181e-06,
      "loss": 1.64338341,
      "memory(GiB)": 97.17,
      "step": 12215,
      "train_speed(iter/s)": 1.631015
    },
    {
      "acc": 0.65183668,
      "epoch": 0.3099949264332826,
      "grad_norm": 6.65625,
      "learning_rate": 9.701611223994596e-06,
      "loss": 1.63149185,
      "memory(GiB)": 97.17,
      "step": 12220,
      "train_speed(iter/s)": 1.631088
    },
    {
      "acc": 0.6511651,
      "epoch": 0.3101217656012177,
      "grad_norm": 4.78125,
      "learning_rate": 9.701254289389529e-06,
      "loss": 1.60745201,
      "memory(GiB)": 97.17,
      "step": 12225,
      "train_speed(iter/s)": 1.631161
    },
    {
      "acc": 0.65827971,
      "epoch": 0.3102486047691527,
      "grad_norm": 6.6875,
      "learning_rate": 9.700897148002308e-06,
      "loss": 1.65823116,
      "memory(GiB)": 97.17,
      "step": 12230,
      "train_speed(iter/s)": 1.631231
    },
    {
      "acc": 0.65242763,
      "epoch": 0.31037544393708777,
      "grad_norm": 5.46875,
      "learning_rate": 9.700539799848645e-06,
      "loss": 1.66842232,
      "memory(GiB)": 97.17,
      "step": 12235,
      "train_speed(iter/s)": 1.631302
    },
    {
      "acc": 0.65059233,
      "epoch": 0.3105022831050228,
      "grad_norm": 4.96875,
      "learning_rate": 9.700182244944252e-06,
      "loss": 1.62010841,
      "memory(GiB)": 97.17,
      "step": 12240,
      "train_speed(iter/s)": 1.631377
    },
    {
      "acc": 0.6480607,
      "epoch": 0.3106291222729579,
      "grad_norm": 5.375,
      "learning_rate": 9.69982448330486e-06,
      "loss": 1.66656265,
      "memory(GiB)": 97.17,
      "step": 12245,
      "train_speed(iter/s)": 1.631447
    },
    {
      "acc": 0.64326897,
      "epoch": 0.31075596144089296,
      "grad_norm": 5.8125,
      "learning_rate": 9.699466514946206e-06,
      "loss": 1.75429573,
      "memory(GiB)": 97.17,
      "step": 12250,
      "train_speed(iter/s)": 1.631516
    },
    {
      "acc": 0.66316671,
      "epoch": 0.310882800608828,
      "grad_norm": 6.75,
      "learning_rate": 9.699108339884032e-06,
      "loss": 1.57260742,
      "memory(GiB)": 97.17,
      "step": 12255,
      "train_speed(iter/s)": 1.631586
    },
    {
      "acc": 0.63127174,
      "epoch": 0.31100963977676305,
      "grad_norm": 5.46875,
      "learning_rate": 9.698749958134093e-06,
      "loss": 1.7066433,
      "memory(GiB)": 97.17,
      "step": 12260,
      "train_speed(iter/s)": 1.631652
    },
    {
      "acc": 0.64093122,
      "epoch": 0.31113647894469815,
      "grad_norm": 6.0,
      "learning_rate": 9.698391369712152e-06,
      "loss": 1.72350273,
      "memory(GiB)": 97.17,
      "step": 12265,
      "train_speed(iter/s)": 1.631722
    },
    {
      "acc": 0.65811157,
      "epoch": 0.3112633181126332,
      "grad_norm": 5.28125,
      "learning_rate": 9.698032574633982e-06,
      "loss": 1.62617531,
      "memory(GiB)": 97.17,
      "step": 12270,
      "train_speed(iter/s)": 1.631793
    },
    {
      "acc": 0.65103946,
      "epoch": 0.31139015728056824,
      "grad_norm": 5.25,
      "learning_rate": 9.697673572915364e-06,
      "loss": 1.62757626,
      "memory(GiB)": 97.17,
      "step": 12275,
      "train_speed(iter/s)": 1.631868
    },
    {
      "acc": 0.6472517,
      "epoch": 0.3115169964485033,
      "grad_norm": 5.4375,
      "learning_rate": 9.697314364572087e-06,
      "loss": 1.64757442,
      "memory(GiB)": 97.17,
      "step": 12280,
      "train_speed(iter/s)": 1.631934
    },
    {
      "acc": 0.65679131,
      "epoch": 0.3116438356164384,
      "grad_norm": 5.625,
      "learning_rate": 9.696954949619955e-06,
      "loss": 1.67645168,
      "memory(GiB)": 97.17,
      "step": 12285,
      "train_speed(iter/s)": 1.632006
    },
    {
      "acc": 0.65980196,
      "epoch": 0.3117706747843734,
      "grad_norm": 5.09375,
      "learning_rate": 9.696595328074774e-06,
      "loss": 1.60186901,
      "memory(GiB)": 97.17,
      "step": 12290,
      "train_speed(iter/s)": 1.632071
    },
    {
      "acc": 0.6604579,
      "epoch": 0.31189751395230847,
      "grad_norm": 5.8125,
      "learning_rate": 9.69623549995236e-06,
      "loss": 1.63972206,
      "memory(GiB)": 97.17,
      "step": 12295,
      "train_speed(iter/s)": 1.632142
    },
    {
      "acc": 0.64192972,
      "epoch": 0.3120243531202435,
      "grad_norm": 5.96875,
      "learning_rate": 9.695875465268543e-06,
      "loss": 1.68872604,
      "memory(GiB)": 97.17,
      "step": 12300,
      "train_speed(iter/s)": 1.632215
    },
    {
      "acc": 0.65491533,
      "epoch": 0.3121511922881786,
      "grad_norm": 7.25,
      "learning_rate": 9.695515224039156e-06,
      "loss": 1.66243534,
      "memory(GiB)": 97.17,
      "step": 12305,
      "train_speed(iter/s)": 1.632283
    },
    {
      "acc": 0.6556776,
      "epoch": 0.31227803145611366,
      "grad_norm": 5.875,
      "learning_rate": 9.695154776280047e-06,
      "loss": 1.60202942,
      "memory(GiB)": 97.17,
      "step": 12310,
      "train_speed(iter/s)": 1.63235
    },
    {
      "acc": 0.64802794,
      "epoch": 0.3124048706240487,
      "grad_norm": 5.625,
      "learning_rate": 9.694794122007067e-06,
      "loss": 1.68175659,
      "memory(GiB)": 97.17,
      "step": 12315,
      "train_speed(iter/s)": 1.632423
    },
    {
      "acc": 0.64842658,
      "epoch": 0.31253170979198375,
      "grad_norm": 5.40625,
      "learning_rate": 9.694433261236083e-06,
      "loss": 1.6339119,
      "memory(GiB)": 97.17,
      "step": 12320,
      "train_speed(iter/s)": 1.632493
    },
    {
      "acc": 0.64380045,
      "epoch": 0.31265854895991885,
      "grad_norm": 5.5,
      "learning_rate": 9.694072193982962e-06,
      "loss": 1.7081356,
      "memory(GiB)": 97.17,
      "step": 12325,
      "train_speed(iter/s)": 1.632561
    },
    {
      "acc": 0.65466681,
      "epoch": 0.3127853881278539,
      "grad_norm": 5.5625,
      "learning_rate": 9.69371092026359e-06,
      "loss": 1.64989376,
      "memory(GiB)": 97.17,
      "step": 12330,
      "train_speed(iter/s)": 1.632635
    },
    {
      "acc": 0.67976747,
      "epoch": 0.31291222729578894,
      "grad_norm": 5.3125,
      "learning_rate": 9.693349440093855e-06,
      "loss": 1.58285637,
      "memory(GiB)": 97.17,
      "step": 12335,
      "train_speed(iter/s)": 1.632708
    },
    {
      "acc": 0.66369729,
      "epoch": 0.313039066463724,
      "grad_norm": 4.6875,
      "learning_rate": 9.69298775348966e-06,
      "loss": 1.52019424,
      "memory(GiB)": 97.17,
      "step": 12340,
      "train_speed(iter/s)": 1.632773
    },
    {
      "acc": 0.65198488,
      "epoch": 0.3131659056316591,
      "grad_norm": 6.0625,
      "learning_rate": 9.69262586046691e-06,
      "loss": 1.63689671,
      "memory(GiB)": 97.17,
      "step": 12345,
      "train_speed(iter/s)": 1.632844
    },
    {
      "acc": 0.65919633,
      "epoch": 0.3132927447995941,
      "grad_norm": 5.84375,
      "learning_rate": 9.692263761041521e-06,
      "loss": 1.65497913,
      "memory(GiB)": 97.17,
      "step": 12350,
      "train_speed(iter/s)": 1.632912
    },
    {
      "acc": 0.64915676,
      "epoch": 0.31341958396752917,
      "grad_norm": 4.59375,
      "learning_rate": 9.691901455229425e-06,
      "loss": 1.62881203,
      "memory(GiB)": 97.17,
      "step": 12355,
      "train_speed(iter/s)": 1.632984
    },
    {
      "acc": 0.6485455,
      "epoch": 0.3135464231354642,
      "grad_norm": 5.40625,
      "learning_rate": 9.691538943046552e-06,
      "loss": 1.64240589,
      "memory(GiB)": 97.17,
      "step": 12360,
      "train_speed(iter/s)": 1.633054
    },
    {
      "acc": 0.6553206,
      "epoch": 0.3136732623033993,
      "grad_norm": 4.75,
      "learning_rate": 9.691176224508853e-06,
      "loss": 1.66156693,
      "memory(GiB)": 97.17,
      "step": 12365,
      "train_speed(iter/s)": 1.633121
    },
    {
      "acc": 0.64690952,
      "epoch": 0.31380010147133436,
      "grad_norm": 5.78125,
      "learning_rate": 9.690813299632278e-06,
      "loss": 1.60425549,
      "memory(GiB)": 97.17,
      "step": 12370,
      "train_speed(iter/s)": 1.633194
    },
    {
      "acc": 0.65716491,
      "epoch": 0.3139269406392694,
      "grad_norm": 5.1875,
      "learning_rate": 9.690450168432793e-06,
      "loss": 1.66567345,
      "memory(GiB)": 97.17,
      "step": 12375,
      "train_speed(iter/s)": 1.63326
    },
    {
      "acc": 0.63407702,
      "epoch": 0.31405377980720445,
      "grad_norm": 5.0625,
      "learning_rate": 9.690086830926366e-06,
      "loss": 1.64914207,
      "memory(GiB)": 97.17,
      "step": 12380,
      "train_speed(iter/s)": 1.633327
    },
    {
      "acc": 0.63304691,
      "epoch": 0.31418061897513955,
      "grad_norm": 5.125,
      "learning_rate": 9.689723287128981e-06,
      "loss": 1.68027935,
      "memory(GiB)": 97.17,
      "step": 12385,
      "train_speed(iter/s)": 1.633393
    },
    {
      "acc": 0.65252461,
      "epoch": 0.3143074581430746,
      "grad_norm": 5.9375,
      "learning_rate": 9.689359537056628e-06,
      "loss": 1.63740005,
      "memory(GiB)": 97.17,
      "step": 12390,
      "train_speed(iter/s)": 1.633464
    },
    {
      "acc": 0.64774771,
      "epoch": 0.31443429731100964,
      "grad_norm": 5.71875,
      "learning_rate": 9.688995580725304e-06,
      "loss": 1.59514341,
      "memory(GiB)": 97.17,
      "step": 12395,
      "train_speed(iter/s)": 1.633533
    },
    {
      "acc": 0.65391817,
      "epoch": 0.3145611364789447,
      "grad_norm": 5.84375,
      "learning_rate": 9.688631418151022e-06,
      "loss": 1.58650837,
      "memory(GiB)": 97.17,
      "step": 12400,
      "train_speed(iter/s)": 1.633604
    },
    {
      "acc": 0.66186643,
      "epoch": 0.3146879756468798,
      "grad_norm": 5.90625,
      "learning_rate": 9.688267049349796e-06,
      "loss": 1.59120388,
      "memory(GiB)": 97.17,
      "step": 12405,
      "train_speed(iter/s)": 1.633675
    },
    {
      "acc": 0.64244699,
      "epoch": 0.3148148148148148,
      "grad_norm": 5.3125,
      "learning_rate": 9.687902474337654e-06,
      "loss": 1.72289429,
      "memory(GiB)": 97.17,
      "step": 12410,
      "train_speed(iter/s)": 1.63374
    },
    {
      "acc": 0.64908428,
      "epoch": 0.31494165398274987,
      "grad_norm": 5.0625,
      "learning_rate": 9.687537693130631e-06,
      "loss": 1.6042099,
      "memory(GiB)": 97.17,
      "step": 12415,
      "train_speed(iter/s)": 1.63381
    },
    {
      "acc": 0.64052429,
      "epoch": 0.3150684931506849,
      "grad_norm": 5.4375,
      "learning_rate": 9.687172705744773e-06,
      "loss": 1.68112698,
      "memory(GiB)": 97.17,
      "step": 12420,
      "train_speed(iter/s)": 1.633878
    },
    {
      "acc": 0.63257399,
      "epoch": 0.31519533231862,
      "grad_norm": 5.875,
      "learning_rate": 9.686807512196132e-06,
      "loss": 1.7116066,
      "memory(GiB)": 97.17,
      "step": 12425,
      "train_speed(iter/s)": 1.633949
    },
    {
      "acc": 0.65500231,
      "epoch": 0.31532217148655506,
      "grad_norm": 5.875,
      "learning_rate": 9.68644211250077e-06,
      "loss": 1.61961002,
      "memory(GiB)": 97.17,
      "step": 12430,
      "train_speed(iter/s)": 1.634019
    },
    {
      "acc": 0.62910328,
      "epoch": 0.3154490106544901,
      "grad_norm": 6.15625,
      "learning_rate": 9.686076506674761e-06,
      "loss": 1.68652954,
      "memory(GiB)": 97.17,
      "step": 12435,
      "train_speed(iter/s)": 1.634088
    },
    {
      "acc": 0.64076643,
      "epoch": 0.31557584982242515,
      "grad_norm": 5.09375,
      "learning_rate": 9.685710694734187e-06,
      "loss": 1.68172646,
      "memory(GiB)": 97.17,
      "step": 12440,
      "train_speed(iter/s)": 1.634161
    },
    {
      "acc": 0.64843788,
      "epoch": 0.31570268899036025,
      "grad_norm": 5.78125,
      "learning_rate": 9.685344676695135e-06,
      "loss": 1.63500252,
      "memory(GiB)": 97.17,
      "step": 12445,
      "train_speed(iter/s)": 1.63423
    },
    {
      "acc": 0.6416647,
      "epoch": 0.3158295281582953,
      "grad_norm": 7.28125,
      "learning_rate": 9.684978452573706e-06,
      "loss": 1.67504749,
      "memory(GiB)": 97.17,
      "step": 12450,
      "train_speed(iter/s)": 1.634295
    },
    {
      "acc": 0.6630147,
      "epoch": 0.31595636732623034,
      "grad_norm": 4.78125,
      "learning_rate": 9.684612022386008e-06,
      "loss": 1.5468152,
      "memory(GiB)": 97.17,
      "step": 12455,
      "train_speed(iter/s)": 1.634358
    },
    {
      "acc": 0.63974771,
      "epoch": 0.3160832064941654,
      "grad_norm": 4.6875,
      "learning_rate": 9.68424538614816e-06,
      "loss": 1.67091637,
      "memory(GiB)": 97.17,
      "step": 12460,
      "train_speed(iter/s)": 1.634426
    },
    {
      "acc": 0.64996028,
      "epoch": 0.3162100456621005,
      "grad_norm": 5.53125,
      "learning_rate": 9.683878543876281e-06,
      "loss": 1.66914978,
      "memory(GiB)": 97.17,
      "step": 12465,
      "train_speed(iter/s)": 1.634497
    },
    {
      "acc": 0.65390582,
      "epoch": 0.3163368848300355,
      "grad_norm": 5.90625,
      "learning_rate": 9.683511495586516e-06,
      "loss": 1.60467873,
      "memory(GiB)": 97.17,
      "step": 12470,
      "train_speed(iter/s)": 1.63457
    },
    {
      "acc": 0.65536156,
      "epoch": 0.31646372399797057,
      "grad_norm": 5.46875,
      "learning_rate": 9.683144241295003e-06,
      "loss": 1.57642612,
      "memory(GiB)": 97.17,
      "step": 12475,
      "train_speed(iter/s)": 1.634627
    },
    {
      "acc": 0.65573554,
      "epoch": 0.3165905631659056,
      "grad_norm": 7.40625,
      "learning_rate": 9.682776781017899e-06,
      "loss": 1.61298599,
      "memory(GiB)": 97.17,
      "step": 12480,
      "train_speed(iter/s)": 1.634694
    },
    {
      "acc": 0.63949137,
      "epoch": 0.3167174023338407,
      "grad_norm": 6.125,
      "learning_rate": 9.682409114771364e-06,
      "loss": 1.67794991,
      "memory(GiB)": 97.17,
      "step": 12485,
      "train_speed(iter/s)": 1.634759
    },
    {
      "acc": 0.63329649,
      "epoch": 0.31684424150177576,
      "grad_norm": 5.5625,
      "learning_rate": 9.682041242571571e-06,
      "loss": 1.64192829,
      "memory(GiB)": 97.17,
      "step": 12490,
      "train_speed(iter/s)": 1.634828
    },
    {
      "acc": 0.65123,
      "epoch": 0.3169710806697108,
      "grad_norm": 6.125,
      "learning_rate": 9.681673164434701e-06,
      "loss": 1.65969276,
      "memory(GiB)": 97.17,
      "step": 12495,
      "train_speed(iter/s)": 1.6349
    },
    {
      "acc": 0.660251,
      "epoch": 0.31709791983764585,
      "grad_norm": 4.53125,
      "learning_rate": 9.681304880376942e-06,
      "loss": 1.60781631,
      "memory(GiB)": 97.17,
      "step": 12500,
      "train_speed(iter/s)": 1.634967
    },
    {
      "acc": 0.65085888,
      "epoch": 0.31722475900558095,
      "grad_norm": 5.5,
      "learning_rate": 9.680936390414495e-06,
      "loss": 1.73761711,
      "memory(GiB)": 97.17,
      "step": 12505,
      "train_speed(iter/s)": 1.635035
    },
    {
      "acc": 0.64507508,
      "epoch": 0.317351598173516,
      "grad_norm": 6.125,
      "learning_rate": 9.680567694563566e-06,
      "loss": 1.64682598,
      "memory(GiB)": 97.17,
      "step": 12510,
      "train_speed(iter/s)": 1.635099
    },
    {
      "acc": 0.65353298,
      "epoch": 0.31747843734145104,
      "grad_norm": 5.625,
      "learning_rate": 9.680198792840371e-06,
      "loss": 1.64877968,
      "memory(GiB)": 97.17,
      "step": 12515,
      "train_speed(iter/s)": 1.635166
    },
    {
      "acc": 0.6431283,
      "epoch": 0.3176052765093861,
      "grad_norm": 5.4375,
      "learning_rate": 9.67982968526114e-06,
      "loss": 1.63001709,
      "memory(GiB)": 97.17,
      "step": 12520,
      "train_speed(iter/s)": 1.635227
    },
    {
      "acc": 0.66182041,
      "epoch": 0.3177321156773212,
      "grad_norm": 5.53125,
      "learning_rate": 9.679460371842104e-06,
      "loss": 1.57468204,
      "memory(GiB)": 97.17,
      "step": 12525,
      "train_speed(iter/s)": 1.635293
    },
    {
      "acc": 0.64837599,
      "epoch": 0.3178589548452562,
      "grad_norm": 5.6875,
      "learning_rate": 9.679090852599508e-06,
      "loss": 1.65184116,
      "memory(GiB)": 97.17,
      "step": 12530,
      "train_speed(iter/s)": 1.635362
    },
    {
      "acc": 0.64117403,
      "epoch": 0.31798579401319127,
      "grad_norm": 5.53125,
      "learning_rate": 9.678721127549608e-06,
      "loss": 1.67416115,
      "memory(GiB)": 97.17,
      "step": 12535,
      "train_speed(iter/s)": 1.635433
    },
    {
      "acc": 0.63163776,
      "epoch": 0.3181126331811263,
      "grad_norm": 6.84375,
      "learning_rate": 9.678351196708662e-06,
      "loss": 1.76682434,
      "memory(GiB)": 97.17,
      "step": 12540,
      "train_speed(iter/s)": 1.6355
    },
    {
      "acc": 0.64566784,
      "epoch": 0.3182394723490614,
      "grad_norm": 6.25,
      "learning_rate": 9.677981060092943e-06,
      "loss": 1.69713898,
      "memory(GiB)": 97.17,
      "step": 12545,
      "train_speed(iter/s)": 1.635566
    },
    {
      "acc": 0.63155918,
      "epoch": 0.31836631151699646,
      "grad_norm": 4.53125,
      "learning_rate": 9.677610717718732e-06,
      "loss": 1.71420555,
      "memory(GiB)": 97.17,
      "step": 12550,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.62976179,
      "epoch": 0.3184931506849315,
      "grad_norm": 6.6875,
      "learning_rate": 9.677240169602317e-06,
      "loss": 1.70730667,
      "memory(GiB)": 97.17,
      "step": 12555,
      "train_speed(iter/s)": 1.635697
    },
    {
      "acc": 0.65112371,
      "epoch": 0.31861998985286655,
      "grad_norm": 6.625,
      "learning_rate": 9.676869415759999e-06,
      "loss": 1.67480469,
      "memory(GiB)": 97.17,
      "step": 12560,
      "train_speed(iter/s)": 1.63576
    },
    {
      "acc": 0.63670521,
      "epoch": 0.31874682902080165,
      "grad_norm": 5.4375,
      "learning_rate": 9.67649845620808e-06,
      "loss": 1.7223175,
      "memory(GiB)": 97.17,
      "step": 12565,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.65497069,
      "epoch": 0.3188736681887367,
      "grad_norm": 5.21875,
      "learning_rate": 9.676127290962883e-06,
      "loss": 1.67975903,
      "memory(GiB)": 97.17,
      "step": 12570,
      "train_speed(iter/s)": 1.635893
    },
    {
      "acc": 0.65211391,
      "epoch": 0.31900050735667174,
      "grad_norm": 5.8125,
      "learning_rate": 9.675755920040728e-06,
      "loss": 1.62615433,
      "memory(GiB)": 97.17,
      "step": 12575,
      "train_speed(iter/s)": 1.635957
    },
    {
      "acc": 0.66108007,
      "epoch": 0.3191273465246068,
      "grad_norm": 5.09375,
      "learning_rate": 9.675384343457954e-06,
      "loss": 1.63112106,
      "memory(GiB)": 97.17,
      "step": 12580,
      "train_speed(iter/s)": 1.636016
    },
    {
      "acc": 0.64188585,
      "epoch": 0.3192541856925419,
      "grad_norm": 5.125,
      "learning_rate": 9.675012561230901e-06,
      "loss": 1.69303932,
      "memory(GiB)": 97.17,
      "step": 12585,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.65562472,
      "epoch": 0.3193810248604769,
      "grad_norm": 5.90625,
      "learning_rate": 9.674640573375924e-06,
      "loss": 1.65416985,
      "memory(GiB)": 97.17,
      "step": 12590,
      "train_speed(iter/s)": 1.636152
    },
    {
      "acc": 0.64916744,
      "epoch": 0.31950786402841197,
      "grad_norm": 6.6875,
      "learning_rate": 9.674268379909383e-06,
      "loss": 1.68290672,
      "memory(GiB)": 97.17,
      "step": 12595,
      "train_speed(iter/s)": 1.636221
    },
    {
      "acc": 0.64814253,
      "epoch": 0.319634703196347,
      "grad_norm": 7.15625,
      "learning_rate": 9.67389598084765e-06,
      "loss": 1.63947773,
      "memory(GiB)": 97.17,
      "step": 12600,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.64805546,
      "epoch": 0.3197615423642821,
      "grad_norm": 7.46875,
      "learning_rate": 9.673523376207103e-06,
      "loss": 1.60097141,
      "memory(GiB)": 97.17,
      "step": 12605,
      "train_speed(iter/s)": 1.636353
    },
    {
      "acc": 0.64015789,
      "epoch": 0.31988838153221716,
      "grad_norm": 7.0,
      "learning_rate": 9.673150566004135e-06,
      "loss": 1.66025257,
      "memory(GiB)": 97.17,
      "step": 12610,
      "train_speed(iter/s)": 1.636425
    },
    {
      "acc": 0.64231443,
      "epoch": 0.3200152207001522,
      "grad_norm": 5.34375,
      "learning_rate": 9.672777550255137e-06,
      "loss": 1.72642555,
      "memory(GiB)": 97.17,
      "step": 12615,
      "train_speed(iter/s)": 1.636493
    },
    {
      "acc": 0.65578756,
      "epoch": 0.32014205986808725,
      "grad_norm": 5.625,
      "learning_rate": 9.672404328976523e-06,
      "loss": 1.66687927,
      "memory(GiB)": 97.17,
      "step": 12620,
      "train_speed(iter/s)": 1.636556
    },
    {
      "acc": 0.65339026,
      "epoch": 0.32026889903602235,
      "grad_norm": 5.71875,
      "learning_rate": 9.672030902184706e-06,
      "loss": 1.64034901,
      "memory(GiB)": 97.17,
      "step": 12625,
      "train_speed(iter/s)": 1.636618
    },
    {
      "acc": 0.64897451,
      "epoch": 0.3203957382039574,
      "grad_norm": 5.0,
      "learning_rate": 9.671657269896108e-06,
      "loss": 1.72187843,
      "memory(GiB)": 97.17,
      "step": 12630,
      "train_speed(iter/s)": 1.636681
    },
    {
      "acc": 0.65473375,
      "epoch": 0.32052257737189244,
      "grad_norm": 4.84375,
      "learning_rate": 9.671283432127169e-06,
      "loss": 1.63135338,
      "memory(GiB)": 97.17,
      "step": 12635,
      "train_speed(iter/s)": 1.636748
    },
    {
      "acc": 0.65806017,
      "epoch": 0.3206494165398275,
      "grad_norm": 5.5625,
      "learning_rate": 9.670909388894328e-06,
      "loss": 1.58408194,
      "memory(GiB)": 97.17,
      "step": 12640,
      "train_speed(iter/s)": 1.636818
    },
    {
      "acc": 0.64611273,
      "epoch": 0.3207762557077626,
      "grad_norm": 4.6875,
      "learning_rate": 9.670535140214037e-06,
      "loss": 1.66608925,
      "memory(GiB)": 97.17,
      "step": 12645,
      "train_speed(iter/s)": 1.636885
    },
    {
      "acc": 0.65427494,
      "epoch": 0.3209030948756976,
      "grad_norm": 5.65625,
      "learning_rate": 9.670160686102759e-06,
      "loss": 1.65067062,
      "memory(GiB)": 97.17,
      "step": 12650,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.64160805,
      "epoch": 0.32102993404363267,
      "grad_norm": 5.40625,
      "learning_rate": 9.669786026576962e-06,
      "loss": 1.68313408,
      "memory(GiB)": 97.17,
      "step": 12655,
      "train_speed(iter/s)": 1.637009
    },
    {
      "acc": 0.6566082,
      "epoch": 0.3211567732115677,
      "grad_norm": 5.46875,
      "learning_rate": 9.669411161653127e-06,
      "loss": 1.63150597,
      "memory(GiB)": 97.17,
      "step": 12660,
      "train_speed(iter/s)": 1.637074
    },
    {
      "acc": 0.64243369,
      "epoch": 0.3212836123795028,
      "grad_norm": 5.3125,
      "learning_rate": 9.669036091347742e-06,
      "loss": 1.61782188,
      "memory(GiB)": 97.17,
      "step": 12665,
      "train_speed(iter/s)": 1.637142
    },
    {
      "acc": 0.65017185,
      "epoch": 0.32141045154743786,
      "grad_norm": 6.4375,
      "learning_rate": 9.668660815677304e-06,
      "loss": 1.59754953,
      "memory(GiB)": 97.17,
      "step": 12670,
      "train_speed(iter/s)": 1.637204
    },
    {
      "acc": 0.67116165,
      "epoch": 0.3215372907153729,
      "grad_norm": 6.53125,
      "learning_rate": 9.668285334658319e-06,
      "loss": 1.57749071,
      "memory(GiB)": 97.17,
      "step": 12675,
      "train_speed(iter/s)": 1.637272
    },
    {
      "acc": 0.6393919,
      "epoch": 0.32166412988330795,
      "grad_norm": 5.1875,
      "learning_rate": 9.667909648307302e-06,
      "loss": 1.68731289,
      "memory(GiB)": 97.17,
      "step": 12680,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.66831617,
      "epoch": 0.32179096905124305,
      "grad_norm": 7.0625,
      "learning_rate": 9.66753375664078e-06,
      "loss": 1.59963207,
      "memory(GiB)": 97.17,
      "step": 12685,
      "train_speed(iter/s)": 1.637401
    },
    {
      "acc": 0.6519949,
      "epoch": 0.3219178082191781,
      "grad_norm": 6.0625,
      "learning_rate": 9.667157659675284e-06,
      "loss": 1.63812599,
      "memory(GiB)": 97.17,
      "step": 12690,
      "train_speed(iter/s)": 1.637464
    },
    {
      "acc": 0.65419474,
      "epoch": 0.32204464738711314,
      "grad_norm": 5.46875,
      "learning_rate": 9.666781357427355e-06,
      "loss": 1.65822525,
      "memory(GiB)": 97.17,
      "step": 12695,
      "train_speed(iter/s)": 1.637528
    },
    {
      "acc": 0.64320679,
      "epoch": 0.3221714865550482,
      "grad_norm": 5.5,
      "learning_rate": 9.666404849913546e-06,
      "loss": 1.65864391,
      "memory(GiB)": 97.17,
      "step": 12700,
      "train_speed(iter/s)": 1.637593
    },
    {
      "acc": 0.65752296,
      "epoch": 0.3222983257229833,
      "grad_norm": 6.6875,
      "learning_rate": 9.66602813715042e-06,
      "loss": 1.60788002,
      "memory(GiB)": 97.17,
      "step": 12705,
      "train_speed(iter/s)": 1.637662
    },
    {
      "acc": 0.65540419,
      "epoch": 0.3224251648909183,
      "grad_norm": 5.65625,
      "learning_rate": 9.665651219154543e-06,
      "loss": 1.6908247,
      "memory(GiB)": 97.17,
      "step": 12710,
      "train_speed(iter/s)": 1.637727
    },
    {
      "acc": 0.6586338,
      "epoch": 0.32255200405885337,
      "grad_norm": 4.875,
      "learning_rate": 9.665274095942495e-06,
      "loss": 1.6695631,
      "memory(GiB)": 97.17,
      "step": 12715,
      "train_speed(iter/s)": 1.637792
    },
    {
      "acc": 0.65187531,
      "epoch": 0.3226788432267884,
      "grad_norm": 7.0,
      "learning_rate": 9.664896767530862e-06,
      "loss": 1.68884411,
      "memory(GiB)": 97.17,
      "step": 12720,
      "train_speed(iter/s)": 1.637858
    },
    {
      "acc": 0.65139108,
      "epoch": 0.3228056823947235,
      "grad_norm": 5.125,
      "learning_rate": 9.664519233936242e-06,
      "loss": 1.67368927,
      "memory(GiB)": 97.17,
      "step": 12725,
      "train_speed(iter/s)": 1.637926
    },
    {
      "acc": 0.64907026,
      "epoch": 0.32293252156265856,
      "grad_norm": 4.90625,
      "learning_rate": 9.664141495175242e-06,
      "loss": 1.57588377,
      "memory(GiB)": 97.17,
      "step": 12730,
      "train_speed(iter/s)": 1.637988
    },
    {
      "acc": 0.65204868,
      "epoch": 0.3230593607305936,
      "grad_norm": 5.09375,
      "learning_rate": 9.663763551264476e-06,
      "loss": 1.66170807,
      "memory(GiB)": 97.17,
      "step": 12735,
      "train_speed(iter/s)": 1.638046
    },
    {
      "acc": 0.65029244,
      "epoch": 0.32318619989852865,
      "grad_norm": 6.1875,
      "learning_rate": 9.663385402220565e-06,
      "loss": 1.60605583,
      "memory(GiB)": 97.17,
      "step": 12740,
      "train_speed(iter/s)": 1.638107
    },
    {
      "acc": 0.65202026,
      "epoch": 0.32331303906646375,
      "grad_norm": 6.3125,
      "learning_rate": 9.663007048060144e-06,
      "loss": 1.70589123,
      "memory(GiB)": 97.17,
      "step": 12745,
      "train_speed(iter/s)": 1.638169
    },
    {
      "acc": 0.66016808,
      "epoch": 0.3234398782343988,
      "grad_norm": 5.34375,
      "learning_rate": 9.662628488799854e-06,
      "loss": 1.5867878,
      "memory(GiB)": 97.17,
      "step": 12750,
      "train_speed(iter/s)": 1.638229
    },
    {
      "acc": 0.64225903,
      "epoch": 0.32356671740233384,
      "grad_norm": 6.03125,
      "learning_rate": 9.662249724456346e-06,
      "loss": 1.62411652,
      "memory(GiB)": 97.17,
      "step": 12755,
      "train_speed(iter/s)": 1.638292
    },
    {
      "acc": 0.65363932,
      "epoch": 0.3236935565702689,
      "grad_norm": 6.34375,
      "learning_rate": 9.661870755046278e-06,
      "loss": 1.64809647,
      "memory(GiB)": 97.17,
      "step": 12760,
      "train_speed(iter/s)": 1.63836
    },
    {
      "acc": 0.64653487,
      "epoch": 0.323820395738204,
      "grad_norm": 4.84375,
      "learning_rate": 9.661491580586322e-06,
      "loss": 1.58892832,
      "memory(GiB)": 97.17,
      "step": 12765,
      "train_speed(iter/s)": 1.638417
    },
    {
      "acc": 0.64173794,
      "epoch": 0.323947234906139,
      "grad_norm": 6.4375,
      "learning_rate": 9.661112201093156e-06,
      "loss": 1.65680523,
      "memory(GiB)": 97.17,
      "step": 12770,
      "train_speed(iter/s)": 1.638486
    },
    {
      "acc": 0.64310226,
      "epoch": 0.32407407407407407,
      "grad_norm": 5.71875,
      "learning_rate": 9.660732616583463e-06,
      "loss": 1.6770319,
      "memory(GiB)": 97.17,
      "step": 12775,
      "train_speed(iter/s)": 1.638548
    },
    {
      "acc": 0.66438527,
      "epoch": 0.3242009132420091,
      "grad_norm": 5.6875,
      "learning_rate": 9.660352827073941e-06,
      "loss": 1.60030174,
      "memory(GiB)": 97.17,
      "step": 12780,
      "train_speed(iter/s)": 1.63861
    },
    {
      "acc": 0.63418293,
      "epoch": 0.3243277524099442,
      "grad_norm": 5.3125,
      "learning_rate": 9.659972832581295e-06,
      "loss": 1.66988316,
      "memory(GiB)": 97.17,
      "step": 12785,
      "train_speed(iter/s)": 1.638677
    },
    {
      "acc": 0.64672856,
      "epoch": 0.32445459157787926,
      "grad_norm": 5.46875,
      "learning_rate": 9.659592633122238e-06,
      "loss": 1.62647552,
      "memory(GiB)": 97.17,
      "step": 12790,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.64901643,
      "epoch": 0.3245814307458143,
      "grad_norm": 8.625,
      "learning_rate": 9.659212228713495e-06,
      "loss": 1.66549568,
      "memory(GiB)": 97.17,
      "step": 12795,
      "train_speed(iter/s)": 1.638812
    },
    {
      "acc": 0.6582293,
      "epoch": 0.32470826991374935,
      "grad_norm": 6.125,
      "learning_rate": 9.658831619371793e-06,
      "loss": 1.61192818,
      "memory(GiB)": 97.17,
      "step": 12800,
      "train_speed(iter/s)": 1.638876
    },
    {
      "acc": 0.64622264,
      "epoch": 0.32483510908168445,
      "grad_norm": 6.21875,
      "learning_rate": 9.658450805113879e-06,
      "loss": 1.70059204,
      "memory(GiB)": 97.17,
      "step": 12805,
      "train_speed(iter/s)": 1.638943
    },
    {
      "acc": 0.66009068,
      "epoch": 0.3249619482496195,
      "grad_norm": 4.78125,
      "learning_rate": 9.6580697859565e-06,
      "loss": 1.58308096,
      "memory(GiB)": 97.17,
      "step": 12810,
      "train_speed(iter/s)": 1.639009
    },
    {
      "acc": 0.67253685,
      "epoch": 0.32508878741755454,
      "grad_norm": 4.84375,
      "learning_rate": 9.657688561916414e-06,
      "loss": 1.49973269,
      "memory(GiB)": 97.17,
      "step": 12815,
      "train_speed(iter/s)": 1.639066
    },
    {
      "acc": 0.66676235,
      "epoch": 0.3252156265854896,
      "grad_norm": 6.03125,
      "learning_rate": 9.65730713301039e-06,
      "loss": 1.57210884,
      "memory(GiB)": 97.17,
      "step": 12820,
      "train_speed(iter/s)": 1.639131
    },
    {
      "acc": 0.64397898,
      "epoch": 0.3253424657534247,
      "grad_norm": 6.375,
      "learning_rate": 9.656925499255206e-06,
      "loss": 1.63543816,
      "memory(GiB)": 97.17,
      "step": 12825,
      "train_speed(iter/s)": 1.639201
    },
    {
      "acc": 0.65438166,
      "epoch": 0.3254693049213597,
      "grad_norm": 5.0625,
      "learning_rate": 9.656543660667646e-06,
      "loss": 1.59673157,
      "memory(GiB)": 97.17,
      "step": 12830,
      "train_speed(iter/s)": 1.639267
    },
    {
      "acc": 0.66472788,
      "epoch": 0.32559614408929477,
      "grad_norm": 5.75,
      "learning_rate": 9.656161617264507e-06,
      "loss": 1.5379261,
      "memory(GiB)": 97.17,
      "step": 12835,
      "train_speed(iter/s)": 1.639332
    },
    {
      "acc": 0.63419309,
      "epoch": 0.3257229832572298,
      "grad_norm": 6.625,
      "learning_rate": 9.65577936906259e-06,
      "loss": 1.64864674,
      "memory(GiB)": 97.17,
      "step": 12840,
      "train_speed(iter/s)": 1.639392
    },
    {
      "acc": 0.63277168,
      "epoch": 0.3258498224251649,
      "grad_norm": 6.90625,
      "learning_rate": 9.65539691607871e-06,
      "loss": 1.64743996,
      "memory(GiB)": 97.17,
      "step": 12845,
      "train_speed(iter/s)": 1.639456
    },
    {
      "acc": 0.64508448,
      "epoch": 0.32597666159309996,
      "grad_norm": 4.8125,
      "learning_rate": 9.65501425832969e-06,
      "loss": 1.677705,
      "memory(GiB)": 97.17,
      "step": 12850,
      "train_speed(iter/s)": 1.639514
    },
    {
      "acc": 0.64293809,
      "epoch": 0.326103500761035,
      "grad_norm": 5.3125,
      "learning_rate": 9.65463139583236e-06,
      "loss": 1.65923309,
      "memory(GiB)": 97.17,
      "step": 12855,
      "train_speed(iter/s)": 1.639579
    },
    {
      "acc": 0.64264169,
      "epoch": 0.32623033992897005,
      "grad_norm": 5.6875,
      "learning_rate": 9.65424832860356e-06,
      "loss": 1.65921211,
      "memory(GiB)": 97.17,
      "step": 12860,
      "train_speed(iter/s)": 1.639647
    },
    {
      "acc": 0.64991336,
      "epoch": 0.32635717909690515,
      "grad_norm": 5.625,
      "learning_rate": 9.653865056660136e-06,
      "loss": 1.64876823,
      "memory(GiB)": 97.17,
      "step": 12865,
      "train_speed(iter/s)": 1.63971
    },
    {
      "acc": 0.64246111,
      "epoch": 0.3264840182648402,
      "grad_norm": 5.5625,
      "learning_rate": 9.653481580018951e-06,
      "loss": 1.6958704,
      "memory(GiB)": 97.17,
      "step": 12870,
      "train_speed(iter/s)": 1.639775
    },
    {
      "acc": 0.6485539,
      "epoch": 0.32661085743277524,
      "grad_norm": 6.375,
      "learning_rate": 9.653097898696869e-06,
      "loss": 1.69870033,
      "memory(GiB)": 97.17,
      "step": 12875,
      "train_speed(iter/s)": 1.639837
    },
    {
      "acc": 0.64272871,
      "epoch": 0.3267376966007103,
      "grad_norm": 6.46875,
      "learning_rate": 9.652714012710766e-06,
      "loss": 1.67902279,
      "memory(GiB)": 97.17,
      "step": 12880,
      "train_speed(iter/s)": 1.639903
    },
    {
      "acc": 0.64827604,
      "epoch": 0.3268645357686454,
      "grad_norm": 6.875,
      "learning_rate": 9.65232992207753e-06,
      "loss": 1.7218359,
      "memory(GiB)": 97.17,
      "step": 12885,
      "train_speed(iter/s)": 1.639972
    },
    {
      "acc": 0.64713454,
      "epoch": 0.3269913749365804,
      "grad_norm": 4.75,
      "learning_rate": 9.651945626814052e-06,
      "loss": 1.68011665,
      "memory(GiB)": 97.17,
      "step": 12890,
      "train_speed(iter/s)": 1.64004
    },
    {
      "acc": 0.65753012,
      "epoch": 0.32711821410451547,
      "grad_norm": 6.53125,
      "learning_rate": 9.651561126937236e-06,
      "loss": 1.61671028,
      "memory(GiB)": 97.17,
      "step": 12895,
      "train_speed(iter/s)": 1.640102
    },
    {
      "acc": 0.64125328,
      "epoch": 0.3272450532724505,
      "grad_norm": 6.625,
      "learning_rate": 9.651176422463994e-06,
      "loss": 1.71301308,
      "memory(GiB)": 97.17,
      "step": 12900,
      "train_speed(iter/s)": 1.640165
    },
    {
      "acc": 0.65696259,
      "epoch": 0.3273718924403856,
      "grad_norm": 4.75,
      "learning_rate": 9.650791513411246e-06,
      "loss": 1.59470024,
      "memory(GiB)": 97.17,
      "step": 12905,
      "train_speed(iter/s)": 1.640227
    },
    {
      "acc": 0.63237667,
      "epoch": 0.32749873160832066,
      "grad_norm": 4.65625,
      "learning_rate": 9.650406399795924e-06,
      "loss": 1.69231205,
      "memory(GiB)": 97.17,
      "step": 12910,
      "train_speed(iter/s)": 1.640295
    },
    {
      "acc": 0.64365764,
      "epoch": 0.3276255707762557,
      "grad_norm": 7.40625,
      "learning_rate": 9.650021081634965e-06,
      "loss": 1.68579178,
      "memory(GiB)": 97.17,
      "step": 12915,
      "train_speed(iter/s)": 1.640364
    },
    {
      "acc": 0.63840742,
      "epoch": 0.32775240994419075,
      "grad_norm": 5.8125,
      "learning_rate": 9.649635558945318e-06,
      "loss": 1.65430431,
      "memory(GiB)": 97.17,
      "step": 12920,
      "train_speed(iter/s)": 1.640429
    },
    {
      "acc": 0.64144392,
      "epoch": 0.32787924911212585,
      "grad_norm": 4.59375,
      "learning_rate": 9.649249831743941e-06,
      "loss": 1.68406563,
      "memory(GiB)": 97.17,
      "step": 12925,
      "train_speed(iter/s)": 1.640488
    },
    {
      "acc": 0.64603863,
      "epoch": 0.3280060882800609,
      "grad_norm": 5.34375,
      "learning_rate": 9.6488639000478e-06,
      "loss": 1.68732719,
      "memory(GiB)": 97.17,
      "step": 12930,
      "train_speed(iter/s)": 1.640549
    },
    {
      "acc": 0.6613368,
      "epoch": 0.32813292744799594,
      "grad_norm": 5.65625,
      "learning_rate": 9.648477763873868e-06,
      "loss": 1.64164238,
      "memory(GiB)": 97.17,
      "step": 12935,
      "train_speed(iter/s)": 1.640607
    },
    {
      "acc": 0.6546916,
      "epoch": 0.328259766615931,
      "grad_norm": 5.75,
      "learning_rate": 9.648091423239128e-06,
      "loss": 1.5988987,
      "memory(GiB)": 97.17,
      "step": 12940,
      "train_speed(iter/s)": 1.640673
    },
    {
      "acc": 0.65834322,
      "epoch": 0.3283866057838661,
      "grad_norm": 5.59375,
      "learning_rate": 9.647704878160576e-06,
      "loss": 1.57230587,
      "memory(GiB)": 97.17,
      "step": 12945,
      "train_speed(iter/s)": 1.640735
    },
    {
      "acc": 0.63474984,
      "epoch": 0.3285134449518011,
      "grad_norm": 5.8125,
      "learning_rate": 9.647318128655213e-06,
      "loss": 1.68502693,
      "memory(GiB)": 97.17,
      "step": 12950,
      "train_speed(iter/s)": 1.640797
    },
    {
      "acc": 0.66642752,
      "epoch": 0.32864028411973617,
      "grad_norm": 4.84375,
      "learning_rate": 9.64693117474005e-06,
      "loss": 1.58760185,
      "memory(GiB)": 97.17,
      "step": 12955,
      "train_speed(iter/s)": 1.64085
    },
    {
      "acc": 0.65560608,
      "epoch": 0.3287671232876712,
      "grad_norm": 5.84375,
      "learning_rate": 9.646544016432109e-06,
      "loss": 1.64631081,
      "memory(GiB)": 97.17,
      "step": 12960,
      "train_speed(iter/s)": 1.640916
    },
    {
      "acc": 0.66386285,
      "epoch": 0.3288939624556063,
      "grad_norm": 6.15625,
      "learning_rate": 9.646156653748415e-06,
      "loss": 1.61627159,
      "memory(GiB)": 97.17,
      "step": 12965,
      "train_speed(iter/s)": 1.640978
    },
    {
      "acc": 0.65677385,
      "epoch": 0.32902080162354136,
      "grad_norm": 5.84375,
      "learning_rate": 9.645769086706008e-06,
      "loss": 1.690905,
      "memory(GiB)": 97.17,
      "step": 12970,
      "train_speed(iter/s)": 1.641036
    },
    {
      "acc": 0.65197349,
      "epoch": 0.3291476407914764,
      "grad_norm": 4.5625,
      "learning_rate": 9.645381315321934e-06,
      "loss": 1.63398762,
      "memory(GiB)": 97.17,
      "step": 12975,
      "train_speed(iter/s)": 1.641102
    },
    {
      "acc": 0.64948888,
      "epoch": 0.32927447995941145,
      "grad_norm": 4.84375,
      "learning_rate": 9.64499333961325e-06,
      "loss": 1.66944027,
      "memory(GiB)": 97.17,
      "step": 12980,
      "train_speed(iter/s)": 1.641166
    },
    {
      "acc": 0.66175227,
      "epoch": 0.32940131912734655,
      "grad_norm": 5.09375,
      "learning_rate": 9.64460515959702e-06,
      "loss": 1.53600693,
      "memory(GiB)": 97.17,
      "step": 12985,
      "train_speed(iter/s)": 1.64123
    },
    {
      "acc": 0.64427519,
      "epoch": 0.3295281582952816,
      "grad_norm": 6.0625,
      "learning_rate": 9.64421677529032e-06,
      "loss": 1.65408115,
      "memory(GiB)": 97.17,
      "step": 12990,
      "train_speed(iter/s)": 1.641296
    },
    {
      "acc": 0.6672967,
      "epoch": 0.32965499746321664,
      "grad_norm": 5.09375,
      "learning_rate": 9.64382818671023e-06,
      "loss": 1.55482807,
      "memory(GiB)": 97.17,
      "step": 12995,
      "train_speed(iter/s)": 1.641357
    },
    {
      "acc": 0.65891047,
      "epoch": 0.3297818366311517,
      "grad_norm": 6.28125,
      "learning_rate": 9.643439393873844e-06,
      "loss": 1.57503719,
      "memory(GiB)": 97.17,
      "step": 13000,
      "train_speed(iter/s)": 1.641425
    },
    {
      "epoch": 0.3297818366311517,
      "eval_acc": 0.6410164203509607,
      "eval_loss": 1.6034586429595947,
      "eval_runtime": 58.9433,
      "eval_samples_per_second": 108.07,
      "eval_steps_per_second": 27.026,
      "step": 13000
    },
    {
      "acc": 0.65530634,
      "epoch": 0.3299086757990868,
      "grad_norm": 5.15625,
      "learning_rate": 9.643050396798262e-06,
      "loss": 1.66781387,
      "memory(GiB)": 97.17,
      "step": 13005,
      "train_speed(iter/s)": 1.628505
    },
    {
      "acc": 0.66238599,
      "epoch": 0.3300355149670218,
      "grad_norm": 8.125,
      "learning_rate": 9.642661195500593e-06,
      "loss": 1.53265553,
      "memory(GiB)": 97.17,
      "step": 13010,
      "train_speed(iter/s)": 1.628571
    },
    {
      "acc": 0.6384572,
      "epoch": 0.33016235413495687,
      "grad_norm": 6.15625,
      "learning_rate": 9.642271789997956e-06,
      "loss": 1.6549612,
      "memory(GiB)": 97.17,
      "step": 13015,
      "train_speed(iter/s)": 1.628634
    },
    {
      "acc": 0.65347447,
      "epoch": 0.3302891933028919,
      "grad_norm": 4.8125,
      "learning_rate": 9.64188218030748e-06,
      "loss": 1.56046085,
      "memory(GiB)": 97.17,
      "step": 13020,
      "train_speed(iter/s)": 1.628699
    },
    {
      "acc": 0.66249228,
      "epoch": 0.330416032470827,
      "grad_norm": 5.1875,
      "learning_rate": 9.641492366446301e-06,
      "loss": 1.62477837,
      "memory(GiB)": 97.17,
      "step": 13025,
      "train_speed(iter/s)": 1.628764
    },
    {
      "acc": 0.64906745,
      "epoch": 0.33054287163876206,
      "grad_norm": 5.09375,
      "learning_rate": 9.641102348431565e-06,
      "loss": 1.62433739,
      "memory(GiB)": 97.17,
      "step": 13030,
      "train_speed(iter/s)": 1.628827
    },
    {
      "acc": 0.64215651,
      "epoch": 0.3306697108066971,
      "grad_norm": 5.34375,
      "learning_rate": 9.640712126280429e-06,
      "loss": 1.65913658,
      "memory(GiB)": 97.17,
      "step": 13035,
      "train_speed(iter/s)": 1.62889
    },
    {
      "acc": 0.64990587,
      "epoch": 0.33079654997463215,
      "grad_norm": 5.125,
      "learning_rate": 9.640321700010053e-06,
      "loss": 1.61545315,
      "memory(GiB)": 97.17,
      "step": 13040,
      "train_speed(iter/s)": 1.628955
    },
    {
      "acc": 0.6462347,
      "epoch": 0.33092338914256725,
      "grad_norm": 5.5625,
      "learning_rate": 9.63993106963761e-06,
      "loss": 1.67008781,
      "memory(GiB)": 97.17,
      "step": 13045,
      "train_speed(iter/s)": 1.629019
    },
    {
      "acc": 0.64295092,
      "epoch": 0.3310502283105023,
      "grad_norm": 5.125,
      "learning_rate": 9.639540235180283e-06,
      "loss": 1.63737774,
      "memory(GiB)": 97.17,
      "step": 13050,
      "train_speed(iter/s)": 1.629081
    },
    {
      "acc": 0.66302996,
      "epoch": 0.33117706747843734,
      "grad_norm": 6.5625,
      "learning_rate": 9.639149196655263e-06,
      "loss": 1.59176388,
      "memory(GiB)": 97.17,
      "step": 13055,
      "train_speed(iter/s)": 1.629143
    },
    {
      "acc": 0.64858561,
      "epoch": 0.3313039066463724,
      "grad_norm": 5.3125,
      "learning_rate": 9.638757954079749e-06,
      "loss": 1.60935822,
      "memory(GiB)": 97.17,
      "step": 13060,
      "train_speed(iter/s)": 1.629206
    },
    {
      "acc": 0.65634975,
      "epoch": 0.3314307458143075,
      "grad_norm": 7.15625,
      "learning_rate": 9.638366507470948e-06,
      "loss": 1.60039139,
      "memory(GiB)": 97.17,
      "step": 13065,
      "train_speed(iter/s)": 1.629272
    },
    {
      "acc": 0.65855961,
      "epoch": 0.3315575849822425,
      "grad_norm": 5.9375,
      "learning_rate": 9.637974856846082e-06,
      "loss": 1.62336674,
      "memory(GiB)": 97.17,
      "step": 13070,
      "train_speed(iter/s)": 1.629333
    },
    {
      "acc": 0.65982628,
      "epoch": 0.33168442415017757,
      "grad_norm": 6.21875,
      "learning_rate": 9.637583002222373e-06,
      "loss": 1.54557648,
      "memory(GiB)": 97.17,
      "step": 13075,
      "train_speed(iter/s)": 1.629392
    },
    {
      "acc": 0.65409746,
      "epoch": 0.3318112633181126,
      "grad_norm": 5.4375,
      "learning_rate": 9.637190943617059e-06,
      "loss": 1.63411102,
      "memory(GiB)": 97.17,
      "step": 13080,
      "train_speed(iter/s)": 1.629456
    },
    {
      "acc": 0.64977961,
      "epoch": 0.3319381024860477,
      "grad_norm": 6.21875,
      "learning_rate": 9.636798681047383e-06,
      "loss": 1.59947138,
      "memory(GiB)": 97.17,
      "step": 13085,
      "train_speed(iter/s)": 1.629506
    },
    {
      "acc": 0.6693881,
      "epoch": 0.33206494165398276,
      "grad_norm": 6.65625,
      "learning_rate": 9.6364062145306e-06,
      "loss": 1.602742,
      "memory(GiB)": 97.17,
      "step": 13090,
      "train_speed(iter/s)": 1.62957
    },
    {
      "acc": 0.6565311,
      "epoch": 0.3321917808219178,
      "grad_norm": 4.75,
      "learning_rate": 9.636013544083971e-06,
      "loss": 1.62767162,
      "memory(GiB)": 97.17,
      "step": 13095,
      "train_speed(iter/s)": 1.629627
    },
    {
      "acc": 0.6459856,
      "epoch": 0.33231861998985285,
      "grad_norm": 5.5,
      "learning_rate": 9.635620669724768e-06,
      "loss": 1.65949135,
      "memory(GiB)": 97.17,
      "step": 13100,
      "train_speed(iter/s)": 1.629699
    },
    {
      "acc": 0.65224152,
      "epoch": 0.33244545915778795,
      "grad_norm": 5.6875,
      "learning_rate": 9.635227591470272e-06,
      "loss": 1.65984306,
      "memory(GiB)": 97.17,
      "step": 13105,
      "train_speed(iter/s)": 1.629768
    },
    {
      "acc": 0.66255636,
      "epoch": 0.332572298325723,
      "grad_norm": 5.5625,
      "learning_rate": 9.63483430933777e-06,
      "loss": 1.59273062,
      "memory(GiB)": 97.17,
      "step": 13110,
      "train_speed(iter/s)": 1.629828
    },
    {
      "acc": 0.65655255,
      "epoch": 0.33269913749365804,
      "grad_norm": 5.8125,
      "learning_rate": 9.634440823344565e-06,
      "loss": 1.65388527,
      "memory(GiB)": 97.17,
      "step": 13115,
      "train_speed(iter/s)": 1.629894
    },
    {
      "acc": 0.66799397,
      "epoch": 0.3328259766615931,
      "grad_norm": 6.03125,
      "learning_rate": 9.634047133507959e-06,
      "loss": 1.55609608,
      "memory(GiB)": 97.17,
      "step": 13120,
      "train_speed(iter/s)": 1.629964
    },
    {
      "acc": 0.66454654,
      "epoch": 0.3329528158295282,
      "grad_norm": 5.625,
      "learning_rate": 9.63365323984527e-06,
      "loss": 1.5684514,
      "memory(GiB)": 97.17,
      "step": 13125,
      "train_speed(iter/s)": 1.630027
    },
    {
      "acc": 0.66204042,
      "epoch": 0.3330796549974632,
      "grad_norm": 5.875,
      "learning_rate": 9.633259142373825e-06,
      "loss": 1.58791218,
      "memory(GiB)": 97.17,
      "step": 13130,
      "train_speed(iter/s)": 1.630093
    },
    {
      "acc": 0.63796101,
      "epoch": 0.33320649416539827,
      "grad_norm": 6.21875,
      "learning_rate": 9.632864841110957e-06,
      "loss": 1.64469719,
      "memory(GiB)": 97.17,
      "step": 13135,
      "train_speed(iter/s)": 1.630156
    },
    {
      "acc": 0.6486742,
      "epoch": 0.3333333333333333,
      "grad_norm": 6.46875,
      "learning_rate": 9.632470336074009e-06,
      "loss": 1.69275875,
      "memory(GiB)": 97.17,
      "step": 13140,
      "train_speed(iter/s)": 1.630225
    },
    {
      "acc": 0.65933766,
      "epoch": 0.3334601725012684,
      "grad_norm": 6.125,
      "learning_rate": 9.632075627280333e-06,
      "loss": 1.57961845,
      "memory(GiB)": 97.17,
      "step": 13145,
      "train_speed(iter/s)": 1.630293
    },
    {
      "acc": 0.64420061,
      "epoch": 0.33358701166920346,
      "grad_norm": 5.71875,
      "learning_rate": 9.631680714747292e-06,
      "loss": 1.65871258,
      "memory(GiB)": 97.17,
      "step": 13150,
      "train_speed(iter/s)": 1.630361
    },
    {
      "acc": 0.66268587,
      "epoch": 0.3337138508371385,
      "grad_norm": 8.0625,
      "learning_rate": 9.63128559849225e-06,
      "loss": 1.57028494,
      "memory(GiB)": 97.17,
      "step": 13155,
      "train_speed(iter/s)": 1.630429
    },
    {
      "acc": 0.6471487,
      "epoch": 0.33384069000507355,
      "grad_norm": 6.6875,
      "learning_rate": 9.630890278532594e-06,
      "loss": 1.59411259,
      "memory(GiB)": 97.17,
      "step": 13160,
      "train_speed(iter/s)": 1.630498
    },
    {
      "acc": 0.6281713,
      "epoch": 0.33396752917300865,
      "grad_norm": 5.0625,
      "learning_rate": 9.630494754885706e-06,
      "loss": 1.72221336,
      "memory(GiB)": 97.17,
      "step": 13165,
      "train_speed(iter/s)": 1.630562
    },
    {
      "acc": 0.65375352,
      "epoch": 0.3340943683409437,
      "grad_norm": 5.375,
      "learning_rate": 9.630099027568986e-06,
      "loss": 1.61772003,
      "memory(GiB)": 97.17,
      "step": 13170,
      "train_speed(iter/s)": 1.630629
    },
    {
      "acc": 0.65680413,
      "epoch": 0.33422120750887874,
      "grad_norm": 5.3125,
      "learning_rate": 9.629703096599839e-06,
      "loss": 1.65999069,
      "memory(GiB)": 97.17,
      "step": 13175,
      "train_speed(iter/s)": 1.6307
    },
    {
      "acc": 0.65099554,
      "epoch": 0.3343480466768138,
      "grad_norm": 5.09375,
      "learning_rate": 9.629306961995678e-06,
      "loss": 1.63689957,
      "memory(GiB)": 97.17,
      "step": 13180,
      "train_speed(iter/s)": 1.630761
    },
    {
      "acc": 0.6357625,
      "epoch": 0.3344748858447489,
      "grad_norm": 5.40625,
      "learning_rate": 9.62891062377393e-06,
      "loss": 1.67008762,
      "memory(GiB)": 97.17,
      "step": 13185,
      "train_speed(iter/s)": 1.630828
    },
    {
      "acc": 0.6549716,
      "epoch": 0.3346017250126839,
      "grad_norm": 6.125,
      "learning_rate": 9.628514081952026e-06,
      "loss": 1.60908508,
      "memory(GiB)": 97.17,
      "step": 13190,
      "train_speed(iter/s)": 1.630891
    },
    {
      "acc": 0.62954302,
      "epoch": 0.33472856418061897,
      "grad_norm": 5.8125,
      "learning_rate": 9.628117336547408e-06,
      "loss": 1.72531414,
      "memory(GiB)": 97.17,
      "step": 13195,
      "train_speed(iter/s)": 1.63096
    },
    {
      "acc": 0.64352145,
      "epoch": 0.334855403348554,
      "grad_norm": 4.53125,
      "learning_rate": 9.627720387577525e-06,
      "loss": 1.6556284,
      "memory(GiB)": 97.17,
      "step": 13200,
      "train_speed(iter/s)": 1.631022
    },
    {
      "acc": 0.64894171,
      "epoch": 0.3349822425164891,
      "grad_norm": 6.46875,
      "learning_rate": 9.62732323505984e-06,
      "loss": 1.69405785,
      "memory(GiB)": 97.17,
      "step": 13205,
      "train_speed(iter/s)": 1.631092
    },
    {
      "acc": 0.65751781,
      "epoch": 0.33510908168442416,
      "grad_norm": 5.09375,
      "learning_rate": 9.62692587901182e-06,
      "loss": 1.57094851,
      "memory(GiB)": 97.17,
      "step": 13210,
      "train_speed(iter/s)": 1.631155
    },
    {
      "acc": 0.6518158,
      "epoch": 0.3352359208523592,
      "grad_norm": 5.90625,
      "learning_rate": 9.62652831945094e-06,
      "loss": 1.65788956,
      "memory(GiB)": 97.17,
      "step": 13215,
      "train_speed(iter/s)": 1.631202
    },
    {
      "acc": 0.6636971,
      "epoch": 0.33536276002029425,
      "grad_norm": 5.75,
      "learning_rate": 9.626130556394689e-06,
      "loss": 1.6660244,
      "memory(GiB)": 97.17,
      "step": 13220,
      "train_speed(iter/s)": 1.631264
    },
    {
      "acc": 0.66110892,
      "epoch": 0.33548959918822935,
      "grad_norm": 5.125,
      "learning_rate": 9.625732589860562e-06,
      "loss": 1.6646822,
      "memory(GiB)": 97.17,
      "step": 13225,
      "train_speed(iter/s)": 1.631333
    },
    {
      "acc": 0.63690033,
      "epoch": 0.3356164383561644,
      "grad_norm": 6.09375,
      "learning_rate": 9.625334419866064e-06,
      "loss": 1.71548882,
      "memory(GiB)": 97.17,
      "step": 13230,
      "train_speed(iter/s)": 1.631403
    },
    {
      "acc": 0.64992247,
      "epoch": 0.33574327752409944,
      "grad_norm": 5.625,
      "learning_rate": 9.624936046428708e-06,
      "loss": 1.61509514,
      "memory(GiB)": 97.17,
      "step": 13235,
      "train_speed(iter/s)": 1.631472
    },
    {
      "acc": 0.66669917,
      "epoch": 0.3358701166920345,
      "grad_norm": 5.65625,
      "learning_rate": 9.624537469566015e-06,
      "loss": 1.61054688,
      "memory(GiB)": 97.17,
      "step": 13240,
      "train_speed(iter/s)": 1.631541
    },
    {
      "acc": 0.65492659,
      "epoch": 0.3359969558599696,
      "grad_norm": 5.9375,
      "learning_rate": 9.624138689295516e-06,
      "loss": 1.58358002,
      "memory(GiB)": 97.17,
      "step": 13245,
      "train_speed(iter/s)": 1.631608
    },
    {
      "acc": 0.6432312,
      "epoch": 0.3361237950279046,
      "grad_norm": 5.53125,
      "learning_rate": 9.623739705634753e-06,
      "loss": 1.60147266,
      "memory(GiB)": 97.17,
      "step": 13250,
      "train_speed(iter/s)": 1.631678
    },
    {
      "acc": 0.65860195,
      "epoch": 0.33625063419583967,
      "grad_norm": 6.34375,
      "learning_rate": 9.623340518601274e-06,
      "loss": 1.63152981,
      "memory(GiB)": 97.17,
      "step": 13255,
      "train_speed(iter/s)": 1.631741
    },
    {
      "acc": 0.64756298,
      "epoch": 0.3363774733637747,
      "grad_norm": 7.3125,
      "learning_rate": 9.622941128212639e-06,
      "loss": 1.72195683,
      "memory(GiB)": 97.17,
      "step": 13260,
      "train_speed(iter/s)": 1.631811
    },
    {
      "acc": 0.65460567,
      "epoch": 0.3365043125317098,
      "grad_norm": 6.09375,
      "learning_rate": 9.622541534486411e-06,
      "loss": 1.61770439,
      "memory(GiB)": 97.17,
      "step": 13265,
      "train_speed(iter/s)": 1.631884
    },
    {
      "acc": 0.6486124,
      "epoch": 0.33663115169964486,
      "grad_norm": 6.625,
      "learning_rate": 9.62214173744017e-06,
      "loss": 1.64007339,
      "memory(GiB)": 97.17,
      "step": 13270,
      "train_speed(iter/s)": 1.631948
    },
    {
      "acc": 0.64311457,
      "epoch": 0.3367579908675799,
      "grad_norm": 5.90625,
      "learning_rate": 9.6217417370915e-06,
      "loss": 1.63710213,
      "memory(GiB)": 97.17,
      "step": 13275,
      "train_speed(iter/s)": 1.63201
    },
    {
      "acc": 0.65351524,
      "epoch": 0.33688483003551495,
      "grad_norm": 6.0625,
      "learning_rate": 9.62134153345799e-06,
      "loss": 1.59720058,
      "memory(GiB)": 97.17,
      "step": 13280,
      "train_speed(iter/s)": 1.632073
    },
    {
      "acc": 0.65047779,
      "epoch": 0.33701166920345005,
      "grad_norm": 5.96875,
      "learning_rate": 9.620941126557248e-06,
      "loss": 1.57065411,
      "memory(GiB)": 97.17,
      "step": 13285,
      "train_speed(iter/s)": 1.632136
    },
    {
      "acc": 0.66537809,
      "epoch": 0.3371385083713851,
      "grad_norm": 4.75,
      "learning_rate": 9.620540516406885e-06,
      "loss": 1.54066076,
      "memory(GiB)": 97.17,
      "step": 13290,
      "train_speed(iter/s)": 1.6322
    },
    {
      "acc": 0.64236665,
      "epoch": 0.33726534753932014,
      "grad_norm": 5.09375,
      "learning_rate": 9.620139703024522e-06,
      "loss": 1.62466125,
      "memory(GiB)": 97.17,
      "step": 13295,
      "train_speed(iter/s)": 1.63227
    },
    {
      "acc": 0.64019966,
      "epoch": 0.3373921867072552,
      "grad_norm": 6.46875,
      "learning_rate": 9.619738686427785e-06,
      "loss": 1.68296986,
      "memory(GiB)": 97.17,
      "step": 13300,
      "train_speed(iter/s)": 1.632339
    },
    {
      "acc": 0.6494771,
      "epoch": 0.3375190258751903,
      "grad_norm": 5.28125,
      "learning_rate": 9.619337466634317e-06,
      "loss": 1.61415253,
      "memory(GiB)": 97.17,
      "step": 13305,
      "train_speed(iter/s)": 1.632408
    },
    {
      "acc": 0.66605744,
      "epoch": 0.3376458650431253,
      "grad_norm": 5.75,
      "learning_rate": 9.618936043661762e-06,
      "loss": 1.58365126,
      "memory(GiB)": 97.17,
      "step": 13310,
      "train_speed(iter/s)": 1.632472
    },
    {
      "acc": 0.63992095,
      "epoch": 0.33777270421106037,
      "grad_norm": 7.96875,
      "learning_rate": 9.618534417527779e-06,
      "loss": 1.64826775,
      "memory(GiB)": 97.17,
      "step": 13315,
      "train_speed(iter/s)": 1.632534
    },
    {
      "acc": 0.65570612,
      "epoch": 0.3378995433789954,
      "grad_norm": 6.09375,
      "learning_rate": 9.61813258825003e-06,
      "loss": 1.5715662,
      "memory(GiB)": 97.17,
      "step": 13320,
      "train_speed(iter/s)": 1.632597
    },
    {
      "acc": 0.64682083,
      "epoch": 0.3380263825469305,
      "grad_norm": 5.53125,
      "learning_rate": 9.617730555846191e-06,
      "loss": 1.64066086,
      "memory(GiB)": 97.17,
      "step": 13325,
      "train_speed(iter/s)": 1.632659
    },
    {
      "acc": 0.64868999,
      "epoch": 0.33815322171486556,
      "grad_norm": 6.25,
      "learning_rate": 9.617328320333947e-06,
      "loss": 1.62098942,
      "memory(GiB)": 97.17,
      "step": 13330,
      "train_speed(iter/s)": 1.632724
    },
    {
      "acc": 0.67242332,
      "epoch": 0.3382800608828006,
      "grad_norm": 5.65625,
      "learning_rate": 9.616925881730989e-06,
      "loss": 1.51666565,
      "memory(GiB)": 97.17,
      "step": 13335,
      "train_speed(iter/s)": 1.632784
    },
    {
      "acc": 0.65879402,
      "epoch": 0.33840690005073565,
      "grad_norm": 5.40625,
      "learning_rate": 9.616523240055017e-06,
      "loss": 1.6054718,
      "memory(GiB)": 97.17,
      "step": 13340,
      "train_speed(iter/s)": 1.632849
    },
    {
      "acc": 0.63616529,
      "epoch": 0.33853373921867075,
      "grad_norm": 5.34375,
      "learning_rate": 9.616120395323743e-06,
      "loss": 1.69087791,
      "memory(GiB)": 97.17,
      "step": 13345,
      "train_speed(iter/s)": 1.632914
    },
    {
      "acc": 0.65138173,
      "epoch": 0.3386605783866058,
      "grad_norm": 5.09375,
      "learning_rate": 9.615717347554882e-06,
      "loss": 1.58977051,
      "memory(GiB)": 97.17,
      "step": 13350,
      "train_speed(iter/s)": 1.632978
    },
    {
      "acc": 0.63408489,
      "epoch": 0.33878741755454084,
      "grad_norm": 5.15625,
      "learning_rate": 9.615314096766166e-06,
      "loss": 1.6426239,
      "memory(GiB)": 97.17,
      "step": 13355,
      "train_speed(iter/s)": 1.633044
    },
    {
      "acc": 0.64465322,
      "epoch": 0.3389142567224759,
      "grad_norm": 6.5625,
      "learning_rate": 9.61491064297533e-06,
      "loss": 1.67834702,
      "memory(GiB)": 97.17,
      "step": 13360,
      "train_speed(iter/s)": 1.633109
    },
    {
      "acc": 0.64876709,
      "epoch": 0.339041095890411,
      "grad_norm": 6.09375,
      "learning_rate": 9.614506986200119e-06,
      "loss": 1.67453194,
      "memory(GiB)": 97.17,
      "step": 13365,
      "train_speed(iter/s)": 1.633175
    },
    {
      "acc": 0.64880538,
      "epoch": 0.339167935058346,
      "grad_norm": 7.1875,
      "learning_rate": 9.61410312645829e-06,
      "loss": 1.63719215,
      "memory(GiB)": 97.17,
      "step": 13370,
      "train_speed(iter/s)": 1.633238
    },
    {
      "acc": 0.65017805,
      "epoch": 0.33929477422628107,
      "grad_norm": 6.4375,
      "learning_rate": 9.613699063767603e-06,
      "loss": 1.67882271,
      "memory(GiB)": 97.17,
      "step": 13375,
      "train_speed(iter/s)": 1.633303
    },
    {
      "acc": 0.64997711,
      "epoch": 0.3394216133942161,
      "grad_norm": 5.65625,
      "learning_rate": 9.613294798145833e-06,
      "loss": 1.66933441,
      "memory(GiB)": 97.17,
      "step": 13380,
      "train_speed(iter/s)": 1.633369
    },
    {
      "acc": 0.6599268,
      "epoch": 0.3395484525621512,
      "grad_norm": 6.25,
      "learning_rate": 9.612890329610762e-06,
      "loss": 1.58168192,
      "memory(GiB)": 97.17,
      "step": 13385,
      "train_speed(iter/s)": 1.633432
    },
    {
      "acc": 0.65642452,
      "epoch": 0.33967529173008626,
      "grad_norm": 5.6875,
      "learning_rate": 9.612485658180178e-06,
      "loss": 1.62011013,
      "memory(GiB)": 97.17,
      "step": 13390,
      "train_speed(iter/s)": 1.6335
    },
    {
      "acc": 0.65850906,
      "epoch": 0.3398021308980213,
      "grad_norm": 5.34375,
      "learning_rate": 9.612080783871882e-06,
      "loss": 1.60980778,
      "memory(GiB)": 97.17,
      "step": 13395,
      "train_speed(iter/s)": 1.633563
    },
    {
      "acc": 0.65071831,
      "epoch": 0.33992897006595635,
      "grad_norm": 5.125,
      "learning_rate": 9.611675706703682e-06,
      "loss": 1.59563951,
      "memory(GiB)": 97.17,
      "step": 13400,
      "train_speed(iter/s)": 1.633627
    },
    {
      "acc": 0.64234285,
      "epoch": 0.34005580923389145,
      "grad_norm": 6.59375,
      "learning_rate": 9.611270426693395e-06,
      "loss": 1.67684059,
      "memory(GiB)": 97.17,
      "step": 13405,
      "train_speed(iter/s)": 1.63369
    },
    {
      "acc": 0.62538528,
      "epoch": 0.3401826484018265,
      "grad_norm": 6.65625,
      "learning_rate": 9.610864943858847e-06,
      "loss": 1.66686058,
      "memory(GiB)": 97.17,
      "step": 13410,
      "train_speed(iter/s)": 1.633758
    },
    {
      "acc": 0.65809946,
      "epoch": 0.34030948756976154,
      "grad_norm": 6.09375,
      "learning_rate": 9.61045925821787e-06,
      "loss": 1.62358341,
      "memory(GiB)": 97.17,
      "step": 13415,
      "train_speed(iter/s)": 1.633829
    },
    {
      "acc": 0.64416828,
      "epoch": 0.3404363267376966,
      "grad_norm": 4.9375,
      "learning_rate": 9.610053369788314e-06,
      "loss": 1.64812698,
      "memory(GiB)": 97.17,
      "step": 13420,
      "train_speed(iter/s)": 1.633891
    },
    {
      "acc": 0.64484787,
      "epoch": 0.3405631659056317,
      "grad_norm": 5.46875,
      "learning_rate": 9.609647278588027e-06,
      "loss": 1.72029305,
      "memory(GiB)": 97.17,
      "step": 13425,
      "train_speed(iter/s)": 1.633958
    },
    {
      "acc": 0.65401459,
      "epoch": 0.3406900050735667,
      "grad_norm": 4.34375,
      "learning_rate": 9.609240984634871e-06,
      "loss": 1.59744959,
      "memory(GiB)": 97.17,
      "step": 13430,
      "train_speed(iter/s)": 1.63402
    },
    {
      "acc": 0.65944738,
      "epoch": 0.34081684424150177,
      "grad_norm": 5.4375,
      "learning_rate": 9.608834487946719e-06,
      "loss": 1.63598385,
      "memory(GiB)": 97.17,
      "step": 13435,
      "train_speed(iter/s)": 1.634085
    },
    {
      "acc": 0.63552823,
      "epoch": 0.3409436834094368,
      "grad_norm": 6.1875,
      "learning_rate": 9.60842778854145e-06,
      "loss": 1.64073486,
      "memory(GiB)": 97.17,
      "step": 13440,
      "train_speed(iter/s)": 1.634152
    },
    {
      "acc": 0.65797157,
      "epoch": 0.3410705225773719,
      "grad_norm": 5.0625,
      "learning_rate": 9.60802088643695e-06,
      "loss": 1.61512642,
      "memory(GiB)": 97.17,
      "step": 13445,
      "train_speed(iter/s)": 1.634217
    },
    {
      "acc": 0.64022861,
      "epoch": 0.34119736174530696,
      "grad_norm": 4.75,
      "learning_rate": 9.60761378165112e-06,
      "loss": 1.6373558,
      "memory(GiB)": 97.17,
      "step": 13450,
      "train_speed(iter/s)": 1.63428
    },
    {
      "acc": 0.67306585,
      "epoch": 0.341324200913242,
      "grad_norm": 5.4375,
      "learning_rate": 9.607206474201863e-06,
      "loss": 1.58249474,
      "memory(GiB)": 97.17,
      "step": 13455,
      "train_speed(iter/s)": 1.634341
    },
    {
      "acc": 0.6459002,
      "epoch": 0.34145104008117705,
      "grad_norm": 5.875,
      "learning_rate": 9.606798964107096e-06,
      "loss": 1.69725952,
      "memory(GiB)": 97.17,
      "step": 13460,
      "train_speed(iter/s)": 1.634406
    },
    {
      "acc": 0.66193967,
      "epoch": 0.34157787924911215,
      "grad_norm": 6.15625,
      "learning_rate": 9.60639125138474e-06,
      "loss": 1.54741488,
      "memory(GiB)": 97.17,
      "step": 13465,
      "train_speed(iter/s)": 1.634474
    },
    {
      "acc": 0.65342951,
      "epoch": 0.3417047184170472,
      "grad_norm": 6.0625,
      "learning_rate": 9.605983336052735e-06,
      "loss": 1.63188744,
      "memory(GiB)": 97.17,
      "step": 13470,
      "train_speed(iter/s)": 1.634538
    },
    {
      "acc": 0.63303299,
      "epoch": 0.34183155758498224,
      "grad_norm": 5.4375,
      "learning_rate": 9.605575218129017e-06,
      "loss": 1.65524559,
      "memory(GiB)": 97.17,
      "step": 13475,
      "train_speed(iter/s)": 1.634607
    },
    {
      "acc": 0.63365688,
      "epoch": 0.3419583967529173,
      "grad_norm": 5.375,
      "learning_rate": 9.605166897631539e-06,
      "loss": 1.70623589,
      "memory(GiB)": 97.17,
      "step": 13480,
      "train_speed(iter/s)": 1.634672
    },
    {
      "acc": 0.65946827,
      "epoch": 0.3420852359208524,
      "grad_norm": 5.6875,
      "learning_rate": 9.604758374578259e-06,
      "loss": 1.60951061,
      "memory(GiB)": 97.17,
      "step": 13485,
      "train_speed(iter/s)": 1.634734
    },
    {
      "acc": 0.6486948,
      "epoch": 0.3422120750887874,
      "grad_norm": 6.9375,
      "learning_rate": 9.604349648987148e-06,
      "loss": 1.62078304,
      "memory(GiB)": 97.17,
      "step": 13490,
      "train_speed(iter/s)": 1.6348
    },
    {
      "acc": 0.64911952,
      "epoch": 0.34233891425672247,
      "grad_norm": 6.09375,
      "learning_rate": 9.603940720876181e-06,
      "loss": 1.66483097,
      "memory(GiB)": 97.17,
      "step": 13495,
      "train_speed(iter/s)": 1.634861
    },
    {
      "acc": 0.64871492,
      "epoch": 0.3424657534246575,
      "grad_norm": 5.59375,
      "learning_rate": 9.603531590263348e-06,
      "loss": 1.63723221,
      "memory(GiB)": 97.17,
      "step": 13500,
      "train_speed(iter/s)": 1.634923
    },
    {
      "acc": 0.64357762,
      "epoch": 0.3425925925925926,
      "grad_norm": 4.8125,
      "learning_rate": 9.603122257166641e-06,
      "loss": 1.64757099,
      "memory(GiB)": 97.17,
      "step": 13505,
      "train_speed(iter/s)": 1.634985
    },
    {
      "acc": 0.66512899,
      "epoch": 0.34271943176052766,
      "grad_norm": 5.375,
      "learning_rate": 9.602712721604066e-06,
      "loss": 1.58341637,
      "memory(GiB)": 97.17,
      "step": 13510,
      "train_speed(iter/s)": 1.635046
    },
    {
      "acc": 0.63635159,
      "epoch": 0.3428462709284627,
      "grad_norm": 6.21875,
      "learning_rate": 9.602302983593637e-06,
      "loss": 1.68867359,
      "memory(GiB)": 97.17,
      "step": 13515,
      "train_speed(iter/s)": 1.635115
    },
    {
      "acc": 0.64827604,
      "epoch": 0.34297311009639775,
      "grad_norm": 6.625,
      "learning_rate": 9.601893043153372e-06,
      "loss": 1.64419937,
      "memory(GiB)": 97.17,
      "step": 13520,
      "train_speed(iter/s)": 1.635181
    },
    {
      "acc": 0.64873228,
      "epoch": 0.34309994926433285,
      "grad_norm": 5.25,
      "learning_rate": 9.601482900301308e-06,
      "loss": 1.62032547,
      "memory(GiB)": 97.17,
      "step": 13525,
      "train_speed(iter/s)": 1.635243
    },
    {
      "acc": 0.65557809,
      "epoch": 0.3432267884322679,
      "grad_norm": 5.625,
      "learning_rate": 9.60107255505548e-06,
      "loss": 1.64220734,
      "memory(GiB)": 97.17,
      "step": 13530,
      "train_speed(iter/s)": 1.635311
    },
    {
      "acc": 0.63922901,
      "epoch": 0.34335362760020294,
      "grad_norm": 4.9375,
      "learning_rate": 9.60066200743394e-06,
      "loss": 1.75861874,
      "memory(GiB)": 97.17,
      "step": 13535,
      "train_speed(iter/s)": 1.635379
    },
    {
      "acc": 0.65102186,
      "epoch": 0.343480466768138,
      "grad_norm": 7.6875,
      "learning_rate": 9.600251257454744e-06,
      "loss": 1.61872768,
      "memory(GiB)": 97.17,
      "step": 13540,
      "train_speed(iter/s)": 1.635447
    },
    {
      "acc": 0.63279428,
      "epoch": 0.3436073059360731,
      "grad_norm": 5.15625,
      "learning_rate": 9.599840305135959e-06,
      "loss": 1.66210136,
      "memory(GiB)": 97.17,
      "step": 13545,
      "train_speed(iter/s)": 1.63551
    },
    {
      "acc": 0.65107059,
      "epoch": 0.3437341451040081,
      "grad_norm": 5.4375,
      "learning_rate": 9.59942915049566e-06,
      "loss": 1.56801577,
      "memory(GiB)": 97.17,
      "step": 13550,
      "train_speed(iter/s)": 1.635574
    },
    {
      "acc": 0.63942881,
      "epoch": 0.34386098427194317,
      "grad_norm": 5.3125,
      "learning_rate": 9.599017793551933e-06,
      "loss": 1.6313673,
      "memory(GiB)": 97.17,
      "step": 13555,
      "train_speed(iter/s)": 1.635642
    },
    {
      "acc": 0.6361475,
      "epoch": 0.3439878234398782,
      "grad_norm": 5.5625,
      "learning_rate": 9.598606234322869e-06,
      "loss": 1.74004631,
      "memory(GiB)": 97.17,
      "step": 13560,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.65669527,
      "epoch": 0.3441146626078133,
      "grad_norm": 5.15625,
      "learning_rate": 9.598194472826574e-06,
      "loss": 1.6196373,
      "memory(GiB)": 97.17,
      "step": 13565,
      "train_speed(iter/s)": 1.635769
    },
    {
      "acc": 0.64783645,
      "epoch": 0.34424150177574836,
      "grad_norm": 6.3125,
      "learning_rate": 9.597782509081154e-06,
      "loss": 1.65286465,
      "memory(GiB)": 97.17,
      "step": 13570,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.65066686,
      "epoch": 0.3443683409436834,
      "grad_norm": 4.9375,
      "learning_rate": 9.597370343104733e-06,
      "loss": 1.58463345,
      "memory(GiB)": 97.17,
      "step": 13575,
      "train_speed(iter/s)": 1.6359
    },
    {
      "acc": 0.6642622,
      "epoch": 0.34449518011161845,
      "grad_norm": 6.90625,
      "learning_rate": 9.596957974915438e-06,
      "loss": 1.5500495,
      "memory(GiB)": 97.17,
      "step": 13580,
      "train_speed(iter/s)": 1.635958
    },
    {
      "acc": 0.64914742,
      "epoch": 0.34462201927955355,
      "grad_norm": 4.40625,
      "learning_rate": 9.596545404531408e-06,
      "loss": 1.63522263,
      "memory(GiB)": 97.17,
      "step": 13585,
      "train_speed(iter/s)": 1.636017
    },
    {
      "acc": 0.63693576,
      "epoch": 0.3447488584474886,
      "grad_norm": 4.875,
      "learning_rate": 9.596132631970788e-06,
      "loss": 1.61731949,
      "memory(GiB)": 97.17,
      "step": 13590,
      "train_speed(iter/s)": 1.636079
    },
    {
      "acc": 0.66151857,
      "epoch": 0.34487569761542364,
      "grad_norm": 5.25,
      "learning_rate": 9.595719657251735e-06,
      "loss": 1.57577553,
      "memory(GiB)": 97.17,
      "step": 13595,
      "train_speed(iter/s)": 1.636145
    },
    {
      "acc": 0.66294804,
      "epoch": 0.3450025367833587,
      "grad_norm": 5.8125,
      "learning_rate": 9.595306480392413e-06,
      "loss": 1.57756386,
      "memory(GiB)": 97.17,
      "step": 13600,
      "train_speed(iter/s)": 1.636202
    },
    {
      "acc": 0.64581871,
      "epoch": 0.3451293759512938,
      "grad_norm": 6.96875,
      "learning_rate": 9.594893101410995e-06,
      "loss": 1.66540737,
      "memory(GiB)": 97.17,
      "step": 13605,
      "train_speed(iter/s)": 1.636264
    },
    {
      "acc": 0.64619398,
      "epoch": 0.3452562151192288,
      "grad_norm": 5.09375,
      "learning_rate": 9.594479520325665e-06,
      "loss": 1.63915825,
      "memory(GiB)": 97.17,
      "step": 13610,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.6353097,
      "epoch": 0.34538305428716387,
      "grad_norm": 5.90625,
      "learning_rate": 9.594065737154611e-06,
      "loss": 1.60954819,
      "memory(GiB)": 97.17,
      "step": 13615,
      "train_speed(iter/s)": 1.636384
    },
    {
      "acc": 0.65606308,
      "epoch": 0.3455098934550989,
      "grad_norm": 6.1875,
      "learning_rate": 9.593651751916037e-06,
      "loss": 1.59537392,
      "memory(GiB)": 97.17,
      "step": 13620,
      "train_speed(iter/s)": 1.636442
    },
    {
      "acc": 0.66525683,
      "epoch": 0.345636732623034,
      "grad_norm": 6.25,
      "learning_rate": 9.593237564628149e-06,
      "loss": 1.52252274,
      "memory(GiB)": 97.17,
      "step": 13625,
      "train_speed(iter/s)": 1.636507
    },
    {
      "acc": 0.66184902,
      "epoch": 0.34576357179096906,
      "grad_norm": 5.5625,
      "learning_rate": 9.592823175309164e-06,
      "loss": 1.50504761,
      "memory(GiB)": 97.17,
      "step": 13630,
      "train_speed(iter/s)": 1.63657
    },
    {
      "acc": 0.65679488,
      "epoch": 0.3458904109589041,
      "grad_norm": 4.625,
      "learning_rate": 9.592408583977311e-06,
      "loss": 1.59319515,
      "memory(GiB)": 97.17,
      "step": 13635,
      "train_speed(iter/s)": 1.636632
    },
    {
      "acc": 0.6349916,
      "epoch": 0.34601725012683915,
      "grad_norm": 5.6875,
      "learning_rate": 9.591993790650826e-06,
      "loss": 1.60629158,
      "memory(GiB)": 97.17,
      "step": 13640,
      "train_speed(iter/s)": 1.636695
    },
    {
      "acc": 0.66123581,
      "epoch": 0.34614408929477425,
      "grad_norm": 6.15625,
      "learning_rate": 9.591578795347952e-06,
      "loss": 1.66320419,
      "memory(GiB)": 97.17,
      "step": 13645,
      "train_speed(iter/s)": 1.636758
    },
    {
      "acc": 0.65161381,
      "epoch": 0.3462709284627093,
      "grad_norm": 6.4375,
      "learning_rate": 9.591163598086943e-06,
      "loss": 1.63374386,
      "memory(GiB)": 97.17,
      "step": 13650,
      "train_speed(iter/s)": 1.636825
    },
    {
      "acc": 0.65236111,
      "epoch": 0.34639776763064434,
      "grad_norm": 5.1875,
      "learning_rate": 9.59074819888606e-06,
      "loss": 1.64519844,
      "memory(GiB)": 97.17,
      "step": 13655,
      "train_speed(iter/s)": 1.636892
    },
    {
      "acc": 0.65328779,
      "epoch": 0.3465246067985794,
      "grad_norm": 5.375,
      "learning_rate": 9.590332597763575e-06,
      "loss": 1.59348421,
      "memory(GiB)": 97.17,
      "step": 13660,
      "train_speed(iter/s)": 1.636954
    },
    {
      "acc": 0.65040708,
      "epoch": 0.3466514459665145,
      "grad_norm": 5.5,
      "learning_rate": 9.589916794737768e-06,
      "loss": 1.67796631,
      "memory(GiB)": 97.17,
      "step": 13665,
      "train_speed(iter/s)": 1.637015
    },
    {
      "acc": 0.65535231,
      "epoch": 0.3467782851344495,
      "grad_norm": 5.84375,
      "learning_rate": 9.589500789826927e-06,
      "loss": 1.69573822,
      "memory(GiB)": 97.17,
      "step": 13670,
      "train_speed(iter/s)": 1.637075
    },
    {
      "acc": 0.6386796,
      "epoch": 0.34690512430238457,
      "grad_norm": 5.84375,
      "learning_rate": 9.589084583049353e-06,
      "loss": 1.73981438,
      "memory(GiB)": 97.17,
      "step": 13675,
      "train_speed(iter/s)": 1.637138
    },
    {
      "acc": 0.6500874,
      "epoch": 0.3470319634703196,
      "grad_norm": 6.59375,
      "learning_rate": 9.588668174423348e-06,
      "loss": 1.62884007,
      "memory(GiB)": 97.17,
      "step": 13680,
      "train_speed(iter/s)": 1.637203
    },
    {
      "acc": 0.64499726,
      "epoch": 0.3471588026382547,
      "grad_norm": 5.28125,
      "learning_rate": 9.588251563967232e-06,
      "loss": 1.70470772,
      "memory(GiB)": 97.17,
      "step": 13685,
      "train_speed(iter/s)": 1.63726
    },
    {
      "acc": 0.64006367,
      "epoch": 0.34728564180618976,
      "grad_norm": 5.28125,
      "learning_rate": 9.587834751699326e-06,
      "loss": 1.71168442,
      "memory(GiB)": 97.17,
      "step": 13690,
      "train_speed(iter/s)": 1.637322
    },
    {
      "acc": 0.64713793,
      "epoch": 0.3474124809741248,
      "grad_norm": 6.4375,
      "learning_rate": 9.587417737637963e-06,
      "loss": 1.67513752,
      "memory(GiB)": 97.17,
      "step": 13695,
      "train_speed(iter/s)": 1.637386
    },
    {
      "acc": 0.65330906,
      "epoch": 0.34753932014205985,
      "grad_norm": 5.65625,
      "learning_rate": 9.587000521801488e-06,
      "loss": 1.60202026,
      "memory(GiB)": 97.17,
      "step": 13700,
      "train_speed(iter/s)": 1.637444
    },
    {
      "acc": 0.65934496,
      "epoch": 0.34766615930999495,
      "grad_norm": 5.90625,
      "learning_rate": 9.58658310420825e-06,
      "loss": 1.61946106,
      "memory(GiB)": 97.17,
      "step": 13705,
      "train_speed(iter/s)": 1.637507
    },
    {
      "acc": 0.64947195,
      "epoch": 0.34779299847793,
      "grad_norm": 5.875,
      "learning_rate": 9.58616548487661e-06,
      "loss": 1.60600033,
      "memory(GiB)": 97.17,
      "step": 13710,
      "train_speed(iter/s)": 1.63757
    },
    {
      "acc": 0.6499331,
      "epoch": 0.34791983764586504,
      "grad_norm": 6.0,
      "learning_rate": 9.585747663824936e-06,
      "loss": 1.68437023,
      "memory(GiB)": 97.17,
      "step": 13715,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.65571041,
      "epoch": 0.3480466768138001,
      "grad_norm": 5.75,
      "learning_rate": 9.585329641071606e-06,
      "loss": 1.60613651,
      "memory(GiB)": 97.17,
      "step": 13720,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.65194998,
      "epoch": 0.3481735159817352,
      "grad_norm": 6.84375,
      "learning_rate": 9.584911416635007e-06,
      "loss": 1.63271618,
      "memory(GiB)": 97.17,
      "step": 13725,
      "train_speed(iter/s)": 1.637746
    },
    {
      "acc": 0.64861517,
      "epoch": 0.3483003551496702,
      "grad_norm": 5.0625,
      "learning_rate": 9.584492990533533e-06,
      "loss": 1.64106178,
      "memory(GiB)": 97.17,
      "step": 13730,
      "train_speed(iter/s)": 1.637806
    },
    {
      "acc": 0.65483723,
      "epoch": 0.34842719431760527,
      "grad_norm": 5.65625,
      "learning_rate": 9.58407436278559e-06,
      "loss": 1.5321784,
      "memory(GiB)": 97.17,
      "step": 13735,
      "train_speed(iter/s)": 1.637866
    },
    {
      "acc": 0.64420929,
      "epoch": 0.3485540334855403,
      "grad_norm": 5.1875,
      "learning_rate": 9.583655533409588e-06,
      "loss": 1.62746487,
      "memory(GiB)": 97.17,
      "step": 13740,
      "train_speed(iter/s)": 1.637924
    },
    {
      "acc": 0.6508462,
      "epoch": 0.3486808726534754,
      "grad_norm": 4.53125,
      "learning_rate": 9.583236502423952e-06,
      "loss": 1.63266335,
      "memory(GiB)": 97.17,
      "step": 13745,
      "train_speed(iter/s)": 1.637986
    },
    {
      "acc": 0.6501246,
      "epoch": 0.34880771182141046,
      "grad_norm": 5.59375,
      "learning_rate": 9.582817269847112e-06,
      "loss": 1.69108353,
      "memory(GiB)": 97.17,
      "step": 13750,
      "train_speed(iter/s)": 1.63805
    },
    {
      "acc": 0.65709333,
      "epoch": 0.3489345509893455,
      "grad_norm": 5.90625,
      "learning_rate": 9.582397835697509e-06,
      "loss": 1.57833624,
      "memory(GiB)": 97.17,
      "step": 13755,
      "train_speed(iter/s)": 1.638106
    },
    {
      "acc": 0.65828834,
      "epoch": 0.34906139015728055,
      "grad_norm": 5.0625,
      "learning_rate": 9.581978199993587e-06,
      "loss": 1.60937443,
      "memory(GiB)": 97.17,
      "step": 13760,
      "train_speed(iter/s)": 1.638162
    },
    {
      "acc": 0.64878697,
      "epoch": 0.34918822932521565,
      "grad_norm": 4.84375,
      "learning_rate": 9.58155836275381e-06,
      "loss": 1.6178339,
      "memory(GiB)": 97.17,
      "step": 13765,
      "train_speed(iter/s)": 1.638224
    },
    {
      "acc": 0.64724798,
      "epoch": 0.3493150684931507,
      "grad_norm": 6.0625,
      "learning_rate": 9.581138323996639e-06,
      "loss": 1.71036758,
      "memory(GiB)": 97.17,
      "step": 13770,
      "train_speed(iter/s)": 1.638288
    },
    {
      "acc": 0.66152434,
      "epoch": 0.34944190766108574,
      "grad_norm": 5.5,
      "learning_rate": 9.580718083740553e-06,
      "loss": 1.68143501,
      "memory(GiB)": 97.17,
      "step": 13775,
      "train_speed(iter/s)": 1.638352
    },
    {
      "acc": 0.68205414,
      "epoch": 0.3495687468290208,
      "grad_norm": 5.28125,
      "learning_rate": 9.580297642004032e-06,
      "loss": 1.54240494,
      "memory(GiB)": 97.17,
      "step": 13780,
      "train_speed(iter/s)": 1.638412
    },
    {
      "acc": 0.65155163,
      "epoch": 0.3496955859969559,
      "grad_norm": 4.90625,
      "learning_rate": 9.579876998805573e-06,
      "loss": 1.6943512,
      "memory(GiB)": 97.17,
      "step": 13785,
      "train_speed(iter/s)": 1.638475
    },
    {
      "acc": 0.65310907,
      "epoch": 0.3498224251648909,
      "grad_norm": 5.84375,
      "learning_rate": 9.579456154163676e-06,
      "loss": 1.64239845,
      "memory(GiB)": 97.17,
      "step": 13790,
      "train_speed(iter/s)": 1.638537
    },
    {
      "acc": 0.6457129,
      "epoch": 0.34994926433282597,
      "grad_norm": 5.28125,
      "learning_rate": 9.57903510809685e-06,
      "loss": 1.60808601,
      "memory(GiB)": 97.17,
      "step": 13795,
      "train_speed(iter/s)": 1.638596
    },
    {
      "acc": 0.65088377,
      "epoch": 0.350076103500761,
      "grad_norm": 6.28125,
      "learning_rate": 9.578613860623617e-06,
      "loss": 1.60699387,
      "memory(GiB)": 97.17,
      "step": 13800,
      "train_speed(iter/s)": 1.638658
    },
    {
      "acc": 0.66416445,
      "epoch": 0.3502029426686961,
      "grad_norm": 5.875,
      "learning_rate": 9.578192411762503e-06,
      "loss": 1.53634377,
      "memory(GiB)": 97.17,
      "step": 13805,
      "train_speed(iter/s)": 1.638723
    },
    {
      "acc": 0.64890122,
      "epoch": 0.35032978183663116,
      "grad_norm": 4.5625,
      "learning_rate": 9.577770761532049e-06,
      "loss": 1.66492214,
      "memory(GiB)": 97.17,
      "step": 13810,
      "train_speed(iter/s)": 1.638784
    },
    {
      "acc": 0.64871078,
      "epoch": 0.3504566210045662,
      "grad_norm": 6.375,
      "learning_rate": 9.577348909950797e-06,
      "loss": 1.65300541,
      "memory(GiB)": 97.17,
      "step": 13815,
      "train_speed(iter/s)": 1.638839
    },
    {
      "acc": 0.66125431,
      "epoch": 0.35058346017250125,
      "grad_norm": 5.34375,
      "learning_rate": 9.576926857037303e-06,
      "loss": 1.4876543,
      "memory(GiB)": 97.17,
      "step": 13820,
      "train_speed(iter/s)": 1.638902
    },
    {
      "acc": 0.66586304,
      "epoch": 0.35071029934043635,
      "grad_norm": 5.71875,
      "learning_rate": 9.576504602810133e-06,
      "loss": 1.63209648,
      "memory(GiB)": 97.17,
      "step": 13825,
      "train_speed(iter/s)": 1.638967
    },
    {
      "acc": 0.65355177,
      "epoch": 0.3508371385083714,
      "grad_norm": 5.28125,
      "learning_rate": 9.576082147287858e-06,
      "loss": 1.6247776,
      "memory(GiB)": 97.17,
      "step": 13830,
      "train_speed(iter/s)": 1.63903
    },
    {
      "acc": 0.65186634,
      "epoch": 0.35096397767630644,
      "grad_norm": 5.96875,
      "learning_rate": 9.575659490489058e-06,
      "loss": 1.58020687,
      "memory(GiB)": 97.17,
      "step": 13835,
      "train_speed(iter/s)": 1.639093
    },
    {
      "acc": 0.65629144,
      "epoch": 0.3510908168442415,
      "grad_norm": 5.5,
      "learning_rate": 9.575236632432325e-06,
      "loss": 1.63607521,
      "memory(GiB)": 97.17,
      "step": 13840,
      "train_speed(iter/s)": 1.639153
    },
    {
      "acc": 0.66264906,
      "epoch": 0.3512176560121766,
      "grad_norm": 6.125,
      "learning_rate": 9.574813573136259e-06,
      "loss": 1.62069511,
      "memory(GiB)": 97.17,
      "step": 13845,
      "train_speed(iter/s)": 1.639213
    },
    {
      "acc": 0.6588851,
      "epoch": 0.3513444951801116,
      "grad_norm": 4.78125,
      "learning_rate": 9.574390312619466e-06,
      "loss": 1.62078972,
      "memory(GiB)": 97.17,
      "step": 13850,
      "train_speed(iter/s)": 1.639271
    },
    {
      "acc": 0.66158595,
      "epoch": 0.35147133434804667,
      "grad_norm": 5.125,
      "learning_rate": 9.573966850900565e-06,
      "loss": 1.61515694,
      "memory(GiB)": 97.17,
      "step": 13855,
      "train_speed(iter/s)": 1.639328
    },
    {
      "acc": 0.64539061,
      "epoch": 0.3515981735159817,
      "grad_norm": 5.90625,
      "learning_rate": 9.57354318799818e-06,
      "loss": 1.58629074,
      "memory(GiB)": 97.17,
      "step": 13860,
      "train_speed(iter/s)": 1.639389
    },
    {
      "acc": 0.6522891,
      "epoch": 0.3517250126839168,
      "grad_norm": 5.46875,
      "learning_rate": 9.573119323930946e-06,
      "loss": 1.62132435,
      "memory(GiB)": 97.17,
      "step": 13865,
      "train_speed(iter/s)": 1.639449
    },
    {
      "acc": 0.65135703,
      "epoch": 0.35185185185185186,
      "grad_norm": 5.3125,
      "learning_rate": 9.572695258717507e-06,
      "loss": 1.64454937,
      "memory(GiB)": 97.17,
      "step": 13870,
      "train_speed(iter/s)": 1.63951
    },
    {
      "acc": 0.65078793,
      "epoch": 0.3519786910197869,
      "grad_norm": 5.25,
      "learning_rate": 9.572270992376513e-06,
      "loss": 1.6489296,
      "memory(GiB)": 97.17,
      "step": 13875,
      "train_speed(iter/s)": 1.639573
    },
    {
      "acc": 0.65659442,
      "epoch": 0.35210553018772195,
      "grad_norm": 5.03125,
      "learning_rate": 9.571846524926629e-06,
      "loss": 1.62971611,
      "memory(GiB)": 97.17,
      "step": 13880,
      "train_speed(iter/s)": 1.639633
    },
    {
      "acc": 0.65870566,
      "epoch": 0.35223236935565705,
      "grad_norm": 5.03125,
      "learning_rate": 9.571421856386522e-06,
      "loss": 1.67318134,
      "memory(GiB)": 97.17,
      "step": 13885,
      "train_speed(iter/s)": 1.639693
    },
    {
      "acc": 0.65042315,
      "epoch": 0.3523592085235921,
      "grad_norm": 5.25,
      "learning_rate": 9.570996986774872e-06,
      "loss": 1.63607979,
      "memory(GiB)": 97.17,
      "step": 13890,
      "train_speed(iter/s)": 1.638787
    },
    {
      "acc": 0.64825239,
      "epoch": 0.35248604769152714,
      "grad_norm": 6.875,
      "learning_rate": 9.570571916110366e-06,
      "loss": 1.66116676,
      "memory(GiB)": 97.17,
      "step": 13895,
      "train_speed(iter/s)": 1.638846
    },
    {
      "acc": 0.65396852,
      "epoch": 0.3526128868594622,
      "grad_norm": 5.09375,
      "learning_rate": 9.570146644411705e-06,
      "loss": 1.62385712,
      "memory(GiB)": 97.17,
      "step": 13900,
      "train_speed(iter/s)": 1.637907
    },
    {
      "acc": 0.65529709,
      "epoch": 0.3527397260273973,
      "grad_norm": 5.5625,
      "learning_rate": 9.569721171697587e-06,
      "loss": 1.63701954,
      "memory(GiB)": 97.17,
      "step": 13905,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.64981155,
      "epoch": 0.3528665651953323,
      "grad_norm": 5.21875,
      "learning_rate": 9.569295497986727e-06,
      "loss": 1.68096714,
      "memory(GiB)": 97.17,
      "step": 13910,
      "train_speed(iter/s)": 1.636975
    },
    {
      "acc": 0.66631312,
      "epoch": 0.35299340436326737,
      "grad_norm": 6.375,
      "learning_rate": 9.568869623297855e-06,
      "loss": 1.60080128,
      "memory(GiB)": 97.17,
      "step": 13915,
      "train_speed(iter/s)": 1.637038
    },
    {
      "acc": 0.64119601,
      "epoch": 0.3531202435312024,
      "grad_norm": 5.1875,
      "learning_rate": 9.568443547649697e-06,
      "loss": 1.72888298,
      "memory(GiB)": 97.17,
      "step": 13920,
      "train_speed(iter/s)": 1.637102
    },
    {
      "acc": 0.6534049,
      "epoch": 0.3532470826991375,
      "grad_norm": 6.125,
      "learning_rate": 9.568017271060994e-06,
      "loss": 1.64671173,
      "memory(GiB)": 97.17,
      "step": 13925,
      "train_speed(iter/s)": 1.637161
    },
    {
      "acc": 0.66404309,
      "epoch": 0.35337392186707256,
      "grad_norm": 7.125,
      "learning_rate": 9.567590793550498e-06,
      "loss": 1.58683472,
      "memory(GiB)": 97.17,
      "step": 13930,
      "train_speed(iter/s)": 1.637219
    },
    {
      "acc": 0.65419855,
      "epoch": 0.3535007610350076,
      "grad_norm": 6.40625,
      "learning_rate": 9.567164115136965e-06,
      "loss": 1.62939644,
      "memory(GiB)": 97.17,
      "step": 13935,
      "train_speed(iter/s)": 1.637283
    },
    {
      "acc": 0.66233878,
      "epoch": 0.35362760020294265,
      "grad_norm": 5.1875,
      "learning_rate": 9.566737235839166e-06,
      "loss": 1.61837234,
      "memory(GiB)": 97.17,
      "step": 13940,
      "train_speed(iter/s)": 1.637346
    },
    {
      "acc": 0.64619679,
      "epoch": 0.35375443937087775,
      "grad_norm": 6.5625,
      "learning_rate": 9.566310155675871e-06,
      "loss": 1.64337883,
      "memory(GiB)": 97.17,
      "step": 13945,
      "train_speed(iter/s)": 1.637411
    },
    {
      "acc": 0.64525166,
      "epoch": 0.3538812785388128,
      "grad_norm": 4.25,
      "learning_rate": 9.56588287466587e-06,
      "loss": 1.58622475,
      "memory(GiB)": 97.17,
      "step": 13950,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.6659534,
      "epoch": 0.35400811770674784,
      "grad_norm": 5.96875,
      "learning_rate": 9.565455392827954e-06,
      "loss": 1.58146667,
      "memory(GiB)": 97.17,
      "step": 13955,
      "train_speed(iter/s)": 1.637531
    },
    {
      "acc": 0.64464064,
      "epoch": 0.3541349568746829,
      "grad_norm": 5.15625,
      "learning_rate": 9.565027710180927e-06,
      "loss": 1.65124092,
      "memory(GiB)": 97.17,
      "step": 13960,
      "train_speed(iter/s)": 1.637589
    },
    {
      "acc": 0.66620688,
      "epoch": 0.354261796042618,
      "grad_norm": 5.9375,
      "learning_rate": 9.5645998267436e-06,
      "loss": 1.51923618,
      "memory(GiB)": 97.17,
      "step": 13965,
      "train_speed(iter/s)": 1.637652
    },
    {
      "acc": 0.64231234,
      "epoch": 0.354388635210553,
      "grad_norm": 4.90625,
      "learning_rate": 9.564171742534794e-06,
      "loss": 1.64211617,
      "memory(GiB)": 97.17,
      "step": 13970,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.65112343,
      "epoch": 0.35451547437848807,
      "grad_norm": 6.28125,
      "learning_rate": 9.563743457573336e-06,
      "loss": 1.67483883,
      "memory(GiB)": 97.17,
      "step": 13975,
      "train_speed(iter/s)": 1.637769
    },
    {
      "acc": 0.64085951,
      "epoch": 0.3546423135464231,
      "grad_norm": 5.1875,
      "learning_rate": 9.563314971878065e-06,
      "loss": 1.65576439,
      "memory(GiB)": 97.17,
      "step": 13980,
      "train_speed(iter/s)": 1.637828
    },
    {
      "acc": 0.64888682,
      "epoch": 0.3547691527143582,
      "grad_norm": 5.5625,
      "learning_rate": 9.562886285467828e-06,
      "loss": 1.6142662,
      "memory(GiB)": 97.17,
      "step": 13985,
      "train_speed(iter/s)": 1.637892
    },
    {
      "acc": 0.64186583,
      "epoch": 0.35489599188229326,
      "grad_norm": 5.75,
      "learning_rate": 9.56245739836148e-06,
      "loss": 1.6557024,
      "memory(GiB)": 97.17,
      "step": 13990,
      "train_speed(iter/s)": 1.637949
    },
    {
      "acc": 0.6515049,
      "epoch": 0.3550228310502283,
      "grad_norm": 6.21875,
      "learning_rate": 9.562028310577887e-06,
      "loss": 1.62775726,
      "memory(GiB)": 97.17,
      "step": 13995,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.64320784,
      "epoch": 0.35514967021816335,
      "grad_norm": 5.5625,
      "learning_rate": 9.56159902213592e-06,
      "loss": 1.65720387,
      "memory(GiB)": 97.17,
      "step": 14000,
      "train_speed(iter/s)": 1.638075
    },
    {
      "epoch": 0.35514967021816335,
      "eval_acc": 0.6414904658384445,
      "eval_loss": 1.6004287004470825,
      "eval_runtime": 59.2539,
      "eval_samples_per_second": 107.503,
      "eval_steps_per_second": 26.884,
      "step": 14000
    },
    {
      "acc": 0.65292993,
      "epoch": 0.35527650938609845,
      "grad_norm": 5.875,
      "learning_rate": 9.561169533054462e-06,
      "loss": 1.63042793,
      "memory(GiB)": 97.17,
      "step": 14005,
      "train_speed(iter/s)": 1.626063
    },
    {
      "acc": 0.66362543,
      "epoch": 0.3554033485540335,
      "grad_norm": 6.84375,
      "learning_rate": 9.560739843352404e-06,
      "loss": 1.62679939,
      "memory(GiB)": 97.17,
      "step": 14010,
      "train_speed(iter/s)": 1.626126
    },
    {
      "acc": 0.64804759,
      "epoch": 0.35553018772196854,
      "grad_norm": 6.09375,
      "learning_rate": 9.560309953048645e-06,
      "loss": 1.6837923,
      "memory(GiB)": 97.17,
      "step": 14015,
      "train_speed(iter/s)": 1.625266
    },
    {
      "acc": 0.63858299,
      "epoch": 0.3556570268899036,
      "grad_norm": 5.3125,
      "learning_rate": 9.559879862162095e-06,
      "loss": 1.69130516,
      "memory(GiB)": 97.17,
      "step": 14020,
      "train_speed(iter/s)": 1.625333
    },
    {
      "acc": 0.65001059,
      "epoch": 0.3557838660578387,
      "grad_norm": 5.46875,
      "learning_rate": 9.55944957071167e-06,
      "loss": 1.66215382,
      "memory(GiB)": 97.17,
      "step": 14025,
      "train_speed(iter/s)": 1.625399
    },
    {
      "acc": 0.64804235,
      "epoch": 0.3559107052257737,
      "grad_norm": 6.875,
      "learning_rate": 9.559019078716295e-06,
      "loss": 1.65577374,
      "memory(GiB)": 97.17,
      "step": 14030,
      "train_speed(iter/s)": 1.625471
    },
    {
      "acc": 0.65114422,
      "epoch": 0.35603754439370877,
      "grad_norm": 7.09375,
      "learning_rate": 9.558588386194907e-06,
      "loss": 1.68736877,
      "memory(GiB)": 97.17,
      "step": 14035,
      "train_speed(iter/s)": 1.625539
    },
    {
      "acc": 0.64714065,
      "epoch": 0.3561643835616438,
      "grad_norm": 6.875,
      "learning_rate": 9.55815749316645e-06,
      "loss": 1.59792538,
      "memory(GiB)": 97.17,
      "step": 14040,
      "train_speed(iter/s)": 1.625605
    },
    {
      "acc": 0.63453717,
      "epoch": 0.3562912227295789,
      "grad_norm": 5.5,
      "learning_rate": 9.557726399649875e-06,
      "loss": 1.66835327,
      "memory(GiB)": 97.17,
      "step": 14045,
      "train_speed(iter/s)": 1.625672
    },
    {
      "acc": 0.64820013,
      "epoch": 0.35641806189751396,
      "grad_norm": 7.1875,
      "learning_rate": 9.557295105664144e-06,
      "loss": 1.64806557,
      "memory(GiB)": 97.17,
      "step": 14050,
      "train_speed(iter/s)": 1.625739
    },
    {
      "acc": 0.63990793,
      "epoch": 0.356544901065449,
      "grad_norm": 5.8125,
      "learning_rate": 9.556863611228228e-06,
      "loss": 1.69207649,
      "memory(GiB)": 97.17,
      "step": 14055,
      "train_speed(iter/s)": 1.625806
    },
    {
      "acc": 0.65620365,
      "epoch": 0.35667174023338405,
      "grad_norm": 5.78125,
      "learning_rate": 9.556431916361105e-06,
      "loss": 1.6343586,
      "memory(GiB)": 97.17,
      "step": 14060,
      "train_speed(iter/s)": 1.625874
    },
    {
      "acc": 0.63775182,
      "epoch": 0.35679857940131915,
      "grad_norm": 5.3125,
      "learning_rate": 9.556000021081764e-06,
      "loss": 1.63741646,
      "memory(GiB)": 97.17,
      "step": 14065,
      "train_speed(iter/s)": 1.625941
    },
    {
      "acc": 0.65458846,
      "epoch": 0.3569254185692542,
      "grad_norm": 7.25,
      "learning_rate": 9.5555679254092e-06,
      "loss": 1.55885248,
      "memory(GiB)": 97.17,
      "step": 14070,
      "train_speed(iter/s)": 1.626004
    },
    {
      "acc": 0.64213009,
      "epoch": 0.35705225773718924,
      "grad_norm": 5.875,
      "learning_rate": 9.55513562936242e-06,
      "loss": 1.7240551,
      "memory(GiB)": 97.17,
      "step": 14075,
      "train_speed(iter/s)": 1.626071
    },
    {
      "acc": 0.63884268,
      "epoch": 0.3571790969051243,
      "grad_norm": 6.09375,
      "learning_rate": 9.554703132960437e-06,
      "loss": 1.74014511,
      "memory(GiB)": 97.17,
      "step": 14080,
      "train_speed(iter/s)": 1.626137
    },
    {
      "acc": 0.6602293,
      "epoch": 0.3573059360730594,
      "grad_norm": 5.09375,
      "learning_rate": 9.554270436222277e-06,
      "loss": 1.56141319,
      "memory(GiB)": 97.17,
      "step": 14085,
      "train_speed(iter/s)": 1.6262
    },
    {
      "acc": 0.64819026,
      "epoch": 0.3574327752409944,
      "grad_norm": 5.25,
      "learning_rate": 9.553837539166969e-06,
      "loss": 1.60231209,
      "memory(GiB)": 97.17,
      "step": 14090,
      "train_speed(iter/s)": 1.626265
    },
    {
      "acc": 0.64707174,
      "epoch": 0.35755961440892947,
      "grad_norm": 6.0,
      "learning_rate": 9.553404441813554e-06,
      "loss": 1.69079437,
      "memory(GiB)": 97.17,
      "step": 14095,
      "train_speed(iter/s)": 1.626332
    },
    {
      "acc": 0.6546186,
      "epoch": 0.3576864535768645,
      "grad_norm": 7.84375,
      "learning_rate": 9.552971144181083e-06,
      "loss": 1.58403511,
      "memory(GiB)": 97.17,
      "step": 14100,
      "train_speed(iter/s)": 1.626398
    },
    {
      "acc": 0.65222545,
      "epoch": 0.3578132927447996,
      "grad_norm": 5.28125,
      "learning_rate": 9.552537646288612e-06,
      "loss": 1.6568119,
      "memory(GiB)": 97.17,
      "step": 14105,
      "train_speed(iter/s)": 1.626462
    },
    {
      "acc": 0.66831675,
      "epoch": 0.35794013191273466,
      "grad_norm": 6.46875,
      "learning_rate": 9.552103948155211e-06,
      "loss": 1.57768946,
      "memory(GiB)": 97.17,
      "step": 14110,
      "train_speed(iter/s)": 1.626529
    },
    {
      "acc": 0.66277895,
      "epoch": 0.3580669710806697,
      "grad_norm": 6.59375,
      "learning_rate": 9.551670049799954e-06,
      "loss": 1.62563763,
      "memory(GiB)": 97.17,
      "step": 14115,
      "train_speed(iter/s)": 1.626598
    },
    {
      "acc": 0.64569521,
      "epoch": 0.35819381024860475,
      "grad_norm": 5.6875,
      "learning_rate": 9.551235951241927e-06,
      "loss": 1.62170048,
      "memory(GiB)": 97.17,
      "step": 14120,
      "train_speed(iter/s)": 1.626663
    },
    {
      "acc": 0.64057965,
      "epoch": 0.35832064941653985,
      "grad_norm": 6.03125,
      "learning_rate": 9.550801652500223e-06,
      "loss": 1.72616081,
      "memory(GiB)": 97.17,
      "step": 14125,
      "train_speed(iter/s)": 1.626714
    },
    {
      "acc": 0.65643935,
      "epoch": 0.3584474885844749,
      "grad_norm": 6.1875,
      "learning_rate": 9.550367153593944e-06,
      "loss": 1.64241409,
      "memory(GiB)": 97.17,
      "step": 14130,
      "train_speed(iter/s)": 1.626783
    },
    {
      "acc": 0.6423604,
      "epoch": 0.35857432775240994,
      "grad_norm": 4.875,
      "learning_rate": 9.549932454542202e-06,
      "loss": 1.64936028,
      "memory(GiB)": 97.17,
      "step": 14135,
      "train_speed(iter/s)": 1.626848
    },
    {
      "acc": 0.66251836,
      "epoch": 0.358701166920345,
      "grad_norm": 7.75,
      "learning_rate": 9.549497555364115e-06,
      "loss": 1.58814211,
      "memory(GiB)": 97.17,
      "step": 14140,
      "train_speed(iter/s)": 1.626917
    },
    {
      "acc": 0.65246286,
      "epoch": 0.3588280060882801,
      "grad_norm": 6.53125,
      "learning_rate": 9.549062456078816e-06,
      "loss": 1.60987892,
      "memory(GiB)": 97.17,
      "step": 14145,
      "train_speed(iter/s)": 1.626982
    },
    {
      "acc": 0.65147219,
      "epoch": 0.3589548452562151,
      "grad_norm": 5.5,
      "learning_rate": 9.54862715670544e-06,
      "loss": 1.62243233,
      "memory(GiB)": 97.17,
      "step": 14150,
      "train_speed(iter/s)": 1.62705
    },
    {
      "acc": 0.65065045,
      "epoch": 0.35908168442415017,
      "grad_norm": 4.9375,
      "learning_rate": 9.548191657263132e-06,
      "loss": 1.61188564,
      "memory(GiB)": 97.17,
      "step": 14155,
      "train_speed(iter/s)": 1.627118
    },
    {
      "acc": 0.66279345,
      "epoch": 0.3592085235920852,
      "grad_norm": 5.15625,
      "learning_rate": 9.547755957771049e-06,
      "loss": 1.62165356,
      "memory(GiB)": 97.17,
      "step": 14160,
      "train_speed(iter/s)": 1.627184
    },
    {
      "acc": 0.65560017,
      "epoch": 0.3593353627600203,
      "grad_norm": 5.03125,
      "learning_rate": 9.547320058248356e-06,
      "loss": 1.59486837,
      "memory(GiB)": 97.17,
      "step": 14165,
      "train_speed(iter/s)": 1.627253
    },
    {
      "acc": 0.65943356,
      "epoch": 0.35946220192795536,
      "grad_norm": 6.1875,
      "learning_rate": 9.546883958714223e-06,
      "loss": 1.66743088,
      "memory(GiB)": 97.17,
      "step": 14170,
      "train_speed(iter/s)": 1.627325
    },
    {
      "acc": 0.62036667,
      "epoch": 0.3595890410958904,
      "grad_norm": 5.75,
      "learning_rate": 9.546447659187834e-06,
      "loss": 1.71007957,
      "memory(GiB)": 97.17,
      "step": 14175,
      "train_speed(iter/s)": 1.627394
    },
    {
      "acc": 0.65520248,
      "epoch": 0.35971588026382545,
      "grad_norm": 5.15625,
      "learning_rate": 9.546011159688377e-06,
      "loss": 1.60203037,
      "memory(GiB)": 97.17,
      "step": 14180,
      "train_speed(iter/s)": 1.627459
    },
    {
      "acc": 0.66012106,
      "epoch": 0.35984271943176055,
      "grad_norm": 5.1875,
      "learning_rate": 9.545574460235055e-06,
      "loss": 1.63000755,
      "memory(GiB)": 97.17,
      "step": 14185,
      "train_speed(iter/s)": 1.627527
    },
    {
      "acc": 0.64758515,
      "epoch": 0.3599695585996956,
      "grad_norm": 5.4375,
      "learning_rate": 9.545137560847071e-06,
      "loss": 1.5833518,
      "memory(GiB)": 97.17,
      "step": 14190,
      "train_speed(iter/s)": 1.627596
    },
    {
      "acc": 0.64470348,
      "epoch": 0.36009639776763064,
      "grad_norm": 5.78125,
      "learning_rate": 9.544700461543647e-06,
      "loss": 1.64939404,
      "memory(GiB)": 97.17,
      "step": 14195,
      "train_speed(iter/s)": 1.627666
    },
    {
      "acc": 0.66848049,
      "epoch": 0.3602232369355657,
      "grad_norm": 5.28125,
      "learning_rate": 9.544263162344005e-06,
      "loss": 1.58201141,
      "memory(GiB)": 97.17,
      "step": 14200,
      "train_speed(iter/s)": 1.627735
    },
    {
      "acc": 0.64273958,
      "epoch": 0.3603500761035008,
      "grad_norm": 5.84375,
      "learning_rate": 9.54382566326738e-06,
      "loss": 1.68439293,
      "memory(GiB)": 97.17,
      "step": 14205,
      "train_speed(iter/s)": 1.627798
    },
    {
      "acc": 0.64344196,
      "epoch": 0.3604769152714358,
      "grad_norm": 5.0625,
      "learning_rate": 9.543387964333018e-06,
      "loss": 1.6235239,
      "memory(GiB)": 97.17,
      "step": 14210,
      "train_speed(iter/s)": 1.627864
    },
    {
      "acc": 0.63097067,
      "epoch": 0.36060375443937087,
      "grad_norm": 5.84375,
      "learning_rate": 9.542950065560165e-06,
      "loss": 1.6433712,
      "memory(GiB)": 97.17,
      "step": 14215,
      "train_speed(iter/s)": 1.627928
    },
    {
      "acc": 0.62943425,
      "epoch": 0.3607305936073059,
      "grad_norm": 5.125,
      "learning_rate": 9.542511966968087e-06,
      "loss": 1.74006004,
      "memory(GiB)": 97.17,
      "step": 14220,
      "train_speed(iter/s)": 1.627993
    },
    {
      "acc": 0.65337658,
      "epoch": 0.360857432775241,
      "grad_norm": 5.9375,
      "learning_rate": 9.542073668576052e-06,
      "loss": 1.55032177,
      "memory(GiB)": 97.17,
      "step": 14225,
      "train_speed(iter/s)": 1.628058
    },
    {
      "acc": 0.65861611,
      "epoch": 0.36098427194317606,
      "grad_norm": 4.59375,
      "learning_rate": 9.541635170403338e-06,
      "loss": 1.58314495,
      "memory(GiB)": 97.17,
      "step": 14230,
      "train_speed(iter/s)": 1.628127
    },
    {
      "acc": 0.64714174,
      "epoch": 0.3611111111111111,
      "grad_norm": 5.21875,
      "learning_rate": 9.541196472469234e-06,
      "loss": 1.62513771,
      "memory(GiB)": 97.17,
      "step": 14235,
      "train_speed(iter/s)": 1.628191
    },
    {
      "acc": 0.65190058,
      "epoch": 0.36123795027904615,
      "grad_norm": 6.6875,
      "learning_rate": 9.540757574793032e-06,
      "loss": 1.6331768,
      "memory(GiB)": 97.17,
      "step": 14240,
      "train_speed(iter/s)": 1.628257
    },
    {
      "acc": 0.64696169,
      "epoch": 0.36136478944698125,
      "grad_norm": 6.03125,
      "learning_rate": 9.540318477394039e-06,
      "loss": 1.64407692,
      "memory(GiB)": 97.17,
      "step": 14245,
      "train_speed(iter/s)": 1.628328
    },
    {
      "acc": 0.66525965,
      "epoch": 0.3614916286149163,
      "grad_norm": 4.8125,
      "learning_rate": 9.539879180291568e-06,
      "loss": 1.58592291,
      "memory(GiB)": 97.17,
      "step": 14250,
      "train_speed(iter/s)": 1.628393
    },
    {
      "acc": 0.6540174,
      "epoch": 0.36161846778285134,
      "grad_norm": 5.3125,
      "learning_rate": 9.539439683504943e-06,
      "loss": 1.61532097,
      "memory(GiB)": 97.17,
      "step": 14255,
      "train_speed(iter/s)": 1.62846
    },
    {
      "acc": 0.64410844,
      "epoch": 0.3617453069507864,
      "grad_norm": 5.46875,
      "learning_rate": 9.538999987053492e-06,
      "loss": 1.70699844,
      "memory(GiB)": 97.17,
      "step": 14260,
      "train_speed(iter/s)": 1.628529
    },
    {
      "acc": 0.64135828,
      "epoch": 0.3618721461187215,
      "grad_norm": 5.03125,
      "learning_rate": 9.538560090956557e-06,
      "loss": 1.67186012,
      "memory(GiB)": 97.17,
      "step": 14265,
      "train_speed(iter/s)": 1.628594
    },
    {
      "acc": 0.6543262,
      "epoch": 0.3619989852866565,
      "grad_norm": 5.78125,
      "learning_rate": 9.538119995233485e-06,
      "loss": 1.61801453,
      "memory(GiB)": 97.17,
      "step": 14270,
      "train_speed(iter/s)": 1.628661
    },
    {
      "acc": 0.6599596,
      "epoch": 0.36212582445459157,
      "grad_norm": 5.03125,
      "learning_rate": 9.537679699903637e-06,
      "loss": 1.64361725,
      "memory(GiB)": 97.17,
      "step": 14275,
      "train_speed(iter/s)": 1.628721
    },
    {
      "acc": 0.67413931,
      "epoch": 0.3622526636225266,
      "grad_norm": 7.0,
      "learning_rate": 9.537239204986375e-06,
      "loss": 1.48503141,
      "memory(GiB)": 97.17,
      "step": 14280,
      "train_speed(iter/s)": 1.628788
    },
    {
      "acc": 0.63579888,
      "epoch": 0.3623795027904617,
      "grad_norm": 5.03125,
      "learning_rate": 9.536798510501075e-06,
      "loss": 1.65280933,
      "memory(GiB)": 97.17,
      "step": 14285,
      "train_speed(iter/s)": 1.628854
    },
    {
      "acc": 0.65138206,
      "epoch": 0.36250634195839676,
      "grad_norm": 9.625,
      "learning_rate": 9.536357616467123e-06,
      "loss": 1.60710125,
      "memory(GiB)": 97.17,
      "step": 14290,
      "train_speed(iter/s)": 1.628924
    },
    {
      "acc": 0.65690031,
      "epoch": 0.3626331811263318,
      "grad_norm": 5.03125,
      "learning_rate": 9.535916522903908e-06,
      "loss": 1.59372406,
      "memory(GiB)": 97.17,
      "step": 14295,
      "train_speed(iter/s)": 1.62899
    },
    {
      "acc": 0.64046726,
      "epoch": 0.36276002029426685,
      "grad_norm": 6.40625,
      "learning_rate": 9.535475229830832e-06,
      "loss": 1.69992867,
      "memory(GiB)": 97.17,
      "step": 14300,
      "train_speed(iter/s)": 1.629054
    },
    {
      "acc": 0.64613805,
      "epoch": 0.36288685946220195,
      "grad_norm": 5.09375,
      "learning_rate": 9.535033737267308e-06,
      "loss": 1.59447412,
      "memory(GiB)": 97.17,
      "step": 14305,
      "train_speed(iter/s)": 1.629118
    },
    {
      "acc": 0.64742351,
      "epoch": 0.363013698630137,
      "grad_norm": 5.9375,
      "learning_rate": 9.534592045232752e-06,
      "loss": 1.56169319,
      "memory(GiB)": 97.17,
      "step": 14310,
      "train_speed(iter/s)": 1.629182
    },
    {
      "acc": 0.64363832,
      "epoch": 0.36314053779807204,
      "grad_norm": 5.6875,
      "learning_rate": 9.534150153746591e-06,
      "loss": 1.70888824,
      "memory(GiB)": 97.17,
      "step": 14315,
      "train_speed(iter/s)": 1.629244
    },
    {
      "acc": 0.64335051,
      "epoch": 0.3632673769660071,
      "grad_norm": 6.4375,
      "learning_rate": 9.533708062828264e-06,
      "loss": 1.62780361,
      "memory(GiB)": 97.17,
      "step": 14320,
      "train_speed(iter/s)": 1.629314
    },
    {
      "acc": 0.65207286,
      "epoch": 0.3633942161339422,
      "grad_norm": 5.40625,
      "learning_rate": 9.533265772497216e-06,
      "loss": 1.67987556,
      "memory(GiB)": 97.17,
      "step": 14325,
      "train_speed(iter/s)": 1.629379
    },
    {
      "acc": 0.65923223,
      "epoch": 0.3635210553018772,
      "grad_norm": 5.875,
      "learning_rate": 9.532823282772899e-06,
      "loss": 1.58446646,
      "memory(GiB)": 97.17,
      "step": 14330,
      "train_speed(iter/s)": 1.629448
    },
    {
      "acc": 0.64248061,
      "epoch": 0.36364789446981227,
      "grad_norm": 6.15625,
      "learning_rate": 9.532380593674775e-06,
      "loss": 1.61728401,
      "memory(GiB)": 97.17,
      "step": 14335,
      "train_speed(iter/s)": 1.629513
    },
    {
      "acc": 0.64716191,
      "epoch": 0.3637747336377473,
      "grad_norm": 6.9375,
      "learning_rate": 9.531937705222319e-06,
      "loss": 1.67005825,
      "memory(GiB)": 97.17,
      "step": 14340,
      "train_speed(iter/s)": 1.629582
    },
    {
      "acc": 0.6364892,
      "epoch": 0.3639015728056824,
      "grad_norm": 5.625,
      "learning_rate": 9.531494617435006e-06,
      "loss": 1.63972435,
      "memory(GiB)": 97.17,
      "step": 14345,
      "train_speed(iter/s)": 1.629645
    },
    {
      "acc": 0.65665822,
      "epoch": 0.36402841197361746,
      "grad_norm": 7.4375,
      "learning_rate": 9.531051330332331e-06,
      "loss": 1.5889123,
      "memory(GiB)": 97.17,
      "step": 14350,
      "train_speed(iter/s)": 1.629712
    },
    {
      "acc": 0.63490801,
      "epoch": 0.3641552511415525,
      "grad_norm": 5.65625,
      "learning_rate": 9.530607843933788e-06,
      "loss": 1.67954674,
      "memory(GiB)": 97.17,
      "step": 14355,
      "train_speed(iter/s)": 1.629778
    },
    {
      "acc": 0.65125618,
      "epoch": 0.36428209030948755,
      "grad_norm": 5.46875,
      "learning_rate": 9.530164158258883e-06,
      "loss": 1.57851257,
      "memory(GiB)": 97.17,
      "step": 14360,
      "train_speed(iter/s)": 1.629843
    },
    {
      "acc": 0.66065435,
      "epoch": 0.36440892947742265,
      "grad_norm": 5.59375,
      "learning_rate": 9.529720273327135e-06,
      "loss": 1.5982132,
      "memory(GiB)": 97.17,
      "step": 14365,
      "train_speed(iter/s)": 1.629907
    },
    {
      "acc": 0.65221481,
      "epoch": 0.3645357686453577,
      "grad_norm": 6.46875,
      "learning_rate": 9.529276189158063e-06,
      "loss": 1.67781277,
      "memory(GiB)": 97.17,
      "step": 14370,
      "train_speed(iter/s)": 1.629972
    },
    {
      "acc": 0.65666084,
      "epoch": 0.36466260781329274,
      "grad_norm": 5.09375,
      "learning_rate": 9.528831905771205e-06,
      "loss": 1.63739128,
      "memory(GiB)": 97.17,
      "step": 14375,
      "train_speed(iter/s)": 1.630032
    },
    {
      "acc": 0.64627666,
      "epoch": 0.3647894469812278,
      "grad_norm": 6.96875,
      "learning_rate": 9.528387423186098e-06,
      "loss": 1.66223564,
      "memory(GiB)": 97.17,
      "step": 14380,
      "train_speed(iter/s)": 1.630098
    },
    {
      "acc": 0.65122247,
      "epoch": 0.3649162861491629,
      "grad_norm": 5.46875,
      "learning_rate": 9.527942741422297e-06,
      "loss": 1.60981102,
      "memory(GiB)": 97.17,
      "step": 14385,
      "train_speed(iter/s)": 1.630165
    },
    {
      "acc": 0.64303856,
      "epoch": 0.3650431253170979,
      "grad_norm": 5.71875,
      "learning_rate": 9.527497860499355e-06,
      "loss": 1.70701275,
      "memory(GiB)": 97.17,
      "step": 14390,
      "train_speed(iter/s)": 1.63023
    },
    {
      "acc": 0.65450912,
      "epoch": 0.36516996448503297,
      "grad_norm": 5.9375,
      "learning_rate": 9.527052780436845e-06,
      "loss": 1.54058809,
      "memory(GiB)": 97.17,
      "step": 14395,
      "train_speed(iter/s)": 1.630296
    },
    {
      "acc": 0.64673252,
      "epoch": 0.365296803652968,
      "grad_norm": 5.9375,
      "learning_rate": 9.52660750125434e-06,
      "loss": 1.62171021,
      "memory(GiB)": 97.17,
      "step": 14400,
      "train_speed(iter/s)": 1.630358
    },
    {
      "acc": 0.64762077,
      "epoch": 0.3654236428209031,
      "grad_norm": 5.0625,
      "learning_rate": 9.52616202297143e-06,
      "loss": 1.64411507,
      "memory(GiB)": 97.17,
      "step": 14405,
      "train_speed(iter/s)": 1.630421
    },
    {
      "acc": 0.64639888,
      "epoch": 0.36555048198883816,
      "grad_norm": 4.65625,
      "learning_rate": 9.525716345607706e-06,
      "loss": 1.62094383,
      "memory(GiB)": 97.17,
      "step": 14410,
      "train_speed(iter/s)": 1.630484
    },
    {
      "acc": 0.62937675,
      "epoch": 0.3656773211567732,
      "grad_norm": 4.78125,
      "learning_rate": 9.52527046918277e-06,
      "loss": 1.70464325,
      "memory(GiB)": 97.17,
      "step": 14415,
      "train_speed(iter/s)": 1.630548
    },
    {
      "acc": 0.63535228,
      "epoch": 0.36580416032470825,
      "grad_norm": 4.8125,
      "learning_rate": 9.524824393716235e-06,
      "loss": 1.70969715,
      "memory(GiB)": 97.17,
      "step": 14420,
      "train_speed(iter/s)": 1.630611
    },
    {
      "acc": 0.646453,
      "epoch": 0.36593099949264335,
      "grad_norm": 6.125,
      "learning_rate": 9.524378119227722e-06,
      "loss": 1.62970886,
      "memory(GiB)": 97.17,
      "step": 14425,
      "train_speed(iter/s)": 1.630675
    },
    {
      "acc": 0.65223055,
      "epoch": 0.3660578386605784,
      "grad_norm": 5.65625,
      "learning_rate": 9.523931645736858e-06,
      "loss": 1.63869209,
      "memory(GiB)": 97.17,
      "step": 14430,
      "train_speed(iter/s)": 1.63074
    },
    {
      "acc": 0.65391784,
      "epoch": 0.36618467782851344,
      "grad_norm": 5.75,
      "learning_rate": 9.523484973263283e-06,
      "loss": 1.65556507,
      "memory(GiB)": 97.17,
      "step": 14435,
      "train_speed(iter/s)": 1.630804
    },
    {
      "acc": 0.66809335,
      "epoch": 0.3663115169964485,
      "grad_norm": 6.40625,
      "learning_rate": 9.523038101826644e-06,
      "loss": 1.58158026,
      "memory(GiB)": 97.17,
      "step": 14440,
      "train_speed(iter/s)": 1.630871
    },
    {
      "acc": 0.62958465,
      "epoch": 0.3664383561643836,
      "grad_norm": 5.78125,
      "learning_rate": 9.522591031446596e-06,
      "loss": 1.71249962,
      "memory(GiB)": 97.17,
      "step": 14445,
      "train_speed(iter/s)": 1.630934
    },
    {
      "acc": 0.65568352,
      "epoch": 0.3665651953323186,
      "grad_norm": 5.84375,
      "learning_rate": 9.522143762142801e-06,
      "loss": 1.59135056,
      "memory(GiB)": 97.17,
      "step": 14450,
      "train_speed(iter/s)": 1.630998
    },
    {
      "acc": 0.66998911,
      "epoch": 0.36669203450025367,
      "grad_norm": 5.71875,
      "learning_rate": 9.521696293934934e-06,
      "loss": 1.59900875,
      "memory(GiB)": 97.17,
      "step": 14455,
      "train_speed(iter/s)": 1.631058
    },
    {
      "acc": 0.65012841,
      "epoch": 0.3668188736681887,
      "grad_norm": 5.21875,
      "learning_rate": 9.521248626842676e-06,
      "loss": 1.66891575,
      "memory(GiB)": 97.17,
      "step": 14460,
      "train_speed(iter/s)": 1.631121
    },
    {
      "acc": 0.63980789,
      "epoch": 0.3669457128361238,
      "grad_norm": 4.875,
      "learning_rate": 9.520800760885716e-06,
      "loss": 1.69302425,
      "memory(GiB)": 97.17,
      "step": 14465,
      "train_speed(iter/s)": 1.631182
    },
    {
      "acc": 0.65004869,
      "epoch": 0.36707255200405886,
      "grad_norm": 5.34375,
      "learning_rate": 9.520352696083756e-06,
      "loss": 1.58977261,
      "memory(GiB)": 97.17,
      "step": 14470,
      "train_speed(iter/s)": 1.631241
    },
    {
      "acc": 0.65741978,
      "epoch": 0.3671993911719939,
      "grad_norm": 5.875,
      "learning_rate": 9.519904432456504e-06,
      "loss": 1.5539629,
      "memory(GiB)": 97.17,
      "step": 14475,
      "train_speed(iter/s)": 1.631306
    },
    {
      "acc": 0.67306805,
      "epoch": 0.36732623033992895,
      "grad_norm": 5.96875,
      "learning_rate": 9.519455970023672e-06,
      "loss": 1.52765646,
      "memory(GiB)": 97.17,
      "step": 14480,
      "train_speed(iter/s)": 1.631369
    },
    {
      "acc": 0.64803562,
      "epoch": 0.36745306950786405,
      "grad_norm": 5.90625,
      "learning_rate": 9.519007308804991e-06,
      "loss": 1.63630543,
      "memory(GiB)": 97.17,
      "step": 14485,
      "train_speed(iter/s)": 1.631431
    },
    {
      "acc": 0.65242558,
      "epoch": 0.3675799086757991,
      "grad_norm": 4.78125,
      "learning_rate": 9.518558448820193e-06,
      "loss": 1.5788681,
      "memory(GiB)": 97.17,
      "step": 14490,
      "train_speed(iter/s)": 1.63149
    },
    {
      "acc": 0.64776564,
      "epoch": 0.36770674784373414,
      "grad_norm": 6.625,
      "learning_rate": 9.518109390089017e-06,
      "loss": 1.64999542,
      "memory(GiB)": 97.17,
      "step": 14495,
      "train_speed(iter/s)": 1.631549
    },
    {
      "acc": 0.64250813,
      "epoch": 0.3678335870116692,
      "grad_norm": 7.0,
      "learning_rate": 9.517660132631222e-06,
      "loss": 1.68008728,
      "memory(GiB)": 97.17,
      "step": 14500,
      "train_speed(iter/s)": 1.63161
    },
    {
      "acc": 0.66387734,
      "epoch": 0.3679604261796043,
      "grad_norm": 5.53125,
      "learning_rate": 9.517210676466561e-06,
      "loss": 1.63639526,
      "memory(GiB)": 97.17,
      "step": 14505,
      "train_speed(iter/s)": 1.631675
    },
    {
      "acc": 0.65841141,
      "epoch": 0.3680872653475393,
      "grad_norm": 4.34375,
      "learning_rate": 9.516761021614809e-06,
      "loss": 1.60073128,
      "memory(GiB)": 97.17,
      "step": 14510,
      "train_speed(iter/s)": 1.631736
    },
    {
      "acc": 0.65091815,
      "epoch": 0.36821410451547437,
      "grad_norm": 4.96875,
      "learning_rate": 9.51631116809574e-06,
      "loss": 1.62698765,
      "memory(GiB)": 97.17,
      "step": 14515,
      "train_speed(iter/s)": 1.631797
    },
    {
      "acc": 0.64248943,
      "epoch": 0.3683409436834094,
      "grad_norm": 5.46875,
      "learning_rate": 9.515861115929144e-06,
      "loss": 1.69752483,
      "memory(GiB)": 97.17,
      "step": 14520,
      "train_speed(iter/s)": 1.63186
    },
    {
      "acc": 0.64379282,
      "epoch": 0.3684677828513445,
      "grad_norm": 8.3125,
      "learning_rate": 9.515410865134812e-06,
      "loss": 1.62120342,
      "memory(GiB)": 97.17,
      "step": 14525,
      "train_speed(iter/s)": 1.631923
    },
    {
      "acc": 0.66134501,
      "epoch": 0.36859462201927956,
      "grad_norm": 5.59375,
      "learning_rate": 9.514960415732551e-06,
      "loss": 1.56136761,
      "memory(GiB)": 97.17,
      "step": 14530,
      "train_speed(iter/s)": 1.631988
    },
    {
      "acc": 0.64463539,
      "epoch": 0.3687214611872146,
      "grad_norm": 6.1875,
      "learning_rate": 9.514509767742172e-06,
      "loss": 1.68267403,
      "memory(GiB)": 97.17,
      "step": 14535,
      "train_speed(iter/s)": 1.632052
    },
    {
      "acc": 0.65353851,
      "epoch": 0.36884830035514965,
      "grad_norm": 8.0625,
      "learning_rate": 9.5140589211835e-06,
      "loss": 1.63590584,
      "memory(GiB)": 97.17,
      "step": 14540,
      "train_speed(iter/s)": 1.632115
    },
    {
      "acc": 0.65548134,
      "epoch": 0.36897513952308475,
      "grad_norm": 6.59375,
      "learning_rate": 9.513607876076363e-06,
      "loss": 1.6759716,
      "memory(GiB)": 97.17,
      "step": 14545,
      "train_speed(iter/s)": 1.632179
    },
    {
      "acc": 0.65960293,
      "epoch": 0.3691019786910198,
      "grad_norm": 6.90625,
      "learning_rate": 9.513156632440598e-06,
      "loss": 1.58911648,
      "memory(GiB)": 97.17,
      "step": 14550,
      "train_speed(iter/s)": 1.632244
    },
    {
      "acc": 0.65421047,
      "epoch": 0.36922881785895484,
      "grad_norm": 5.09375,
      "learning_rate": 9.512705190296055e-06,
      "loss": 1.63908081,
      "memory(GiB)": 97.17,
      "step": 14555,
      "train_speed(iter/s)": 1.632302
    },
    {
      "acc": 0.65789351,
      "epoch": 0.3693556570268899,
      "grad_norm": 5.4375,
      "learning_rate": 9.512253549662588e-06,
      "loss": 1.58075638,
      "memory(GiB)": 97.17,
      "step": 14560,
      "train_speed(iter/s)": 1.632364
    },
    {
      "acc": 0.6556778,
      "epoch": 0.369482496194825,
      "grad_norm": 5.71875,
      "learning_rate": 9.511801710560066e-06,
      "loss": 1.68199329,
      "memory(GiB)": 97.17,
      "step": 14565,
      "train_speed(iter/s)": 1.632427
    },
    {
      "acc": 0.65963564,
      "epoch": 0.36960933536276,
      "grad_norm": 6.625,
      "learning_rate": 9.511349673008364e-06,
      "loss": 1.584758,
      "memory(GiB)": 97.17,
      "step": 14570,
      "train_speed(iter/s)": 1.632488
    },
    {
      "acc": 0.6469985,
      "epoch": 0.36973617453069507,
      "grad_norm": 8.8125,
      "learning_rate": 9.510897437027358e-06,
      "loss": 1.64005203,
      "memory(GiB)": 97.17,
      "step": 14575,
      "train_speed(iter/s)": 1.632548
    },
    {
      "acc": 0.6561111,
      "epoch": 0.3698630136986301,
      "grad_norm": 6.0,
      "learning_rate": 9.510445002636943e-06,
      "loss": 1.62161083,
      "memory(GiB)": 97.17,
      "step": 14580,
      "train_speed(iter/s)": 1.632609
    },
    {
      "acc": 0.67015715,
      "epoch": 0.3699898528665652,
      "grad_norm": 5.8125,
      "learning_rate": 9.50999236985702e-06,
      "loss": 1.60314808,
      "memory(GiB)": 97.17,
      "step": 14585,
      "train_speed(iter/s)": 1.632669
    },
    {
      "acc": 0.65853643,
      "epoch": 0.37011669203450026,
      "grad_norm": 6.28125,
      "learning_rate": 9.509539538707497e-06,
      "loss": 1.68197575,
      "memory(GiB)": 97.17,
      "step": 14590,
      "train_speed(iter/s)": 1.632729
    },
    {
      "acc": 0.64739695,
      "epoch": 0.3702435312024353,
      "grad_norm": 7.3125,
      "learning_rate": 9.50908650920829e-06,
      "loss": 1.64297028,
      "memory(GiB)": 97.17,
      "step": 14595,
      "train_speed(iter/s)": 1.632788
    },
    {
      "acc": 0.64998484,
      "epoch": 0.37037037037037035,
      "grad_norm": 4.84375,
      "learning_rate": 9.50863328137933e-06,
      "loss": 1.60108604,
      "memory(GiB)": 97.17,
      "step": 14600,
      "train_speed(iter/s)": 1.632848
    },
    {
      "acc": 0.65740128,
      "epoch": 0.37049720953830545,
      "grad_norm": 5.5,
      "learning_rate": 9.508179855240545e-06,
      "loss": 1.59114885,
      "memory(GiB)": 97.17,
      "step": 14605,
      "train_speed(iter/s)": 1.632908
    },
    {
      "acc": 0.65801759,
      "epoch": 0.3706240487062405,
      "grad_norm": 5.375,
      "learning_rate": 9.507726230811884e-06,
      "loss": 1.68662777,
      "memory(GiB)": 97.17,
      "step": 14610,
      "train_speed(iter/s)": 1.632971
    },
    {
      "acc": 0.64627829,
      "epoch": 0.37075088787417554,
      "grad_norm": 5.78125,
      "learning_rate": 9.507272408113298e-06,
      "loss": 1.60258789,
      "memory(GiB)": 97.17,
      "step": 14615,
      "train_speed(iter/s)": 1.633027
    },
    {
      "acc": 0.63934393,
      "epoch": 0.3708777270421106,
      "grad_norm": 4.59375,
      "learning_rate": 9.506818387164748e-06,
      "loss": 1.67579861,
      "memory(GiB)": 97.17,
      "step": 14620,
      "train_speed(iter/s)": 1.633087
    },
    {
      "acc": 0.65653715,
      "epoch": 0.3710045662100457,
      "grad_norm": 5.71875,
      "learning_rate": 9.506364167986204e-06,
      "loss": 1.65439224,
      "memory(GiB)": 97.17,
      "step": 14625,
      "train_speed(iter/s)": 1.633145
    },
    {
      "acc": 0.65933456,
      "epoch": 0.3711314053779807,
      "grad_norm": 6.1875,
      "learning_rate": 9.505909750597644e-06,
      "loss": 1.65304337,
      "memory(GiB)": 97.17,
      "step": 14630,
      "train_speed(iter/s)": 1.63321
    },
    {
      "acc": 0.64524608,
      "epoch": 0.37125824454591577,
      "grad_norm": 5.90625,
      "learning_rate": 9.505455135019055e-06,
      "loss": 1.60034714,
      "memory(GiB)": 97.17,
      "step": 14635,
      "train_speed(iter/s)": 1.633273
    },
    {
      "acc": 0.6503952,
      "epoch": 0.3713850837138508,
      "grad_norm": 8.6875,
      "learning_rate": 9.505000321270435e-06,
      "loss": 1.59648886,
      "memory(GiB)": 97.17,
      "step": 14640,
      "train_speed(iter/s)": 1.633336
    },
    {
      "acc": 0.6597218,
      "epoch": 0.3715119228817859,
      "grad_norm": 6.28125,
      "learning_rate": 9.504545309371786e-06,
      "loss": 1.62139435,
      "memory(GiB)": 97.17,
      "step": 14645,
      "train_speed(iter/s)": 1.633397
    },
    {
      "acc": 0.63369408,
      "epoch": 0.37163876204972096,
      "grad_norm": 4.96875,
      "learning_rate": 9.504090099343125e-06,
      "loss": 1.68813248,
      "memory(GiB)": 97.17,
      "step": 14650,
      "train_speed(iter/s)": 1.633462
    },
    {
      "acc": 0.67516389,
      "epoch": 0.371765601217656,
      "grad_norm": 5.96875,
      "learning_rate": 9.50363469120447e-06,
      "loss": 1.54997845,
      "memory(GiB)": 97.17,
      "step": 14655,
      "train_speed(iter/s)": 1.633521
    },
    {
      "acc": 0.62684927,
      "epoch": 0.37189244038559105,
      "grad_norm": 5.59375,
      "learning_rate": 9.503179084975855e-06,
      "loss": 1.68973427,
      "memory(GiB)": 97.17,
      "step": 14660,
      "train_speed(iter/s)": 1.633586
    },
    {
      "acc": 0.64531069,
      "epoch": 0.37201927955352615,
      "grad_norm": 5.25,
      "learning_rate": 9.502723280677319e-06,
      "loss": 1.67327309,
      "memory(GiB)": 97.17,
      "step": 14665,
      "train_speed(iter/s)": 1.633646
    },
    {
      "acc": 0.66482682,
      "epoch": 0.3721461187214612,
      "grad_norm": 6.625,
      "learning_rate": 9.50226727832891e-06,
      "loss": 1.67276001,
      "memory(GiB)": 97.17,
      "step": 14670,
      "train_speed(iter/s)": 1.633707
    },
    {
      "acc": 0.63452954,
      "epoch": 0.37227295788939624,
      "grad_norm": 6.375,
      "learning_rate": 9.501811077950685e-06,
      "loss": 1.68142776,
      "memory(GiB)": 97.17,
      "step": 14675,
      "train_speed(iter/s)": 1.633768
    },
    {
      "acc": 0.65397015,
      "epoch": 0.3723997970573313,
      "grad_norm": 5.78125,
      "learning_rate": 9.501354679562708e-06,
      "loss": 1.62402115,
      "memory(GiB)": 97.17,
      "step": 14680,
      "train_speed(iter/s)": 1.63383
    },
    {
      "acc": 0.65664773,
      "epoch": 0.3725266362252664,
      "grad_norm": 5.75,
      "learning_rate": 9.500898083185058e-06,
      "loss": 1.6172451,
      "memory(GiB)": 97.17,
      "step": 14685,
      "train_speed(iter/s)": 1.63389
    },
    {
      "acc": 0.65130043,
      "epoch": 0.3726534753932014,
      "grad_norm": 5.15625,
      "learning_rate": 9.500441288837812e-06,
      "loss": 1.59556112,
      "memory(GiB)": 97.17,
      "step": 14690,
      "train_speed(iter/s)": 1.633947
    },
    {
      "acc": 0.64774413,
      "epoch": 0.37278031456113647,
      "grad_norm": 6.125,
      "learning_rate": 9.499984296541066e-06,
      "loss": 1.67401886,
      "memory(GiB)": 97.17,
      "step": 14695,
      "train_speed(iter/s)": 1.634009
    },
    {
      "acc": 0.63403406,
      "epoch": 0.3729071537290715,
      "grad_norm": 6.25,
      "learning_rate": 9.49952710631492e-06,
      "loss": 1.76184998,
      "memory(GiB)": 97.17,
      "step": 14700,
      "train_speed(iter/s)": 1.63407
    },
    {
      "acc": 0.64683886,
      "epoch": 0.3730339928970066,
      "grad_norm": 6.0,
      "learning_rate": 9.499069718179484e-06,
      "loss": 1.62151756,
      "memory(GiB)": 97.17,
      "step": 14705,
      "train_speed(iter/s)": 1.634131
    },
    {
      "acc": 0.66225176,
      "epoch": 0.37316083206494166,
      "grad_norm": 5.96875,
      "learning_rate": 9.498612132154874e-06,
      "loss": 1.56371212,
      "memory(GiB)": 97.17,
      "step": 14710,
      "train_speed(iter/s)": 1.634192
    },
    {
      "acc": 0.65066142,
      "epoch": 0.3732876712328767,
      "grad_norm": 4.15625,
      "learning_rate": 9.498154348261217e-06,
      "loss": 1.60862694,
      "memory(GiB)": 97.17,
      "step": 14715,
      "train_speed(iter/s)": 1.634253
    },
    {
      "acc": 0.65171728,
      "epoch": 0.37341451040081175,
      "grad_norm": 5.71875,
      "learning_rate": 9.497696366518649e-06,
      "loss": 1.63322906,
      "memory(GiB)": 97.17,
      "step": 14720,
      "train_speed(iter/s)": 1.634314
    },
    {
      "acc": 0.64669833,
      "epoch": 0.37354134956874685,
      "grad_norm": 4.90625,
      "learning_rate": 9.497238186947315e-06,
      "loss": 1.66693687,
      "memory(GiB)": 97.17,
      "step": 14725,
      "train_speed(iter/s)": 1.634375
    },
    {
      "acc": 0.63985062,
      "epoch": 0.3736681887366819,
      "grad_norm": 5.6875,
      "learning_rate": 9.496779809567367e-06,
      "loss": 1.69202881,
      "memory(GiB)": 97.17,
      "step": 14730,
      "train_speed(iter/s)": 1.634436
    },
    {
      "acc": 0.6521153,
      "epoch": 0.37379502790461694,
      "grad_norm": 4.625,
      "learning_rate": 9.496321234398967e-06,
      "loss": 1.56692915,
      "memory(GiB)": 97.17,
      "step": 14735,
      "train_speed(iter/s)": 1.634496
    },
    {
      "acc": 0.65274858,
      "epoch": 0.373921867072552,
      "grad_norm": 4.5625,
      "learning_rate": 9.495862461462282e-06,
      "loss": 1.62136841,
      "memory(GiB)": 97.17,
      "step": 14740,
      "train_speed(iter/s)": 1.634554
    },
    {
      "acc": 0.65033445,
      "epoch": 0.3740487062404871,
      "grad_norm": 5.03125,
      "learning_rate": 9.495403490777495e-06,
      "loss": 1.6391552,
      "memory(GiB)": 97.17,
      "step": 14745,
      "train_speed(iter/s)": 1.634608
    },
    {
      "acc": 0.67271242,
      "epoch": 0.3741755454084221,
      "grad_norm": 5.46875,
      "learning_rate": 9.49494432236479e-06,
      "loss": 1.56324282,
      "memory(GiB)": 97.17,
      "step": 14750,
      "train_speed(iter/s)": 1.634667
    },
    {
      "acc": 0.65324721,
      "epoch": 0.37430238457635717,
      "grad_norm": 5.25,
      "learning_rate": 9.494484956244368e-06,
      "loss": 1.63767223,
      "memory(GiB)": 97.17,
      "step": 14755,
      "train_speed(iter/s)": 1.634727
    },
    {
      "acc": 0.64981432,
      "epoch": 0.3744292237442922,
      "grad_norm": 6.5625,
      "learning_rate": 9.49402539243643e-06,
      "loss": 1.65517578,
      "memory(GiB)": 97.17,
      "step": 14760,
      "train_speed(iter/s)": 1.634791
    },
    {
      "acc": 0.64459352,
      "epoch": 0.3745560629122273,
      "grad_norm": 5.3125,
      "learning_rate": 9.49356563096119e-06,
      "loss": 1.64013443,
      "memory(GiB)": 97.17,
      "step": 14765,
      "train_speed(iter/s)": 1.634848
    },
    {
      "acc": 0.65208397,
      "epoch": 0.37468290208016236,
      "grad_norm": 5.09375,
      "learning_rate": 9.49310567183887e-06,
      "loss": 1.60190315,
      "memory(GiB)": 97.17,
      "step": 14770,
      "train_speed(iter/s)": 1.63491
    },
    {
      "acc": 0.65274296,
      "epoch": 0.3748097412480974,
      "grad_norm": 6.0625,
      "learning_rate": 9.492645515089706e-06,
      "loss": 1.62087288,
      "memory(GiB)": 97.17,
      "step": 14775,
      "train_speed(iter/s)": 1.634972
    },
    {
      "acc": 0.65257802,
      "epoch": 0.37493658041603245,
      "grad_norm": 6.375,
      "learning_rate": 9.492185160733934e-06,
      "loss": 1.62115135,
      "memory(GiB)": 97.17,
      "step": 14780,
      "train_speed(iter/s)": 1.635033
    },
    {
      "acc": 0.64833503,
      "epoch": 0.37506341958396755,
      "grad_norm": 6.65625,
      "learning_rate": 9.491724608791798e-06,
      "loss": 1.65103951,
      "memory(GiB)": 97.17,
      "step": 14785,
      "train_speed(iter/s)": 1.635093
    },
    {
      "acc": 0.64485178,
      "epoch": 0.3751902587519026,
      "grad_norm": 5.5625,
      "learning_rate": 9.491263859283563e-06,
      "loss": 1.58244686,
      "memory(GiB)": 97.17,
      "step": 14790,
      "train_speed(iter/s)": 1.635151
    },
    {
      "acc": 0.65921292,
      "epoch": 0.37531709791983764,
      "grad_norm": 4.375,
      "learning_rate": 9.490802912229491e-06,
      "loss": 1.61254768,
      "memory(GiB)": 97.17,
      "step": 14795,
      "train_speed(iter/s)": 1.635211
    },
    {
      "acc": 0.63699598,
      "epoch": 0.3754439370877727,
      "grad_norm": 4.59375,
      "learning_rate": 9.490341767649858e-06,
      "loss": 1.69989891,
      "memory(GiB)": 97.17,
      "step": 14800,
      "train_speed(iter/s)": 1.635272
    },
    {
      "acc": 0.65551848,
      "epoch": 0.3755707762557078,
      "grad_norm": 6.0,
      "learning_rate": 9.489880425564944e-06,
      "loss": 1.57976818,
      "memory(GiB)": 97.17,
      "step": 14805,
      "train_speed(iter/s)": 1.635332
    },
    {
      "acc": 0.64182749,
      "epoch": 0.37569761542364283,
      "grad_norm": 4.71875,
      "learning_rate": 9.489418885995043e-06,
      "loss": 1.62107086,
      "memory(GiB)": 97.17,
      "step": 14810,
      "train_speed(iter/s)": 1.635393
    },
    {
      "acc": 0.63403397,
      "epoch": 0.3758244545915779,
      "grad_norm": 5.375,
      "learning_rate": 9.488957148960457e-06,
      "loss": 1.72306919,
      "memory(GiB)": 97.17,
      "step": 14815,
      "train_speed(iter/s)": 1.635455
    },
    {
      "acc": 0.63592906,
      "epoch": 0.3759512937595129,
      "grad_norm": 5.28125,
      "learning_rate": 9.488495214481494e-06,
      "loss": 1.61382446,
      "memory(GiB)": 97.17,
      "step": 14820,
      "train_speed(iter/s)": 1.635516
    },
    {
      "acc": 0.65910792,
      "epoch": 0.376078132927448,
      "grad_norm": 5.28125,
      "learning_rate": 9.48803308257847e-06,
      "loss": 1.62439651,
      "memory(GiB)": 97.17,
      "step": 14825,
      "train_speed(iter/s)": 1.635574
    },
    {
      "acc": 0.66102428,
      "epoch": 0.37620497209538306,
      "grad_norm": 5.6875,
      "learning_rate": 9.487570753271716e-06,
      "loss": 1.59010477,
      "memory(GiB)": 97.17,
      "step": 14830,
      "train_speed(iter/s)": 1.635637
    },
    {
      "acc": 0.64529419,
      "epoch": 0.3763318112633181,
      "grad_norm": 6.71875,
      "learning_rate": 9.487108226581564e-06,
      "loss": 1.65264645,
      "memory(GiB)": 97.17,
      "step": 14835,
      "train_speed(iter/s)": 1.6357
    },
    {
      "acc": 0.65747051,
      "epoch": 0.37645865043125315,
      "grad_norm": 4.84375,
      "learning_rate": 9.486645502528355e-06,
      "loss": 1.62201042,
      "memory(GiB)": 97.17,
      "step": 14840,
      "train_speed(iter/s)": 1.635758
    },
    {
      "acc": 0.65689774,
      "epoch": 0.37658548959918825,
      "grad_norm": 4.84375,
      "learning_rate": 9.486182581132449e-06,
      "loss": 1.58362923,
      "memory(GiB)": 97.17,
      "step": 14845,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.651126,
      "epoch": 0.3767123287671233,
      "grad_norm": 6.4375,
      "learning_rate": 9.485719462414202e-06,
      "loss": 1.57980499,
      "memory(GiB)": 97.17,
      "step": 14850,
      "train_speed(iter/s)": 1.63588
    },
    {
      "acc": 0.6693182,
      "epoch": 0.37683916793505834,
      "grad_norm": 7.15625,
      "learning_rate": 9.485256146393987e-06,
      "loss": 1.66769447,
      "memory(GiB)": 97.17,
      "step": 14855,
      "train_speed(iter/s)": 1.635943
    },
    {
      "acc": 0.65831304,
      "epoch": 0.3769660071029934,
      "grad_norm": 6.40625,
      "learning_rate": 9.484792633092182e-06,
      "loss": 1.62320271,
      "memory(GiB)": 97.17,
      "step": 14860,
      "train_speed(iter/s)": 1.636
    },
    {
      "acc": 0.66238065,
      "epoch": 0.3770928462709285,
      "grad_norm": 5.84375,
      "learning_rate": 9.484328922529172e-06,
      "loss": 1.55749445,
      "memory(GiB)": 97.17,
      "step": 14865,
      "train_speed(iter/s)": 1.636061
    },
    {
      "acc": 0.64541454,
      "epoch": 0.37721968543886353,
      "grad_norm": 5.03125,
      "learning_rate": 9.483865014725356e-06,
      "loss": 1.54721661,
      "memory(GiB)": 97.17,
      "step": 14870,
      "train_speed(iter/s)": 1.63612
    },
    {
      "acc": 0.65263486,
      "epoch": 0.3773465246067986,
      "grad_norm": 7.46875,
      "learning_rate": 9.483400909701139e-06,
      "loss": 1.65839577,
      "memory(GiB)": 97.17,
      "step": 14875,
      "train_speed(iter/s)": 1.63618
    },
    {
      "acc": 0.64702287,
      "epoch": 0.3774733637747336,
      "grad_norm": 5.65625,
      "learning_rate": 9.482936607476931e-06,
      "loss": 1.64633999,
      "memory(GiB)": 97.17,
      "step": 14880,
      "train_speed(iter/s)": 1.63624
    },
    {
      "acc": 0.64630709,
      "epoch": 0.3776002029426687,
      "grad_norm": 6.4375,
      "learning_rate": 9.482472108073157e-06,
      "loss": 1.69578705,
      "memory(GiB)": 97.17,
      "step": 14885,
      "train_speed(iter/s)": 1.636303
    },
    {
      "acc": 0.64325972,
      "epoch": 0.37772704211060376,
      "grad_norm": 5.78125,
      "learning_rate": 9.482007411510245e-06,
      "loss": 1.62746124,
      "memory(GiB)": 97.17,
      "step": 14890,
      "train_speed(iter/s)": 1.63636
    },
    {
      "acc": 0.65429587,
      "epoch": 0.3778538812785388,
      "grad_norm": 5.375,
      "learning_rate": 9.48154251780864e-06,
      "loss": 1.57414808,
      "memory(GiB)": 97.17,
      "step": 14895,
      "train_speed(iter/s)": 1.636419
    },
    {
      "acc": 0.65453629,
      "epoch": 0.37798072044647385,
      "grad_norm": 5.75,
      "learning_rate": 9.481077426988782e-06,
      "loss": 1.64011116,
      "memory(GiB)": 97.17,
      "step": 14900,
      "train_speed(iter/s)": 1.63648
    },
    {
      "acc": 0.63851681,
      "epoch": 0.37810755961440895,
      "grad_norm": 5.6875,
      "learning_rate": 9.480612139071134e-06,
      "loss": 1.64953651,
      "memory(GiB)": 97.17,
      "step": 14905,
      "train_speed(iter/s)": 1.636538
    },
    {
      "acc": 0.64264445,
      "epoch": 0.378234398782344,
      "grad_norm": 6.03125,
      "learning_rate": 9.48014665407616e-06,
      "loss": 1.65910034,
      "memory(GiB)": 97.17,
      "step": 14910,
      "train_speed(iter/s)": 1.636597
    },
    {
      "acc": 0.6703825,
      "epoch": 0.37836123795027904,
      "grad_norm": 5.65625,
      "learning_rate": 9.479680972024334e-06,
      "loss": 1.59870377,
      "memory(GiB)": 97.17,
      "step": 14915,
      "train_speed(iter/s)": 1.636656
    },
    {
      "acc": 0.66856813,
      "epoch": 0.3784880771182141,
      "grad_norm": 6.125,
      "learning_rate": 9.47921509293614e-06,
      "loss": 1.57689972,
      "memory(GiB)": 97.17,
      "step": 14920,
      "train_speed(iter/s)": 1.636717
    },
    {
      "acc": 0.65607595,
      "epoch": 0.3786149162861492,
      "grad_norm": 5.5,
      "learning_rate": 9.478749016832066e-06,
      "loss": 1.67704678,
      "memory(GiB)": 97.17,
      "step": 14925,
      "train_speed(iter/s)": 1.636775
    },
    {
      "acc": 0.65648675,
      "epoch": 0.37874175545408423,
      "grad_norm": 4.625,
      "learning_rate": 9.478282743732613e-06,
      "loss": 1.57121677,
      "memory(GiB)": 97.17,
      "step": 14930,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.65071754,
      "epoch": 0.3788685946220193,
      "grad_norm": 5.78125,
      "learning_rate": 9.477816273658293e-06,
      "loss": 1.63837528,
      "memory(GiB)": 97.17,
      "step": 14935,
      "train_speed(iter/s)": 1.636884
    },
    {
      "acc": 0.64984107,
      "epoch": 0.3789954337899543,
      "grad_norm": 5.34375,
      "learning_rate": 9.47734960662962e-06,
      "loss": 1.67655354,
      "memory(GiB)": 97.17,
      "step": 14940,
      "train_speed(iter/s)": 1.63694
    },
    {
      "acc": 0.65793409,
      "epoch": 0.3791222729578894,
      "grad_norm": 8.125,
      "learning_rate": 9.476882742667122e-06,
      "loss": 1.60423203,
      "memory(GiB)": 97.17,
      "step": 14945,
      "train_speed(iter/s)": 1.636998
    },
    {
      "acc": 0.64778004,
      "epoch": 0.37924911212582446,
      "grad_norm": 5.5,
      "learning_rate": 9.476415681791333e-06,
      "loss": 1.59187326,
      "memory(GiB)": 97.17,
      "step": 14950,
      "train_speed(iter/s)": 1.637054
    },
    {
      "acc": 0.65125122,
      "epoch": 0.3793759512937595,
      "grad_norm": 5.78125,
      "learning_rate": 9.475948424022798e-06,
      "loss": 1.59747,
      "memory(GiB)": 97.17,
      "step": 14955,
      "train_speed(iter/s)": 1.637113
    },
    {
      "acc": 0.66960096,
      "epoch": 0.37950279046169455,
      "grad_norm": 5.15625,
      "learning_rate": 9.475480969382065e-06,
      "loss": 1.56869583,
      "memory(GiB)": 97.17,
      "step": 14960,
      "train_speed(iter/s)": 1.637171
    },
    {
      "acc": 0.6516613,
      "epoch": 0.37962962962962965,
      "grad_norm": 5.21875,
      "learning_rate": 9.475013317889699e-06,
      "loss": 1.60754051,
      "memory(GiB)": 97.17,
      "step": 14965,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.64712429,
      "epoch": 0.3797564687975647,
      "grad_norm": 5.5625,
      "learning_rate": 9.474545469566267e-06,
      "loss": 1.60855618,
      "memory(GiB)": 97.17,
      "step": 14970,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.64248848,
      "epoch": 0.37988330796549974,
      "grad_norm": 5.3125,
      "learning_rate": 9.474077424432348e-06,
      "loss": 1.63334675,
      "memory(GiB)": 97.17,
      "step": 14975,
      "train_speed(iter/s)": 1.637349
    },
    {
      "acc": 0.64489765,
      "epoch": 0.3800101471334348,
      "grad_norm": 5.875,
      "learning_rate": 9.47360918250853e-06,
      "loss": 1.70478287,
      "memory(GiB)": 97.17,
      "step": 14980,
      "train_speed(iter/s)": 1.637406
    },
    {
      "acc": 0.64993458,
      "epoch": 0.3801369863013699,
      "grad_norm": 6.90625,
      "learning_rate": 9.473140743815405e-06,
      "loss": 1.70229359,
      "memory(GiB)": 97.17,
      "step": 14985,
      "train_speed(iter/s)": 1.637467
    },
    {
      "acc": 0.67129154,
      "epoch": 0.38026382546930493,
      "grad_norm": 5.1875,
      "learning_rate": 9.47267210837358e-06,
      "loss": 1.57331486,
      "memory(GiB)": 97.17,
      "step": 14990,
      "train_speed(iter/s)": 1.637527
    },
    {
      "acc": 0.65323563,
      "epoch": 0.38039066463724,
      "grad_norm": 6.78125,
      "learning_rate": 9.472203276203667e-06,
      "loss": 1.641152,
      "memory(GiB)": 97.17,
      "step": 14995,
      "train_speed(iter/s)": 1.637588
    },
    {
      "acc": 0.65414772,
      "epoch": 0.380517503805175,
      "grad_norm": 5.46875,
      "learning_rate": 9.471734247326284e-06,
      "loss": 1.66306038,
      "memory(GiB)": 97.17,
      "step": 15000,
      "train_speed(iter/s)": 1.637649
    },
    {
      "epoch": 0.380517503805175,
      "eval_acc": 0.6420889743702191,
      "eval_loss": 1.5977007150650024,
      "eval_runtime": 58.6084,
      "eval_samples_per_second": 108.687,
      "eval_steps_per_second": 27.18,
      "step": 15000
    },
    {
      "acc": 0.65199609,
      "epoch": 0.3806443429731101,
      "grad_norm": 6.03125,
      "learning_rate": 9.471265021762067e-06,
      "loss": 1.61440811,
      "memory(GiB)": 97.17,
      "step": 15005,
      "train_speed(iter/s)": 1.626547
    },
    {
      "acc": 0.67658763,
      "epoch": 0.38077118214104516,
      "grad_norm": 6.5,
      "learning_rate": 9.47079559953165e-06,
      "loss": 1.50812092,
      "memory(GiB)": 97.17,
      "step": 15010,
      "train_speed(iter/s)": 1.626606
    },
    {
      "acc": 0.63453341,
      "epoch": 0.3808980213089802,
      "grad_norm": 5.9375,
      "learning_rate": 9.470325980655683e-06,
      "loss": 1.68023796,
      "memory(GiB)": 97.17,
      "step": 15015,
      "train_speed(iter/s)": 1.626665
    },
    {
      "acc": 0.6445303,
      "epoch": 0.38102486047691525,
      "grad_norm": 8.25,
      "learning_rate": 9.46985616515482e-06,
      "loss": 1.68390846,
      "memory(GiB)": 97.17,
      "step": 15020,
      "train_speed(iter/s)": 1.626722
    },
    {
      "acc": 0.64478121,
      "epoch": 0.38115169964485035,
      "grad_norm": 6.0,
      "learning_rate": 9.469386153049727e-06,
      "loss": 1.62552166,
      "memory(GiB)": 97.17,
      "step": 15025,
      "train_speed(iter/s)": 1.626778
    },
    {
      "acc": 0.65908871,
      "epoch": 0.3812785388127854,
      "grad_norm": 7.59375,
      "learning_rate": 9.468915944361076e-06,
      "loss": 1.55891399,
      "memory(GiB)": 97.17,
      "step": 15030,
      "train_speed(iter/s)": 1.626836
    },
    {
      "acc": 0.65880489,
      "epoch": 0.38140537798072044,
      "grad_norm": 4.53125,
      "learning_rate": 9.468445539109551e-06,
      "loss": 1.63597965,
      "memory(GiB)": 97.17,
      "step": 15035,
      "train_speed(iter/s)": 1.626895
    },
    {
      "acc": 0.6407299,
      "epoch": 0.3815322171486555,
      "grad_norm": 5.59375,
      "learning_rate": 9.46797493731584e-06,
      "loss": 1.70256844,
      "memory(GiB)": 97.17,
      "step": 15040,
      "train_speed(iter/s)": 1.62694
    },
    {
      "acc": 0.66172566,
      "epoch": 0.3816590563165906,
      "grad_norm": 4.875,
      "learning_rate": 9.467504139000642e-06,
      "loss": 1.61686726,
      "memory(GiB)": 97.17,
      "step": 15045,
      "train_speed(iter/s)": 1.626995
    },
    {
      "acc": 0.65152993,
      "epoch": 0.38178589548452563,
      "grad_norm": 4.84375,
      "learning_rate": 9.467033144184667e-06,
      "loss": 1.67275543,
      "memory(GiB)": 97.17,
      "step": 15050,
      "train_speed(iter/s)": 1.627045
    },
    {
      "acc": 0.6585288,
      "epoch": 0.3819127346524607,
      "grad_norm": 6.46875,
      "learning_rate": 9.466561952888632e-06,
      "loss": 1.6463604,
      "memory(GiB)": 97.17,
      "step": 15055,
      "train_speed(iter/s)": 1.627096
    },
    {
      "acc": 0.64626746,
      "epoch": 0.3820395738203957,
      "grad_norm": 6.28125,
      "learning_rate": 9.466090565133259e-06,
      "loss": 1.62370682,
      "memory(GiB)": 97.17,
      "step": 15060,
      "train_speed(iter/s)": 1.627154
    },
    {
      "acc": 0.65925994,
      "epoch": 0.3821664129883308,
      "grad_norm": 4.75,
      "learning_rate": 9.465618980939284e-06,
      "loss": 1.59799404,
      "memory(GiB)": 97.17,
      "step": 15065,
      "train_speed(iter/s)": 1.627212
    },
    {
      "acc": 0.66021891,
      "epoch": 0.38229325215626586,
      "grad_norm": 5.5,
      "learning_rate": 9.465147200327446e-06,
      "loss": 1.65960159,
      "memory(GiB)": 97.17,
      "step": 15070,
      "train_speed(iter/s)": 1.627264
    },
    {
      "acc": 0.65524716,
      "epoch": 0.3824200913242009,
      "grad_norm": 6.09375,
      "learning_rate": 9.464675223318503e-06,
      "loss": 1.55825253,
      "memory(GiB)": 97.17,
      "step": 15075,
      "train_speed(iter/s)": 1.627323
    },
    {
      "acc": 0.66851449,
      "epoch": 0.38254693049213595,
      "grad_norm": 6.40625,
      "learning_rate": 9.464203049933207e-06,
      "loss": 1.60118256,
      "memory(GiB)": 97.17,
      "step": 15080,
      "train_speed(iter/s)": 1.627377
    },
    {
      "acc": 0.68179879,
      "epoch": 0.38267376966007105,
      "grad_norm": 6.125,
      "learning_rate": 9.463730680192332e-06,
      "loss": 1.57144289,
      "memory(GiB)": 97.17,
      "step": 15085,
      "train_speed(iter/s)": 1.627435
    },
    {
      "acc": 0.64977884,
      "epoch": 0.3828006088280061,
      "grad_norm": 5.375,
      "learning_rate": 9.46325811411665e-06,
      "loss": 1.67484436,
      "memory(GiB)": 97.17,
      "step": 15090,
      "train_speed(iter/s)": 1.62749
    },
    {
      "acc": 0.65589213,
      "epoch": 0.38292744799594114,
      "grad_norm": 6.9375,
      "learning_rate": 9.462785351726951e-06,
      "loss": 1.61144466,
      "memory(GiB)": 97.17,
      "step": 15095,
      "train_speed(iter/s)": 1.627548
    },
    {
      "acc": 0.65560389,
      "epoch": 0.3830542871638762,
      "grad_norm": 5.375,
      "learning_rate": 9.462312393044027e-06,
      "loss": 1.60794601,
      "memory(GiB)": 97.17,
      "step": 15100,
      "train_speed(iter/s)": 1.627603
    },
    {
      "acc": 0.66775174,
      "epoch": 0.3831811263318113,
      "grad_norm": 5.5,
      "learning_rate": 9.46183923808868e-06,
      "loss": 1.58283987,
      "memory(GiB)": 97.17,
      "step": 15105,
      "train_speed(iter/s)": 1.627658
    },
    {
      "acc": 0.65754728,
      "epoch": 0.38330796549974633,
      "grad_norm": 5.46875,
      "learning_rate": 9.461365886881724e-06,
      "loss": 1.64416447,
      "memory(GiB)": 97.17,
      "step": 15110,
      "train_speed(iter/s)": 1.627709
    },
    {
      "acc": 0.65614381,
      "epoch": 0.3834348046676814,
      "grad_norm": 5.375,
      "learning_rate": 9.460892339443977e-06,
      "loss": 1.67810593,
      "memory(GiB)": 97.17,
      "step": 15115,
      "train_speed(iter/s)": 1.627766
    },
    {
      "acc": 0.65124025,
      "epoch": 0.3835616438356164,
      "grad_norm": 7.5625,
      "learning_rate": 9.460418595796268e-06,
      "loss": 1.61333828,
      "memory(GiB)": 97.17,
      "step": 15120,
      "train_speed(iter/s)": 1.62782
    },
    {
      "acc": 0.63805156,
      "epoch": 0.3836884830035515,
      "grad_norm": 4.71875,
      "learning_rate": 9.459944655959437e-06,
      "loss": 1.6731926,
      "memory(GiB)": 97.17,
      "step": 15125,
      "train_speed(iter/s)": 1.627875
    },
    {
      "acc": 0.66052732,
      "epoch": 0.38381532217148656,
      "grad_norm": 5.625,
      "learning_rate": 9.459470519954325e-06,
      "loss": 1.54127674,
      "memory(GiB)": 97.17,
      "step": 15130,
      "train_speed(iter/s)": 1.627933
    },
    {
      "acc": 0.65779576,
      "epoch": 0.3839421613394216,
      "grad_norm": 6.875,
      "learning_rate": 9.458996187801791e-06,
      "loss": 1.59846478,
      "memory(GiB)": 97.17,
      "step": 15135,
      "train_speed(iter/s)": 1.627992
    },
    {
      "acc": 0.63993168,
      "epoch": 0.38406900050735665,
      "grad_norm": 5.96875,
      "learning_rate": 9.458521659522697e-06,
      "loss": 1.67858791,
      "memory(GiB)": 97.17,
      "step": 15140,
      "train_speed(iter/s)": 1.628052
    },
    {
      "acc": 0.64192643,
      "epoch": 0.38419583967529175,
      "grad_norm": 5.46875,
      "learning_rate": 9.458046935137913e-06,
      "loss": 1.70274601,
      "memory(GiB)": 97.17,
      "step": 15145,
      "train_speed(iter/s)": 1.628112
    },
    {
      "acc": 0.66031561,
      "epoch": 0.3843226788432268,
      "grad_norm": 5.34375,
      "learning_rate": 9.457572014668323e-06,
      "loss": 1.63343391,
      "memory(GiB)": 97.17,
      "step": 15150,
      "train_speed(iter/s)": 1.62817
    },
    {
      "acc": 0.65354128,
      "epoch": 0.38444951801116184,
      "grad_norm": 4.96875,
      "learning_rate": 9.457096898134813e-06,
      "loss": 1.61082172,
      "memory(GiB)": 97.17,
      "step": 15155,
      "train_speed(iter/s)": 1.628224
    },
    {
      "acc": 0.66112552,
      "epoch": 0.3845763571790969,
      "grad_norm": 7.125,
      "learning_rate": 9.45662158555828e-06,
      "loss": 1.63481731,
      "memory(GiB)": 97.17,
      "step": 15160,
      "train_speed(iter/s)": 1.628283
    },
    {
      "acc": 0.65172977,
      "epoch": 0.384703196347032,
      "grad_norm": 7.9375,
      "learning_rate": 9.456146076959636e-06,
      "loss": 1.63406162,
      "memory(GiB)": 97.17,
      "step": 15165,
      "train_speed(iter/s)": 1.628343
    },
    {
      "acc": 0.66097212,
      "epoch": 0.38483003551496703,
      "grad_norm": 5.3125,
      "learning_rate": 9.455670372359791e-06,
      "loss": 1.66054554,
      "memory(GiB)": 97.17,
      "step": 15170,
      "train_speed(iter/s)": 1.628396
    },
    {
      "acc": 0.64635258,
      "epoch": 0.3849568746829021,
      "grad_norm": 4.96875,
      "learning_rate": 9.45519447177967e-06,
      "loss": 1.64068184,
      "memory(GiB)": 97.17,
      "step": 15175,
      "train_speed(iter/s)": 1.62845
    },
    {
      "acc": 0.65124454,
      "epoch": 0.3850837138508371,
      "grad_norm": 5.5625,
      "learning_rate": 9.454718375240204e-06,
      "loss": 1.67692146,
      "memory(GiB)": 97.17,
      "step": 15180,
      "train_speed(iter/s)": 1.628506
    },
    {
      "acc": 0.66304941,
      "epoch": 0.3852105530187722,
      "grad_norm": 6.875,
      "learning_rate": 9.454242082762336e-06,
      "loss": 1.58205814,
      "memory(GiB)": 97.17,
      "step": 15185,
      "train_speed(iter/s)": 1.628567
    },
    {
      "acc": 0.66110382,
      "epoch": 0.38533739218670726,
      "grad_norm": 7.0,
      "learning_rate": 9.453765594367014e-06,
      "loss": 1.62625275,
      "memory(GiB)": 97.17,
      "step": 15190,
      "train_speed(iter/s)": 1.628625
    },
    {
      "acc": 0.65365376,
      "epoch": 0.3854642313546423,
      "grad_norm": 5.40625,
      "learning_rate": 9.453288910075196e-06,
      "loss": 1.60047874,
      "memory(GiB)": 97.17,
      "step": 15195,
      "train_speed(iter/s)": 1.628684
    },
    {
      "acc": 0.65300121,
      "epoch": 0.38559107052257735,
      "grad_norm": 6.5,
      "learning_rate": 9.452812029907849e-06,
      "loss": 1.65511322,
      "memory(GiB)": 97.17,
      "step": 15200,
      "train_speed(iter/s)": 1.628738
    },
    {
      "acc": 0.64740038,
      "epoch": 0.38571790969051245,
      "grad_norm": 5.46875,
      "learning_rate": 9.452334953885951e-06,
      "loss": 1.64176769,
      "memory(GiB)": 97.17,
      "step": 15205,
      "train_speed(iter/s)": 1.628796
    },
    {
      "acc": 0.64705348,
      "epoch": 0.3858447488584475,
      "grad_norm": 6.0,
      "learning_rate": 9.451857682030481e-06,
      "loss": 1.66733418,
      "memory(GiB)": 97.17,
      "step": 15210,
      "train_speed(iter/s)": 1.628851
    },
    {
      "acc": 0.66407166,
      "epoch": 0.38597158802638254,
      "grad_norm": 6.84375,
      "learning_rate": 9.451380214362436e-06,
      "loss": 1.64468498,
      "memory(GiB)": 97.17,
      "step": 15215,
      "train_speed(iter/s)": 1.628906
    },
    {
      "acc": 0.65807333,
      "epoch": 0.3860984271943176,
      "grad_norm": 4.84375,
      "learning_rate": 9.450902550902814e-06,
      "loss": 1.61301842,
      "memory(GiB)": 97.17,
      "step": 15220,
      "train_speed(iter/s)": 1.62896
    },
    {
      "acc": 0.64157362,
      "epoch": 0.3862252663622527,
      "grad_norm": 5.34375,
      "learning_rate": 9.450424691672626e-06,
      "loss": 1.70941715,
      "memory(GiB)": 97.17,
      "step": 15225,
      "train_speed(iter/s)": 1.629015
    },
    {
      "acc": 0.65545411,
      "epoch": 0.38635210553018773,
      "grad_norm": 5.8125,
      "learning_rate": 9.449946636692891e-06,
      "loss": 1.59882326,
      "memory(GiB)": 97.17,
      "step": 15230,
      "train_speed(iter/s)": 1.629069
    },
    {
      "acc": 0.65746436,
      "epoch": 0.3864789446981228,
      "grad_norm": 6.125,
      "learning_rate": 9.449468385984634e-06,
      "loss": 1.60253696,
      "memory(GiB)": 97.17,
      "step": 15235,
      "train_speed(iter/s)": 1.629128
    },
    {
      "acc": 0.66188722,
      "epoch": 0.3866057838660578,
      "grad_norm": 6.71875,
      "learning_rate": 9.448989939568892e-06,
      "loss": 1.61767769,
      "memory(GiB)": 97.17,
      "step": 15240,
      "train_speed(iter/s)": 1.629186
    },
    {
      "acc": 0.63285279,
      "epoch": 0.3867326230339929,
      "grad_norm": 5.125,
      "learning_rate": 9.448511297466708e-06,
      "loss": 1.69206161,
      "memory(GiB)": 97.17,
      "step": 15245,
      "train_speed(iter/s)": 1.62924
    },
    {
      "acc": 0.6526329,
      "epoch": 0.38685946220192796,
      "grad_norm": 5.3125,
      "learning_rate": 9.448032459699139e-06,
      "loss": 1.59504023,
      "memory(GiB)": 97.17,
      "step": 15250,
      "train_speed(iter/s)": 1.629297
    },
    {
      "acc": 0.63574696,
      "epoch": 0.386986301369863,
      "grad_norm": 5.28125,
      "learning_rate": 9.447553426287244e-06,
      "loss": 1.70259533,
      "memory(GiB)": 97.17,
      "step": 15255,
      "train_speed(iter/s)": 1.629353
    },
    {
      "acc": 0.65058651,
      "epoch": 0.38711314053779805,
      "grad_norm": 4.78125,
      "learning_rate": 9.44707419725209e-06,
      "loss": 1.59770164,
      "memory(GiB)": 97.17,
      "step": 15260,
      "train_speed(iter/s)": 1.629407
    },
    {
      "acc": 0.6444521,
      "epoch": 0.38723997970573315,
      "grad_norm": 5.90625,
      "learning_rate": 9.446594772614759e-06,
      "loss": 1.69053574,
      "memory(GiB)": 97.17,
      "step": 15265,
      "train_speed(iter/s)": 1.629462
    },
    {
      "acc": 0.64778633,
      "epoch": 0.3873668188736682,
      "grad_norm": 5.28125,
      "learning_rate": 9.446115152396335e-06,
      "loss": 1.6015049,
      "memory(GiB)": 97.17,
      "step": 15270,
      "train_speed(iter/s)": 1.629516
    },
    {
      "acc": 0.66978006,
      "epoch": 0.38749365804160324,
      "grad_norm": 7.34375,
      "learning_rate": 9.445635336617919e-06,
      "loss": 1.55565643,
      "memory(GiB)": 97.17,
      "step": 15275,
      "train_speed(iter/s)": 1.629566
    },
    {
      "acc": 0.64932642,
      "epoch": 0.3876204972095383,
      "grad_norm": 4.21875,
      "learning_rate": 9.445155325300612e-06,
      "loss": 1.65540466,
      "memory(GiB)": 97.17,
      "step": 15280,
      "train_speed(iter/s)": 1.629619
    },
    {
      "acc": 0.65902166,
      "epoch": 0.3877473363774734,
      "grad_norm": 5.8125,
      "learning_rate": 9.444675118465528e-06,
      "loss": 1.64618721,
      "memory(GiB)": 97.17,
      "step": 15285,
      "train_speed(iter/s)": 1.629671
    },
    {
      "acc": 0.64727697,
      "epoch": 0.38787417554540843,
      "grad_norm": 5.15625,
      "learning_rate": 9.444194716133785e-06,
      "loss": 1.69308586,
      "memory(GiB)": 97.17,
      "step": 15290,
      "train_speed(iter/s)": 1.629722
    },
    {
      "acc": 0.65821066,
      "epoch": 0.3880010147133435,
      "grad_norm": 5.34375,
      "learning_rate": 9.44371411832652e-06,
      "loss": 1.57123346,
      "memory(GiB)": 97.17,
      "step": 15295,
      "train_speed(iter/s)": 1.629777
    },
    {
      "acc": 0.63835669,
      "epoch": 0.3881278538812785,
      "grad_norm": 5.4375,
      "learning_rate": 9.443233325064867e-06,
      "loss": 1.6920599,
      "memory(GiB)": 97.17,
      "step": 15300,
      "train_speed(iter/s)": 1.629828
    },
    {
      "acc": 0.64770451,
      "epoch": 0.3882546930492136,
      "grad_norm": 4.875,
      "learning_rate": 9.442752336369976e-06,
      "loss": 1.60383701,
      "memory(GiB)": 97.17,
      "step": 15305,
      "train_speed(iter/s)": 1.629884
    },
    {
      "acc": 0.67474632,
      "epoch": 0.38838153221714866,
      "grad_norm": 5.125,
      "learning_rate": 9.442271152263e-06,
      "loss": 1.54049063,
      "memory(GiB)": 97.17,
      "step": 15310,
      "train_speed(iter/s)": 1.629934
    },
    {
      "acc": 0.65135574,
      "epoch": 0.3885083713850837,
      "grad_norm": 4.96875,
      "learning_rate": 9.441789772765107e-06,
      "loss": 1.64121819,
      "memory(GiB)": 97.17,
      "step": 15315,
      "train_speed(iter/s)": 1.62999
    },
    {
      "acc": 0.64553738,
      "epoch": 0.38863521055301875,
      "grad_norm": 5.3125,
      "learning_rate": 9.441308197897467e-06,
      "loss": 1.63902283,
      "memory(GiB)": 97.17,
      "step": 15320,
      "train_speed(iter/s)": 1.630046
    },
    {
      "acc": 0.66621041,
      "epoch": 0.38876204972095385,
      "grad_norm": 6.375,
      "learning_rate": 9.440826427681264e-06,
      "loss": 1.66489449,
      "memory(GiB)": 97.17,
      "step": 15325,
      "train_speed(iter/s)": 1.630099
    },
    {
      "acc": 0.6516644,
      "epoch": 0.3888888888888889,
      "grad_norm": 6.6875,
      "learning_rate": 9.44034446213769e-06,
      "loss": 1.62715588,
      "memory(GiB)": 97.17,
      "step": 15330,
      "train_speed(iter/s)": 1.630155
    },
    {
      "acc": 0.65201325,
      "epoch": 0.38901572805682394,
      "grad_norm": 5.8125,
      "learning_rate": 9.439862301287939e-06,
      "loss": 1.61978149,
      "memory(GiB)": 97.17,
      "step": 15335,
      "train_speed(iter/s)": 1.630212
    },
    {
      "acc": 0.64715352,
      "epoch": 0.389142567224759,
      "grad_norm": 6.03125,
      "learning_rate": 9.439379945153223e-06,
      "loss": 1.69087639,
      "memory(GiB)": 97.17,
      "step": 15340,
      "train_speed(iter/s)": 1.630272
    },
    {
      "acc": 0.66509418,
      "epoch": 0.3892694063926941,
      "grad_norm": 6.34375,
      "learning_rate": 9.438897393754755e-06,
      "loss": 1.54576006,
      "memory(GiB)": 97.17,
      "step": 15345,
      "train_speed(iter/s)": 1.630333
    },
    {
      "acc": 0.6554986,
      "epoch": 0.38939624556062913,
      "grad_norm": 5.21875,
      "learning_rate": 9.438414647113762e-06,
      "loss": 1.60549564,
      "memory(GiB)": 97.17,
      "step": 15350,
      "train_speed(iter/s)": 1.63039
    },
    {
      "acc": 0.64943905,
      "epoch": 0.3895230847285642,
      "grad_norm": 5.46875,
      "learning_rate": 9.437931705251478e-06,
      "loss": 1.62872963,
      "memory(GiB)": 97.17,
      "step": 15355,
      "train_speed(iter/s)": 1.630448
    },
    {
      "acc": 0.65276423,
      "epoch": 0.3896499238964992,
      "grad_norm": 4.96875,
      "learning_rate": 9.437448568189142e-06,
      "loss": 1.55095549,
      "memory(GiB)": 97.17,
      "step": 15360,
      "train_speed(iter/s)": 1.630502
    },
    {
      "acc": 0.65781813,
      "epoch": 0.3897767630644343,
      "grad_norm": 5.03125,
      "learning_rate": 9.436965235948008e-06,
      "loss": 1.57868671,
      "memory(GiB)": 97.17,
      "step": 15365,
      "train_speed(iter/s)": 1.630559
    },
    {
      "acc": 0.65866489,
      "epoch": 0.38990360223236936,
      "grad_norm": 4.96875,
      "learning_rate": 9.436481708549332e-06,
      "loss": 1.64397202,
      "memory(GiB)": 97.17,
      "step": 15370,
      "train_speed(iter/s)": 1.630617
    },
    {
      "acc": 0.64708924,
      "epoch": 0.3900304414003044,
      "grad_norm": 5.40625,
      "learning_rate": 9.435997986014382e-06,
      "loss": 1.69053593,
      "memory(GiB)": 97.17,
      "step": 15375,
      "train_speed(iter/s)": 1.630675
    },
    {
      "acc": 0.6348774,
      "epoch": 0.39015728056823945,
      "grad_norm": 6.1875,
      "learning_rate": 9.435514068364437e-06,
      "loss": 1.72300053,
      "memory(GiB)": 97.17,
      "step": 15380,
      "train_speed(iter/s)": 1.630728
    },
    {
      "acc": 0.65201349,
      "epoch": 0.39028411973617455,
      "grad_norm": 5.28125,
      "learning_rate": 9.43502995562078e-06,
      "loss": 1.59301605,
      "memory(GiB)": 97.17,
      "step": 15385,
      "train_speed(iter/s)": 1.630781
    },
    {
      "acc": 0.65294523,
      "epoch": 0.3904109589041096,
      "grad_norm": 7.03125,
      "learning_rate": 9.434545647804703e-06,
      "loss": 1.65312443,
      "memory(GiB)": 97.17,
      "step": 15390,
      "train_speed(iter/s)": 1.63084
    },
    {
      "acc": 0.66024895,
      "epoch": 0.39053779807204464,
      "grad_norm": 6.3125,
      "learning_rate": 9.434061144937512e-06,
      "loss": 1.58912821,
      "memory(GiB)": 97.17,
      "step": 15395,
      "train_speed(iter/s)": 1.630895
    },
    {
      "acc": 0.66142116,
      "epoch": 0.3906646372399797,
      "grad_norm": 6.03125,
      "learning_rate": 9.433576447040513e-06,
      "loss": 1.56118126,
      "memory(GiB)": 97.17,
      "step": 15400,
      "train_speed(iter/s)": 1.630951
    },
    {
      "acc": 0.65852776,
      "epoch": 0.3907914764079148,
      "grad_norm": 5.1875,
      "learning_rate": 9.433091554135029e-06,
      "loss": 1.66822338,
      "memory(GiB)": 97.17,
      "step": 15405,
      "train_speed(iter/s)": 1.631006
    },
    {
      "acc": 0.64996738,
      "epoch": 0.39091831557584983,
      "grad_norm": 5.84375,
      "learning_rate": 9.432606466242384e-06,
      "loss": 1.68477383,
      "memory(GiB)": 97.17,
      "step": 15410,
      "train_speed(iter/s)": 1.63106
    },
    {
      "acc": 0.65065851,
      "epoch": 0.3910451547437849,
      "grad_norm": 4.875,
      "learning_rate": 9.43212118338392e-06,
      "loss": 1.59774828,
      "memory(GiB)": 97.17,
      "step": 15415,
      "train_speed(iter/s)": 1.631117
    },
    {
      "acc": 0.65187855,
      "epoch": 0.3911719939117199,
      "grad_norm": 5.25,
      "learning_rate": 9.431635705580975e-06,
      "loss": 1.61890888,
      "memory(GiB)": 97.17,
      "step": 15420,
      "train_speed(iter/s)": 1.631172
    },
    {
      "acc": 0.65411873,
      "epoch": 0.391298833079655,
      "grad_norm": 6.9375,
      "learning_rate": 9.431150032854907e-06,
      "loss": 1.61599655,
      "memory(GiB)": 97.17,
      "step": 15425,
      "train_speed(iter/s)": 1.631232
    },
    {
      "acc": 0.65631323,
      "epoch": 0.39142567224759006,
      "grad_norm": 6.375,
      "learning_rate": 9.430664165227077e-06,
      "loss": 1.55557957,
      "memory(GiB)": 97.17,
      "step": 15430,
      "train_speed(iter/s)": 1.631287
    },
    {
      "acc": 0.65505667,
      "epoch": 0.3915525114155251,
      "grad_norm": 5.28125,
      "learning_rate": 9.430178102718857e-06,
      "loss": 1.64248047,
      "memory(GiB)": 97.17,
      "step": 15435,
      "train_speed(iter/s)": 1.631342
    },
    {
      "acc": 0.66421671,
      "epoch": 0.39167935058346015,
      "grad_norm": 5.8125,
      "learning_rate": 9.429691845351623e-06,
      "loss": 1.57862911,
      "memory(GiB)": 97.17,
      "step": 15440,
      "train_speed(iter/s)": 1.631397
    },
    {
      "acc": 0.65027418,
      "epoch": 0.39180618975139525,
      "grad_norm": 5.1875,
      "learning_rate": 9.429205393146763e-06,
      "loss": 1.56260729,
      "memory(GiB)": 97.17,
      "step": 15445,
      "train_speed(iter/s)": 1.631451
    },
    {
      "acc": 0.64492798,
      "epoch": 0.3919330289193303,
      "grad_norm": 5.5625,
      "learning_rate": 9.428718746125678e-06,
      "loss": 1.67092323,
      "memory(GiB)": 97.17,
      "step": 15450,
      "train_speed(iter/s)": 1.63151
    },
    {
      "acc": 0.65922537,
      "epoch": 0.39205986808726534,
      "grad_norm": 5.5,
      "learning_rate": 9.428231904309768e-06,
      "loss": 1.66560287,
      "memory(GiB)": 97.17,
      "step": 15455,
      "train_speed(iter/s)": 1.631564
    },
    {
      "acc": 0.64279432,
      "epoch": 0.3921867072552004,
      "grad_norm": 5.46875,
      "learning_rate": 9.427744867720448e-06,
      "loss": 1.67148151,
      "memory(GiB)": 97.17,
      "step": 15460,
      "train_speed(iter/s)": 1.631617
    },
    {
      "acc": 0.65166764,
      "epoch": 0.3923135464231355,
      "grad_norm": 5.71875,
      "learning_rate": 9.42725763637914e-06,
      "loss": 1.63665276,
      "memory(GiB)": 97.17,
      "step": 15465,
      "train_speed(iter/s)": 1.631671
    },
    {
      "acc": 0.66557617,
      "epoch": 0.39244038559107053,
      "grad_norm": 5.65625,
      "learning_rate": 9.426770210307277e-06,
      "loss": 1.58535719,
      "memory(GiB)": 97.17,
      "step": 15470,
      "train_speed(iter/s)": 1.631726
    },
    {
      "acc": 0.64976187,
      "epoch": 0.3925672247590056,
      "grad_norm": 5.71875,
      "learning_rate": 9.426282589526294e-06,
      "loss": 1.64792614,
      "memory(GiB)": 97.17,
      "step": 15475,
      "train_speed(iter/s)": 1.631779
    },
    {
      "acc": 0.6637393,
      "epoch": 0.3926940639269406,
      "grad_norm": 4.90625,
      "learning_rate": 9.425794774057641e-06,
      "loss": 1.5826395,
      "memory(GiB)": 97.17,
      "step": 15480,
      "train_speed(iter/s)": 1.631836
    },
    {
      "acc": 0.65702786,
      "epoch": 0.3928209030948757,
      "grad_norm": 5.0625,
      "learning_rate": 9.425306763922775e-06,
      "loss": 1.61587639,
      "memory(GiB)": 97.17,
      "step": 15485,
      "train_speed(iter/s)": 1.631893
    },
    {
      "acc": 0.63265066,
      "epoch": 0.39294774226281076,
      "grad_norm": 7.125,
      "learning_rate": 9.42481855914316e-06,
      "loss": 1.67759323,
      "memory(GiB)": 97.17,
      "step": 15490,
      "train_speed(iter/s)": 1.631951
    },
    {
      "acc": 0.63959403,
      "epoch": 0.3930745814307458,
      "grad_norm": 7.1875,
      "learning_rate": 9.424330159740269e-06,
      "loss": 1.64781876,
      "memory(GiB)": 97.17,
      "step": 15495,
      "train_speed(iter/s)": 1.632001
    },
    {
      "acc": 0.6650527,
      "epoch": 0.39320142059868085,
      "grad_norm": 5.03125,
      "learning_rate": 9.423841565735582e-06,
      "loss": 1.50305843,
      "memory(GiB)": 97.17,
      "step": 15500,
      "train_speed(iter/s)": 1.632054
    },
    {
      "acc": 0.65494719,
      "epoch": 0.39332825976661595,
      "grad_norm": 5.4375,
      "learning_rate": 9.423352777150597e-06,
      "loss": 1.67718582,
      "memory(GiB)": 97.17,
      "step": 15505,
      "train_speed(iter/s)": 1.632108
    },
    {
      "acc": 0.66393356,
      "epoch": 0.393455098934551,
      "grad_norm": 6.53125,
      "learning_rate": 9.422863794006804e-06,
      "loss": 1.59034262,
      "memory(GiB)": 97.17,
      "step": 15510,
      "train_speed(iter/s)": 1.632163
    },
    {
      "acc": 0.64157934,
      "epoch": 0.39358193810248604,
      "grad_norm": 5.0625,
      "learning_rate": 9.422374616325716e-06,
      "loss": 1.67116318,
      "memory(GiB)": 97.17,
      "step": 15515,
      "train_speed(iter/s)": 1.632218
    },
    {
      "acc": 0.65052934,
      "epoch": 0.3937087772704211,
      "grad_norm": 6.0,
      "learning_rate": 9.421885244128847e-06,
      "loss": 1.64150238,
      "memory(GiB)": 97.17,
      "step": 15520,
      "train_speed(iter/s)": 1.632269
    },
    {
      "acc": 0.65489674,
      "epoch": 0.3938356164383562,
      "grad_norm": 5.71875,
      "learning_rate": 9.421395677437724e-06,
      "loss": 1.60311394,
      "memory(GiB)": 97.17,
      "step": 15525,
      "train_speed(iter/s)": 1.632321
    },
    {
      "acc": 0.66780558,
      "epoch": 0.39396245560629123,
      "grad_norm": 6.53125,
      "learning_rate": 9.42090591627388e-06,
      "loss": 1.58155308,
      "memory(GiB)": 97.17,
      "step": 15530,
      "train_speed(iter/s)": 1.632374
    },
    {
      "acc": 0.65116453,
      "epoch": 0.3940892947742263,
      "grad_norm": 6.6875,
      "learning_rate": 9.420415960658853e-06,
      "loss": 1.60542164,
      "memory(GiB)": 97.17,
      "step": 15535,
      "train_speed(iter/s)": 1.632422
    },
    {
      "acc": 0.65156603,
      "epoch": 0.3942161339421613,
      "grad_norm": 5.59375,
      "learning_rate": 9.419925810614196e-06,
      "loss": 1.63556404,
      "memory(GiB)": 97.17,
      "step": 15540,
      "train_speed(iter/s)": 1.632473
    },
    {
      "acc": 0.64251719,
      "epoch": 0.3943429731100964,
      "grad_norm": 8.8125,
      "learning_rate": 9.419435466161471e-06,
      "loss": 1.63585396,
      "memory(GiB)": 97.17,
      "step": 15545,
      "train_speed(iter/s)": 1.632529
    },
    {
      "acc": 0.64899192,
      "epoch": 0.39446981227803146,
      "grad_norm": 5.59375,
      "learning_rate": 9.418944927322242e-06,
      "loss": 1.70206203,
      "memory(GiB)": 97.17,
      "step": 15550,
      "train_speed(iter/s)": 1.632587
    },
    {
      "acc": 0.65994377,
      "epoch": 0.3945966514459665,
      "grad_norm": 5.875,
      "learning_rate": 9.418454194118085e-06,
      "loss": 1.5923357,
      "memory(GiB)": 97.17,
      "step": 15555,
      "train_speed(iter/s)": 1.632644
    },
    {
      "acc": 0.6623775,
      "epoch": 0.39472349061390155,
      "grad_norm": 5.78125,
      "learning_rate": 9.417963266570587e-06,
      "loss": 1.66040497,
      "memory(GiB)": 97.17,
      "step": 15560,
      "train_speed(iter/s)": 1.632696
    },
    {
      "acc": 0.65084915,
      "epoch": 0.39485032978183665,
      "grad_norm": 4.53125,
      "learning_rate": 9.417472144701338e-06,
      "loss": 1.61573162,
      "memory(GiB)": 97.17,
      "step": 15565,
      "train_speed(iter/s)": 1.632753
    },
    {
      "acc": 0.66019616,
      "epoch": 0.3949771689497717,
      "grad_norm": 5.5,
      "learning_rate": 9.416980828531944e-06,
      "loss": 1.5662322,
      "memory(GiB)": 97.17,
      "step": 15570,
      "train_speed(iter/s)": 1.632813
    },
    {
      "acc": 0.64340067,
      "epoch": 0.39510400811770674,
      "grad_norm": 5.78125,
      "learning_rate": 9.41648931808401e-06,
      "loss": 1.66267509,
      "memory(GiB)": 97.17,
      "step": 15575,
      "train_speed(iter/s)": 1.632871
    },
    {
      "acc": 0.64938602,
      "epoch": 0.3952308472856418,
      "grad_norm": 5.53125,
      "learning_rate": 9.415997613379159e-06,
      "loss": 1.61800346,
      "memory(GiB)": 97.17,
      "step": 15580,
      "train_speed(iter/s)": 1.632926
    },
    {
      "acc": 0.65750771,
      "epoch": 0.3953576864535769,
      "grad_norm": 5.8125,
      "learning_rate": 9.415505714439016e-06,
      "loss": 1.60547295,
      "memory(GiB)": 97.17,
      "step": 15585,
      "train_speed(iter/s)": 1.632984
    },
    {
      "acc": 0.64576778,
      "epoch": 0.39548452562151193,
      "grad_norm": 4.875,
      "learning_rate": 9.415013621285219e-06,
      "loss": 1.64392223,
      "memory(GiB)": 97.17,
      "step": 15590,
      "train_speed(iter/s)": 1.633039
    },
    {
      "acc": 0.6579072,
      "epoch": 0.395611364789447,
      "grad_norm": 6.09375,
      "learning_rate": 9.41452133393941e-06,
      "loss": 1.54888668,
      "memory(GiB)": 97.17,
      "step": 15595,
      "train_speed(iter/s)": 1.633095
    },
    {
      "acc": 0.66567664,
      "epoch": 0.395738203957382,
      "grad_norm": 4.78125,
      "learning_rate": 9.414028852423245e-06,
      "loss": 1.5840951,
      "memory(GiB)": 97.17,
      "step": 15600,
      "train_speed(iter/s)": 1.633147
    },
    {
      "acc": 0.66496782,
      "epoch": 0.3958650431253171,
      "grad_norm": 7.625,
      "learning_rate": 9.413536176758384e-06,
      "loss": 1.57238636,
      "memory(GiB)": 97.17,
      "step": 15605,
      "train_speed(iter/s)": 1.633205
    },
    {
      "acc": 0.64426823,
      "epoch": 0.39599188229325216,
      "grad_norm": 5.75,
      "learning_rate": 9.413043306966496e-06,
      "loss": 1.64144936,
      "memory(GiB)": 97.17,
      "step": 15610,
      "train_speed(iter/s)": 1.633264
    },
    {
      "acc": 0.67001653,
      "epoch": 0.3961187214611872,
      "grad_norm": 6.59375,
      "learning_rate": 9.41255024306926e-06,
      "loss": 1.57618198,
      "memory(GiB)": 97.17,
      "step": 15615,
      "train_speed(iter/s)": 1.633323
    },
    {
      "acc": 0.6681159,
      "epoch": 0.39624556062912225,
      "grad_norm": 5.90625,
      "learning_rate": 9.412056985088364e-06,
      "loss": 1.6572546,
      "memory(GiB)": 97.17,
      "step": 15620,
      "train_speed(iter/s)": 1.63338
    },
    {
      "acc": 0.64795952,
      "epoch": 0.39637239979705735,
      "grad_norm": 6.5,
      "learning_rate": 9.411563533045505e-06,
      "loss": 1.63303547,
      "memory(GiB)": 97.17,
      "step": 15625,
      "train_speed(iter/s)": 1.633437
    },
    {
      "acc": 0.64200125,
      "epoch": 0.3964992389649924,
      "grad_norm": 6.21875,
      "learning_rate": 9.411069886962383e-06,
      "loss": 1.68100014,
      "memory(GiB)": 97.17,
      "step": 15630,
      "train_speed(iter/s)": 1.63349
    },
    {
      "acc": 0.63818402,
      "epoch": 0.39662607813292744,
      "grad_norm": 7.1875,
      "learning_rate": 9.410576046860716e-06,
      "loss": 1.65536861,
      "memory(GiB)": 97.17,
      "step": 15635,
      "train_speed(iter/s)": 1.633549
    },
    {
      "acc": 0.6690546,
      "epoch": 0.3967529173008625,
      "grad_norm": 5.46875,
      "learning_rate": 9.41008201276222e-06,
      "loss": 1.58703251,
      "memory(GiB)": 97.17,
      "step": 15640,
      "train_speed(iter/s)": 1.633603
    },
    {
      "acc": 0.66416597,
      "epoch": 0.3968797564687976,
      "grad_norm": 5.3125,
      "learning_rate": 9.409587784688629e-06,
      "loss": 1.64223862,
      "memory(GiB)": 97.17,
      "step": 15645,
      "train_speed(iter/s)": 1.633657
    },
    {
      "acc": 0.62723894,
      "epoch": 0.39700659563673263,
      "grad_norm": 5.46875,
      "learning_rate": 9.409093362661678e-06,
      "loss": 1.7032486,
      "memory(GiB)": 97.17,
      "step": 15650,
      "train_speed(iter/s)": 1.633713
    },
    {
      "acc": 0.65629663,
      "epoch": 0.3971334348046677,
      "grad_norm": 6.15625,
      "learning_rate": 9.408598746703119e-06,
      "loss": 1.60428791,
      "memory(GiB)": 97.17,
      "step": 15655,
      "train_speed(iter/s)": 1.633768
    },
    {
      "acc": 0.65424824,
      "epoch": 0.3972602739726027,
      "grad_norm": 6.28125,
      "learning_rate": 9.408103936834703e-06,
      "loss": 1.65331917,
      "memory(GiB)": 97.17,
      "step": 15660,
      "train_speed(iter/s)": 1.633825
    },
    {
      "acc": 0.65117049,
      "epoch": 0.3973871131405378,
      "grad_norm": 5.6875,
      "learning_rate": 9.407608933078194e-06,
      "loss": 1.65230179,
      "memory(GiB)": 97.17,
      "step": 15665,
      "train_speed(iter/s)": 1.633878
    },
    {
      "acc": 0.65586481,
      "epoch": 0.39751395230847286,
      "grad_norm": 7.0,
      "learning_rate": 9.407113735455366e-06,
      "loss": 1.61515751,
      "memory(GiB)": 97.17,
      "step": 15670,
      "train_speed(iter/s)": 1.633933
    },
    {
      "acc": 0.65823298,
      "epoch": 0.3976407914764079,
      "grad_norm": 6.34375,
      "learning_rate": 9.406618343988e-06,
      "loss": 1.59263668,
      "memory(GiB)": 97.17,
      "step": 15675,
      "train_speed(iter/s)": 1.633988
    },
    {
      "acc": 0.65400338,
      "epoch": 0.39776763064434295,
      "grad_norm": 5.53125,
      "learning_rate": 9.406122758697885e-06,
      "loss": 1.63175106,
      "memory(GiB)": 97.17,
      "step": 15680,
      "train_speed(iter/s)": 1.634041
    },
    {
      "acc": 0.65228043,
      "epoch": 0.39789446981227805,
      "grad_norm": 4.90625,
      "learning_rate": 9.405626979606819e-06,
      "loss": 1.68690357,
      "memory(GiB)": 97.17,
      "step": 15685,
      "train_speed(iter/s)": 1.634093
    },
    {
      "acc": 0.65005369,
      "epoch": 0.3980213089802131,
      "grad_norm": 5.09375,
      "learning_rate": 9.405131006736608e-06,
      "loss": 1.61571674,
      "memory(GiB)": 97.17,
      "step": 15690,
      "train_speed(iter/s)": 1.634152
    },
    {
      "acc": 0.66304731,
      "epoch": 0.39814814814814814,
      "grad_norm": 5.59375,
      "learning_rate": 9.404634840109069e-06,
      "loss": 1.63206615,
      "memory(GiB)": 97.17,
      "step": 15695,
      "train_speed(iter/s)": 1.634209
    },
    {
      "acc": 0.6460711,
      "epoch": 0.3982749873160832,
      "grad_norm": 5.34375,
      "learning_rate": 9.404138479746022e-06,
      "loss": 1.63198643,
      "memory(GiB)": 97.17,
      "step": 15700,
      "train_speed(iter/s)": 1.63426
    },
    {
      "acc": 0.65446758,
      "epoch": 0.3984018264840183,
      "grad_norm": 5.875,
      "learning_rate": 9.403641925669304e-06,
      "loss": 1.61774788,
      "memory(GiB)": 97.17,
      "step": 15705,
      "train_speed(iter/s)": 1.634313
    },
    {
      "acc": 0.65057478,
      "epoch": 0.39852866565195333,
      "grad_norm": 4.28125,
      "learning_rate": 9.403145177900752e-06,
      "loss": 1.66977463,
      "memory(GiB)": 97.17,
      "step": 15710,
      "train_speed(iter/s)": 1.634365
    },
    {
      "acc": 0.6444972,
      "epoch": 0.3986555048198884,
      "grad_norm": 6.0,
      "learning_rate": 9.402648236462217e-06,
      "loss": 1.62636948,
      "memory(GiB)": 97.17,
      "step": 15715,
      "train_speed(iter/s)": 1.634423
    },
    {
      "acc": 0.65888119,
      "epoch": 0.3987823439878234,
      "grad_norm": 4.5625,
      "learning_rate": 9.402151101375557e-06,
      "loss": 1.63502407,
      "memory(GiB)": 97.17,
      "step": 15720,
      "train_speed(iter/s)": 1.634476
    },
    {
      "acc": 0.64015155,
      "epoch": 0.3989091831557585,
      "grad_norm": 5.1875,
      "learning_rate": 9.401653772662638e-06,
      "loss": 1.62309017,
      "memory(GiB)": 97.17,
      "step": 15725,
      "train_speed(iter/s)": 1.634532
    },
    {
      "acc": 0.65206957,
      "epoch": 0.39903602232369356,
      "grad_norm": 5.03125,
      "learning_rate": 9.401156250345331e-06,
      "loss": 1.6193222,
      "memory(GiB)": 97.17,
      "step": 15730,
      "train_speed(iter/s)": 1.634587
    },
    {
      "acc": 0.65838399,
      "epoch": 0.3991628614916286,
      "grad_norm": 5.46875,
      "learning_rate": 9.400658534445524e-06,
      "loss": 1.59883022,
      "memory(GiB)": 97.17,
      "step": 15735,
      "train_speed(iter/s)": 1.634638
    },
    {
      "acc": 0.63716321,
      "epoch": 0.39928970065956365,
      "grad_norm": 5.84375,
      "learning_rate": 9.40016062498511e-06,
      "loss": 1.63768959,
      "memory(GiB)": 97.17,
      "step": 15740,
      "train_speed(iter/s)": 1.63469
    },
    {
      "acc": 0.64570274,
      "epoch": 0.39941653982749875,
      "grad_norm": 6.0625,
      "learning_rate": 9.399662521985982e-06,
      "loss": 1.62347889,
      "memory(GiB)": 97.17,
      "step": 15745,
      "train_speed(iter/s)": 1.634736
    },
    {
      "acc": 0.65868735,
      "epoch": 0.3995433789954338,
      "grad_norm": 8.125,
      "learning_rate": 9.399164225470055e-06,
      "loss": 1.65992889,
      "memory(GiB)": 97.17,
      "step": 15750,
      "train_speed(iter/s)": 1.634786
    },
    {
      "acc": 0.65334015,
      "epoch": 0.39967021816336884,
      "grad_norm": 5.84375,
      "learning_rate": 9.398665735459245e-06,
      "loss": 1.65760117,
      "memory(GiB)": 97.17,
      "step": 15755,
      "train_speed(iter/s)": 1.634837
    },
    {
      "acc": 0.63274751,
      "epoch": 0.3997970573313039,
      "grad_norm": 4.59375,
      "learning_rate": 9.398167051975475e-06,
      "loss": 1.67615051,
      "memory(GiB)": 97.17,
      "step": 15760,
      "train_speed(iter/s)": 1.634888
    },
    {
      "acc": 0.6517416,
      "epoch": 0.399923896499239,
      "grad_norm": 6.40625,
      "learning_rate": 9.397668175040684e-06,
      "loss": 1.61494884,
      "memory(GiB)": 97.17,
      "step": 15765,
      "train_speed(iter/s)": 1.634939
    },
    {
      "acc": 0.63603888,
      "epoch": 0.40005073566717403,
      "grad_norm": 5.1875,
      "learning_rate": 9.397169104676813e-06,
      "loss": 1.66938114,
      "memory(GiB)": 97.17,
      "step": 15770,
      "train_speed(iter/s)": 1.634993
    },
    {
      "acc": 0.63894658,
      "epoch": 0.4001775748351091,
      "grad_norm": 5.4375,
      "learning_rate": 9.39666984090581e-06,
      "loss": 1.6665844,
      "memory(GiB)": 97.17,
      "step": 15775,
      "train_speed(iter/s)": 1.635045
    },
    {
      "acc": 0.66393557,
      "epoch": 0.4003044140030441,
      "grad_norm": 5.375,
      "learning_rate": 9.396170383749642e-06,
      "loss": 1.61611996,
      "memory(GiB)": 97.17,
      "step": 15780,
      "train_speed(iter/s)": 1.635099
    },
    {
      "acc": 0.65847626,
      "epoch": 0.4004312531709792,
      "grad_norm": 4.84375,
      "learning_rate": 9.39567073323027e-06,
      "loss": 1.60259666,
      "memory(GiB)": 97.17,
      "step": 15785,
      "train_speed(iter/s)": 1.63515
    },
    {
      "acc": 0.63672585,
      "epoch": 0.40055809233891426,
      "grad_norm": 5.4375,
      "learning_rate": 9.395170889369674e-06,
      "loss": 1.71591072,
      "memory(GiB)": 97.17,
      "step": 15790,
      "train_speed(iter/s)": 1.635205
    },
    {
      "acc": 0.65618362,
      "epoch": 0.4006849315068493,
      "grad_norm": 6.0,
      "learning_rate": 9.39467085218984e-06,
      "loss": 1.64862404,
      "memory(GiB)": 97.17,
      "step": 15795,
      "train_speed(iter/s)": 1.635256
    },
    {
      "acc": 0.64691854,
      "epoch": 0.40081177067478435,
      "grad_norm": 5.03125,
      "learning_rate": 9.39417062171276e-06,
      "loss": 1.62944794,
      "memory(GiB)": 97.17,
      "step": 15800,
      "train_speed(iter/s)": 1.635311
    },
    {
      "acc": 0.65817442,
      "epoch": 0.40093860984271945,
      "grad_norm": 5.875,
      "learning_rate": 9.393670197960439e-06,
      "loss": 1.63574867,
      "memory(GiB)": 97.17,
      "step": 15805,
      "train_speed(iter/s)": 1.635361
    },
    {
      "acc": 0.64882078,
      "epoch": 0.4010654490106545,
      "grad_norm": 5.1875,
      "learning_rate": 9.393169580954884e-06,
      "loss": 1.59303999,
      "memory(GiB)": 97.17,
      "step": 15810,
      "train_speed(iter/s)": 1.635415
    },
    {
      "acc": 0.64674039,
      "epoch": 0.40119228817858954,
      "grad_norm": 4.59375,
      "learning_rate": 9.392668770718118e-06,
      "loss": 1.68373737,
      "memory(GiB)": 97.17,
      "step": 15815,
      "train_speed(iter/s)": 1.635471
    },
    {
      "acc": 0.65925636,
      "epoch": 0.4013191273465246,
      "grad_norm": 5.5625,
      "learning_rate": 9.392167767272169e-06,
      "loss": 1.64430122,
      "memory(GiB)": 97.17,
      "step": 15820,
      "train_speed(iter/s)": 1.635527
    },
    {
      "acc": 0.64930334,
      "epoch": 0.4014459665144597,
      "grad_norm": 5.0625,
      "learning_rate": 9.39166657063907e-06,
      "loss": 1.66579933,
      "memory(GiB)": 97.17,
      "step": 15825,
      "train_speed(iter/s)": 1.635579
    },
    {
      "acc": 0.64079461,
      "epoch": 0.40157280568239473,
      "grad_norm": 4.90625,
      "learning_rate": 9.391165180840869e-06,
      "loss": 1.64372616,
      "memory(GiB)": 97.17,
      "step": 15830,
      "train_speed(iter/s)": 1.635633
    },
    {
      "acc": 0.65208101,
      "epoch": 0.4016996448503298,
      "grad_norm": 6.5,
      "learning_rate": 9.390663597899619e-06,
      "loss": 1.6628273,
      "memory(GiB)": 97.17,
      "step": 15835,
      "train_speed(iter/s)": 1.635687
    },
    {
      "acc": 0.66142225,
      "epoch": 0.4018264840182648,
      "grad_norm": 5.5,
      "learning_rate": 9.39016182183738e-06,
      "loss": 1.55630836,
      "memory(GiB)": 97.17,
      "step": 15840,
      "train_speed(iter/s)": 1.635743
    },
    {
      "acc": 0.66288266,
      "epoch": 0.4019533231861999,
      "grad_norm": 6.0,
      "learning_rate": 9.389659852676223e-06,
      "loss": 1.60005302,
      "memory(GiB)": 97.17,
      "step": 15845,
      "train_speed(iter/s)": 1.635794
    },
    {
      "acc": 0.67552366,
      "epoch": 0.40208016235413496,
      "grad_norm": 6.34375,
      "learning_rate": 9.389157690438228e-06,
      "loss": 1.50747881,
      "memory(GiB)": 97.17,
      "step": 15850,
      "train_speed(iter/s)": 1.635846
    },
    {
      "acc": 0.65831046,
      "epoch": 0.40220700152207,
      "grad_norm": 5.375,
      "learning_rate": 9.38865533514548e-06,
      "loss": 1.55039349,
      "memory(GiB)": 97.17,
      "step": 15855,
      "train_speed(iter/s)": 1.635901
    },
    {
      "acc": 0.66160603,
      "epoch": 0.40233384069000505,
      "grad_norm": 5.0625,
      "learning_rate": 9.388152786820078e-06,
      "loss": 1.55197296,
      "memory(GiB)": 97.17,
      "step": 15860,
      "train_speed(iter/s)": 1.635953
    },
    {
      "acc": 0.63546963,
      "epoch": 0.40246067985794015,
      "grad_norm": 5.40625,
      "learning_rate": 9.387650045484124e-06,
      "loss": 1.75525208,
      "memory(GiB)": 97.17,
      "step": 15865,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.6519228,
      "epoch": 0.4025875190258752,
      "grad_norm": 5.9375,
      "learning_rate": 9.387147111159734e-06,
      "loss": 1.61350021,
      "memory(GiB)": 97.17,
      "step": 15870,
      "train_speed(iter/s)": 1.636059
    },
    {
      "acc": 0.64786277,
      "epoch": 0.40271435819381024,
      "grad_norm": 7.15625,
      "learning_rate": 9.386643983869025e-06,
      "loss": 1.62597084,
      "memory(GiB)": 97.17,
      "step": 15875,
      "train_speed(iter/s)": 1.636115
    },
    {
      "acc": 0.65504198,
      "epoch": 0.4028411973617453,
      "grad_norm": 7.46875,
      "learning_rate": 9.386140663634127e-06,
      "loss": 1.68262711,
      "memory(GiB)": 97.17,
      "step": 15880,
      "train_speed(iter/s)": 1.636171
    },
    {
      "acc": 0.64914284,
      "epoch": 0.4029680365296804,
      "grad_norm": 5.625,
      "learning_rate": 9.385637150477182e-06,
      "loss": 1.6466259,
      "memory(GiB)": 97.17,
      "step": 15885,
      "train_speed(iter/s)": 1.636223
    },
    {
      "acc": 0.64096508,
      "epoch": 0.40309487569761543,
      "grad_norm": 6.25,
      "learning_rate": 9.385133444420333e-06,
      "loss": 1.71891117,
      "memory(GiB)": 97.17,
      "step": 15890,
      "train_speed(iter/s)": 1.636279
    },
    {
      "acc": 0.66728516,
      "epoch": 0.4032217148655505,
      "grad_norm": 6.5625,
      "learning_rate": 9.384629545485738e-06,
      "loss": 1.51819839,
      "memory(GiB)": 97.17,
      "step": 15895,
      "train_speed(iter/s)": 1.636334
    },
    {
      "acc": 0.64217792,
      "epoch": 0.4033485540334855,
      "grad_norm": 5.34375,
      "learning_rate": 9.38412545369556e-06,
      "loss": 1.63828697,
      "memory(GiB)": 97.17,
      "step": 15900,
      "train_speed(iter/s)": 1.636384
    },
    {
      "acc": 0.64461389,
      "epoch": 0.4034753932014206,
      "grad_norm": 7.8125,
      "learning_rate": 9.383621169071971e-06,
      "loss": 1.61118851,
      "memory(GiB)": 97.17,
      "step": 15905,
      "train_speed(iter/s)": 1.636441
    },
    {
      "acc": 0.66804571,
      "epoch": 0.40360223236935566,
      "grad_norm": 5.21875,
      "learning_rate": 9.38311669163715e-06,
      "loss": 1.64611588,
      "memory(GiB)": 97.17,
      "step": 15910,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.65931616,
      "epoch": 0.4037290715372907,
      "grad_norm": 6.84375,
      "learning_rate": 9.38261202141329e-06,
      "loss": 1.60034523,
      "memory(GiB)": 97.17,
      "step": 15915,
      "train_speed(iter/s)": 1.636546
    },
    {
      "acc": 0.64861155,
      "epoch": 0.40385591070522575,
      "grad_norm": 6.75,
      "learning_rate": 9.382107158422585e-06,
      "loss": 1.63590164,
      "memory(GiB)": 97.17,
      "step": 15920,
      "train_speed(iter/s)": 1.6366
    },
    {
      "acc": 0.65154324,
      "epoch": 0.40398274987316085,
      "grad_norm": 5.78125,
      "learning_rate": 9.381602102687241e-06,
      "loss": 1.6102478,
      "memory(GiB)": 97.17,
      "step": 15925,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.66373606,
      "epoch": 0.4041095890410959,
      "grad_norm": 5.9375,
      "learning_rate": 9.381096854229476e-06,
      "loss": 1.62204723,
      "memory(GiB)": 97.17,
      "step": 15930,
      "train_speed(iter/s)": 1.636701
    },
    {
      "acc": 0.66498594,
      "epoch": 0.40423642820903094,
      "grad_norm": 5.0625,
      "learning_rate": 9.38059141307151e-06,
      "loss": 1.60091343,
      "memory(GiB)": 97.17,
      "step": 15935,
      "train_speed(iter/s)": 1.636751
    },
    {
      "acc": 0.65378904,
      "epoch": 0.404363267376966,
      "grad_norm": 5.21875,
      "learning_rate": 9.380085779235577e-06,
      "loss": 1.60037174,
      "memory(GiB)": 97.17,
      "step": 15940,
      "train_speed(iter/s)": 1.636806
    },
    {
      "acc": 0.6509573,
      "epoch": 0.4044901065449011,
      "grad_norm": 4.75,
      "learning_rate": 9.379579952743916e-06,
      "loss": 1.63471603,
      "memory(GiB)": 97.17,
      "step": 15945,
      "train_speed(iter/s)": 1.63686
    },
    {
      "acc": 0.64465246,
      "epoch": 0.40461694571283613,
      "grad_norm": 5.875,
      "learning_rate": 9.379073933618774e-06,
      "loss": 1.61435032,
      "memory(GiB)": 97.17,
      "step": 15950,
      "train_speed(iter/s)": 1.636918
    },
    {
      "acc": 0.64981737,
      "epoch": 0.4047437848807712,
      "grad_norm": 5.28125,
      "learning_rate": 9.37856772188241e-06,
      "loss": 1.65799713,
      "memory(GiB)": 97.17,
      "step": 15955,
      "train_speed(iter/s)": 1.636972
    },
    {
      "acc": 0.64226127,
      "epoch": 0.4048706240487062,
      "grad_norm": 5.4375,
      "learning_rate": 9.378061317557088e-06,
      "loss": 1.60751457,
      "memory(GiB)": 97.17,
      "step": 15960,
      "train_speed(iter/s)": 1.637026
    },
    {
      "acc": 0.67459617,
      "epoch": 0.4049974632166413,
      "grad_norm": 5.875,
      "learning_rate": 9.377554720665083e-06,
      "loss": 1.60703316,
      "memory(GiB)": 97.17,
      "step": 15965,
      "train_speed(iter/s)": 1.637077
    },
    {
      "acc": 0.64980779,
      "epoch": 0.40512430238457636,
      "grad_norm": 5.09375,
      "learning_rate": 9.377047931228677e-06,
      "loss": 1.67774754,
      "memory(GiB)": 97.17,
      "step": 15970,
      "train_speed(iter/s)": 1.637131
    },
    {
      "acc": 0.66006041,
      "epoch": 0.4052511415525114,
      "grad_norm": 6.875,
      "learning_rate": 9.376540949270161e-06,
      "loss": 1.58426991,
      "memory(GiB)": 97.17,
      "step": 15975,
      "train_speed(iter/s)": 1.637185
    },
    {
      "acc": 0.65663538,
      "epoch": 0.40537798072044645,
      "grad_norm": 5.40625,
      "learning_rate": 9.376033774811833e-06,
      "loss": 1.5896637,
      "memory(GiB)": 97.17,
      "step": 15980,
      "train_speed(iter/s)": 1.637242
    },
    {
      "acc": 0.66919851,
      "epoch": 0.40550481988838155,
      "grad_norm": 6.0,
      "learning_rate": 9.375526407876003e-06,
      "loss": 1.57763729,
      "memory(GiB)": 97.17,
      "step": 15985,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.64241471,
      "epoch": 0.4056316590563166,
      "grad_norm": 6.09375,
      "learning_rate": 9.375018848484987e-06,
      "loss": 1.67906704,
      "memory(GiB)": 97.17,
      "step": 15990,
      "train_speed(iter/s)": 1.637346
    },
    {
      "acc": 0.66374674,
      "epoch": 0.40575849822425164,
      "grad_norm": 4.75,
      "learning_rate": 9.374511096661108e-06,
      "loss": 1.60427475,
      "memory(GiB)": 97.17,
      "step": 15995,
      "train_speed(iter/s)": 1.637396
    },
    {
      "acc": 0.64524317,
      "epoch": 0.4058853373921867,
      "grad_norm": 5.0,
      "learning_rate": 9.374003152426701e-06,
      "loss": 1.64672585,
      "memory(GiB)": 97.17,
      "step": 16000,
      "train_speed(iter/s)": 1.637447
    },
    {
      "epoch": 0.4058853373921867,
      "eval_acc": 0.6424895114825512,
      "eval_loss": 1.5949453115463257,
      "eval_runtime": 58.7857,
      "eval_samples_per_second": 108.36,
      "eval_steps_per_second": 27.098,
      "step": 16000
    },
    {
      "acc": 0.64903059,
      "epoch": 0.4060121765601218,
      "grad_norm": 6.09375,
      "learning_rate": 9.373495015804106e-06,
      "loss": 1.60470886,
      "memory(GiB)": 97.17,
      "step": 16005,
      "train_speed(iter/s)": 1.627009
    },
    {
      "acc": 0.64625101,
      "epoch": 0.40613901572805683,
      "grad_norm": 5.75,
      "learning_rate": 9.372986686815674e-06,
      "loss": 1.66825714,
      "memory(GiB)": 97.17,
      "step": 16010,
      "train_speed(iter/s)": 1.627063
    },
    {
      "acc": 0.6507658,
      "epoch": 0.4062658548959919,
      "grad_norm": 6.09375,
      "learning_rate": 9.372478165483763e-06,
      "loss": 1.59801474,
      "memory(GiB)": 97.17,
      "step": 16015,
      "train_speed(iter/s)": 1.627117
    },
    {
      "acc": 0.65511169,
      "epoch": 0.4063926940639269,
      "grad_norm": 5.5625,
      "learning_rate": 9.371969451830743e-06,
      "loss": 1.6586689,
      "memory(GiB)": 97.17,
      "step": 16020,
      "train_speed(iter/s)": 1.627168
    },
    {
      "acc": 0.65719781,
      "epoch": 0.406519533231862,
      "grad_norm": 5.375,
      "learning_rate": 9.371460545878986e-06,
      "loss": 1.60247536,
      "memory(GiB)": 97.17,
      "step": 16025,
      "train_speed(iter/s)": 1.627223
    },
    {
      "acc": 0.65161324,
      "epoch": 0.40664637239979706,
      "grad_norm": 5.9375,
      "learning_rate": 9.370951447650875e-06,
      "loss": 1.57683468,
      "memory(GiB)": 97.17,
      "step": 16030,
      "train_speed(iter/s)": 1.627273
    },
    {
      "acc": 0.63173494,
      "epoch": 0.4067732115677321,
      "grad_norm": 5.28125,
      "learning_rate": 9.370442157168806e-06,
      "loss": 1.68955822,
      "memory(GiB)": 97.17,
      "step": 16035,
      "train_speed(iter/s)": 1.627326
    },
    {
      "acc": 0.64423733,
      "epoch": 0.40690005073566715,
      "grad_norm": 5.21875,
      "learning_rate": 9.369932674455177e-06,
      "loss": 1.65101261,
      "memory(GiB)": 97.17,
      "step": 16040,
      "train_speed(iter/s)": 1.62738
    },
    {
      "acc": 0.65857348,
      "epoch": 0.40702688990360225,
      "grad_norm": 5.53125,
      "learning_rate": 9.3694229995324e-06,
      "loss": 1.5409668,
      "memory(GiB)": 97.17,
      "step": 16045,
      "train_speed(iter/s)": 1.627428
    },
    {
      "acc": 0.66051173,
      "epoch": 0.4071537290715373,
      "grad_norm": 6.3125,
      "learning_rate": 9.368913132422891e-06,
      "loss": 1.60270748,
      "memory(GiB)": 97.17,
      "step": 16050,
      "train_speed(iter/s)": 1.627482
    },
    {
      "acc": 0.65240541,
      "epoch": 0.40728056823947234,
      "grad_norm": 5.875,
      "learning_rate": 9.368403073149079e-06,
      "loss": 1.65346661,
      "memory(GiB)": 97.17,
      "step": 16055,
      "train_speed(iter/s)": 1.627534
    },
    {
      "acc": 0.64385419,
      "epoch": 0.4074074074074074,
      "grad_norm": 6.65625,
      "learning_rate": 9.367892821733393e-06,
      "loss": 1.60929718,
      "memory(GiB)": 97.17,
      "step": 16060,
      "train_speed(iter/s)": 1.627583
    },
    {
      "acc": 0.65275679,
      "epoch": 0.4075342465753425,
      "grad_norm": 5.4375,
      "learning_rate": 9.367382378198282e-06,
      "loss": 1.62133484,
      "memory(GiB)": 97.17,
      "step": 16065,
      "train_speed(iter/s)": 1.627638
    },
    {
      "acc": 0.65248361,
      "epoch": 0.40766108574327753,
      "grad_norm": 6.96875,
      "learning_rate": 9.366871742566193e-06,
      "loss": 1.58335047,
      "memory(GiB)": 97.17,
      "step": 16070,
      "train_speed(iter/s)": 1.627689
    },
    {
      "acc": 0.63841887,
      "epoch": 0.4077879249112126,
      "grad_norm": 5.96875,
      "learning_rate": 9.36636091485959e-06,
      "loss": 1.65226517,
      "memory(GiB)": 97.17,
      "step": 16075,
      "train_speed(iter/s)": 1.627743
    },
    {
      "acc": 0.65547657,
      "epoch": 0.4079147640791476,
      "grad_norm": 5.46875,
      "learning_rate": 9.365849895100939e-06,
      "loss": 1.59378586,
      "memory(GiB)": 97.17,
      "step": 16080,
      "train_speed(iter/s)": 1.627795
    },
    {
      "acc": 0.66481771,
      "epoch": 0.4080416032470827,
      "grad_norm": 7.78125,
      "learning_rate": 9.365338683312714e-06,
      "loss": 1.58041439,
      "memory(GiB)": 97.17,
      "step": 16085,
      "train_speed(iter/s)": 1.627852
    },
    {
      "acc": 0.65277824,
      "epoch": 0.40816844241501776,
      "grad_norm": 4.96875,
      "learning_rate": 9.364827279517408e-06,
      "loss": 1.62052765,
      "memory(GiB)": 97.17,
      "step": 16090,
      "train_speed(iter/s)": 1.627907
    },
    {
      "acc": 0.64046726,
      "epoch": 0.4082952815829528,
      "grad_norm": 4.96875,
      "learning_rate": 9.36431568373751e-06,
      "loss": 1.67497082,
      "memory(GiB)": 97.17,
      "step": 16095,
      "train_speed(iter/s)": 1.627959
    },
    {
      "acc": 0.65935326,
      "epoch": 0.40842212075088785,
      "grad_norm": 5.78125,
      "learning_rate": 9.363803895995522e-06,
      "loss": 1.57289524,
      "memory(GiB)": 97.17,
      "step": 16100,
      "train_speed(iter/s)": 1.628014
    },
    {
      "acc": 0.6536623,
      "epoch": 0.40854895991882295,
      "grad_norm": 5.5625,
      "learning_rate": 9.363291916313955e-06,
      "loss": 1.60950966,
      "memory(GiB)": 97.17,
      "step": 16105,
      "train_speed(iter/s)": 1.628068
    },
    {
      "acc": 0.64895964,
      "epoch": 0.408675799086758,
      "grad_norm": 4.84375,
      "learning_rate": 9.362779744715332e-06,
      "loss": 1.63531952,
      "memory(GiB)": 97.17,
      "step": 16110,
      "train_speed(iter/s)": 1.628119
    },
    {
      "acc": 0.6486104,
      "epoch": 0.40880263825469304,
      "grad_norm": 6.03125,
      "learning_rate": 9.362267381222174e-06,
      "loss": 1.65392284,
      "memory(GiB)": 97.17,
      "step": 16115,
      "train_speed(iter/s)": 1.628175
    },
    {
      "acc": 0.65669584,
      "epoch": 0.4089294774226281,
      "grad_norm": 5.09375,
      "learning_rate": 9.361754825857022e-06,
      "loss": 1.62155495,
      "memory(GiB)": 97.17,
      "step": 16120,
      "train_speed(iter/s)": 1.628226
    },
    {
      "acc": 0.6453598,
      "epoch": 0.4090563165905632,
      "grad_norm": 5.5,
      "learning_rate": 9.36124207864242e-06,
      "loss": 1.64764557,
      "memory(GiB)": 97.17,
      "step": 16125,
      "train_speed(iter/s)": 1.628281
    },
    {
      "acc": 0.64444523,
      "epoch": 0.40918315575849823,
      "grad_norm": 5.6875,
      "learning_rate": 9.360729139600917e-06,
      "loss": 1.65023556,
      "memory(GiB)": 97.17,
      "step": 16130,
      "train_speed(iter/s)": 1.62834
    },
    {
      "acc": 0.65969701,
      "epoch": 0.4093099949264333,
      "grad_norm": 5.25,
      "learning_rate": 9.36021600875508e-06,
      "loss": 1.63509293,
      "memory(GiB)": 97.17,
      "step": 16135,
      "train_speed(iter/s)": 1.628394
    },
    {
      "acc": 0.6343359,
      "epoch": 0.4094368340943683,
      "grad_norm": 5.625,
      "learning_rate": 9.359702686127474e-06,
      "loss": 1.68029633,
      "memory(GiB)": 97.17,
      "step": 16140,
      "train_speed(iter/s)": 1.628448
    },
    {
      "acc": 0.65928364,
      "epoch": 0.4095636732623034,
      "grad_norm": 5.15625,
      "learning_rate": 9.359189171740679e-06,
      "loss": 1.63962669,
      "memory(GiB)": 97.17,
      "step": 16145,
      "train_speed(iter/s)": 1.628504
    },
    {
      "acc": 0.64124012,
      "epoch": 0.40969051243023846,
      "grad_norm": 5.75,
      "learning_rate": 9.358675465617283e-06,
      "loss": 1.66341,
      "memory(GiB)": 97.17,
      "step": 16150,
      "train_speed(iter/s)": 1.62856
    },
    {
      "acc": 0.64791241,
      "epoch": 0.4098173515981735,
      "grad_norm": 5.75,
      "learning_rate": 9.35816156777988e-06,
      "loss": 1.65935631,
      "memory(GiB)": 97.17,
      "step": 16155,
      "train_speed(iter/s)": 1.628614
    },
    {
      "acc": 0.6422976,
      "epoch": 0.40994419076610855,
      "grad_norm": 6.96875,
      "learning_rate": 9.357647478251072e-06,
      "loss": 1.64075851,
      "memory(GiB)": 97.17,
      "step": 16160,
      "train_speed(iter/s)": 1.628669
    },
    {
      "acc": 0.64418387,
      "epoch": 0.41007102993404365,
      "grad_norm": 6.1875,
      "learning_rate": 9.357133197053475e-06,
      "loss": 1.69670563,
      "memory(GiB)": 97.17,
      "step": 16165,
      "train_speed(iter/s)": 1.628727
    },
    {
      "acc": 0.66096764,
      "epoch": 0.4101978691019787,
      "grad_norm": 9.4375,
      "learning_rate": 9.356618724209704e-06,
      "loss": 1.60868568,
      "memory(GiB)": 97.17,
      "step": 16170,
      "train_speed(iter/s)": 1.628781
    },
    {
      "acc": 0.65348921,
      "epoch": 0.41032470826991374,
      "grad_norm": 6.90625,
      "learning_rate": 9.356104059742392e-06,
      "loss": 1.59801893,
      "memory(GiB)": 97.17,
      "step": 16175,
      "train_speed(iter/s)": 1.628836
    },
    {
      "acc": 0.65164595,
      "epoch": 0.4104515474378488,
      "grad_norm": 5.46875,
      "learning_rate": 9.355589203674175e-06,
      "loss": 1.6303215,
      "memory(GiB)": 97.17,
      "step": 16180,
      "train_speed(iter/s)": 1.628889
    },
    {
      "acc": 0.66172752,
      "epoch": 0.4105783866057839,
      "grad_norm": 6.84375,
      "learning_rate": 9.355074156027699e-06,
      "loss": 1.58874168,
      "memory(GiB)": 97.17,
      "step": 16185,
      "train_speed(iter/s)": 1.628946
    },
    {
      "acc": 0.64384103,
      "epoch": 0.41070522577371893,
      "grad_norm": 5.5625,
      "learning_rate": 9.354558916825616e-06,
      "loss": 1.68263836,
      "memory(GiB)": 97.17,
      "step": 16190,
      "train_speed(iter/s)": 1.628999
    },
    {
      "acc": 0.65688691,
      "epoch": 0.410832064941654,
      "grad_norm": 5.03125,
      "learning_rate": 9.354043486090592e-06,
      "loss": 1.59741135,
      "memory(GiB)": 97.17,
      "step": 16195,
      "train_speed(iter/s)": 1.62905
    },
    {
      "acc": 0.64088583,
      "epoch": 0.410958904109589,
      "grad_norm": 5.59375,
      "learning_rate": 9.353527863845296e-06,
      "loss": 1.63057804,
      "memory(GiB)": 97.17,
      "step": 16200,
      "train_speed(iter/s)": 1.629105
    },
    {
      "acc": 0.66965623,
      "epoch": 0.4110857432775241,
      "grad_norm": 4.96875,
      "learning_rate": 9.353012050112405e-06,
      "loss": 1.5989255,
      "memory(GiB)": 97.17,
      "step": 16205,
      "train_speed(iter/s)": 1.629159
    },
    {
      "acc": 0.64907036,
      "epoch": 0.41121258244545916,
      "grad_norm": 5.25,
      "learning_rate": 9.352496044914611e-06,
      "loss": 1.63676128,
      "memory(GiB)": 97.17,
      "step": 16210,
      "train_speed(iter/s)": 1.629209
    },
    {
      "acc": 0.65607471,
      "epoch": 0.4113394216133942,
      "grad_norm": 7.375,
      "learning_rate": 9.351979848274608e-06,
      "loss": 1.62216969,
      "memory(GiB)": 97.17,
      "step": 16215,
      "train_speed(iter/s)": 1.629262
    },
    {
      "acc": 0.66906185,
      "epoch": 0.41146626078132925,
      "grad_norm": 4.875,
      "learning_rate": 9.351463460215102e-06,
      "loss": 1.52141571,
      "memory(GiB)": 97.17,
      "step": 16220,
      "train_speed(iter/s)": 1.629303
    },
    {
      "acc": 0.66006007,
      "epoch": 0.41159309994926435,
      "grad_norm": 6.4375,
      "learning_rate": 9.350946880758804e-06,
      "loss": 1.56992044,
      "memory(GiB)": 97.17,
      "step": 16225,
      "train_speed(iter/s)": 1.629356
    },
    {
      "acc": 0.65177946,
      "epoch": 0.4117199391171994,
      "grad_norm": 5.3125,
      "learning_rate": 9.350430109928437e-06,
      "loss": 1.61403236,
      "memory(GiB)": 97.17,
      "step": 16230,
      "train_speed(iter/s)": 1.62941
    },
    {
      "acc": 0.65621557,
      "epoch": 0.41184677828513444,
      "grad_norm": 6.78125,
      "learning_rate": 9.349913147746731e-06,
      "loss": 1.68774567,
      "memory(GiB)": 97.17,
      "step": 16235,
      "train_speed(iter/s)": 1.629464
    },
    {
      "acc": 0.66037645,
      "epoch": 0.4119736174530695,
      "grad_norm": 5.1875,
      "learning_rate": 9.349395994236423e-06,
      "loss": 1.57108192,
      "memory(GiB)": 97.17,
      "step": 16240,
      "train_speed(iter/s)": 1.629517
    },
    {
      "acc": 0.6542994,
      "epoch": 0.4121004566210046,
      "grad_norm": 5.4375,
      "learning_rate": 9.348878649420262e-06,
      "loss": 1.69590435,
      "memory(GiB)": 97.17,
      "step": 16245,
      "train_speed(iter/s)": 1.629573
    },
    {
      "acc": 0.64440928,
      "epoch": 0.41222729578893963,
      "grad_norm": 5.9375,
      "learning_rate": 9.348361113321e-06,
      "loss": 1.64741974,
      "memory(GiB)": 97.17,
      "step": 16250,
      "train_speed(iter/s)": 1.629623
    },
    {
      "acc": 0.65193729,
      "epoch": 0.4123541349568747,
      "grad_norm": 5.4375,
      "learning_rate": 9.347843385961403e-06,
      "loss": 1.64539108,
      "memory(GiB)": 97.17,
      "step": 16255,
      "train_speed(iter/s)": 1.629677
    },
    {
      "acc": 0.63490973,
      "epoch": 0.4124809741248097,
      "grad_norm": 5.875,
      "learning_rate": 9.347325467364242e-06,
      "loss": 1.6491951,
      "memory(GiB)": 97.17,
      "step": 16260,
      "train_speed(iter/s)": 1.629731
    },
    {
      "acc": 0.66243963,
      "epoch": 0.4126078132927448,
      "grad_norm": 5.28125,
      "learning_rate": 9.346807357552296e-06,
      "loss": 1.59512444,
      "memory(GiB)": 97.17,
      "step": 16265,
      "train_speed(iter/s)": 1.629784
    },
    {
      "acc": 0.64454498,
      "epoch": 0.41273465246067986,
      "grad_norm": 4.96875,
      "learning_rate": 9.346289056548357e-06,
      "loss": 1.65488186,
      "memory(GiB)": 97.17,
      "step": 16270,
      "train_speed(iter/s)": 1.62984
    },
    {
      "acc": 0.66899223,
      "epoch": 0.4128614916286149,
      "grad_norm": 5.5625,
      "learning_rate": 9.345770564375221e-06,
      "loss": 1.565625,
      "memory(GiB)": 97.17,
      "step": 16275,
      "train_speed(iter/s)": 1.629891
    },
    {
      "acc": 0.64188261,
      "epoch": 0.41298833079654995,
      "grad_norm": 4.90625,
      "learning_rate": 9.345251881055692e-06,
      "loss": 1.63279953,
      "memory(GiB)": 97.17,
      "step": 16280,
      "train_speed(iter/s)": 1.629946
    },
    {
      "acc": 0.65170283,
      "epoch": 0.41311516996448505,
      "grad_norm": 6.21875,
      "learning_rate": 9.344733006612585e-06,
      "loss": 1.67423515,
      "memory(GiB)": 97.17,
      "step": 16285,
      "train_speed(iter/s)": 1.629999
    },
    {
      "acc": 0.65542493,
      "epoch": 0.4132420091324201,
      "grad_norm": 5.53125,
      "learning_rate": 9.344213941068724e-06,
      "loss": 1.61397209,
      "memory(GiB)": 97.17,
      "step": 16290,
      "train_speed(iter/s)": 1.630053
    },
    {
      "acc": 0.64678516,
      "epoch": 0.41336884830035514,
      "grad_norm": 5.875,
      "learning_rate": 9.343694684446937e-06,
      "loss": 1.64910469,
      "memory(GiB)": 97.17,
      "step": 16295,
      "train_speed(iter/s)": 1.630105
    },
    {
      "acc": 0.6524024,
      "epoch": 0.4134956874682902,
      "grad_norm": 5.1875,
      "learning_rate": 9.343175236770065e-06,
      "loss": 1.56770487,
      "memory(GiB)": 97.17,
      "step": 16300,
      "train_speed(iter/s)": 1.630159
    },
    {
      "acc": 0.66147881,
      "epoch": 0.4136225266362253,
      "grad_norm": 4.46875,
      "learning_rate": 9.342655598060955e-06,
      "loss": 1.59089117,
      "memory(GiB)": 97.17,
      "step": 16305,
      "train_speed(iter/s)": 1.630213
    },
    {
      "acc": 0.6551693,
      "epoch": 0.41374936580416033,
      "grad_norm": 4.5625,
      "learning_rate": 9.342135768342464e-06,
      "loss": 1.58866253,
      "memory(GiB)": 97.17,
      "step": 16310,
      "train_speed(iter/s)": 1.630265
    },
    {
      "acc": 0.66414738,
      "epoch": 0.4138762049720954,
      "grad_norm": 5.5,
      "learning_rate": 9.341615747637454e-06,
      "loss": 1.58338203,
      "memory(GiB)": 97.17,
      "step": 16315,
      "train_speed(iter/s)": 1.630319
    },
    {
      "acc": 0.64451828,
      "epoch": 0.4140030441400304,
      "grad_norm": 5.09375,
      "learning_rate": 9.3410955359688e-06,
      "loss": 1.66825867,
      "memory(GiB)": 97.17,
      "step": 16320,
      "train_speed(iter/s)": 1.630374
    },
    {
      "acc": 0.66090703,
      "epoch": 0.4141298833079655,
      "grad_norm": 6.90625,
      "learning_rate": 9.340575133359385e-06,
      "loss": 1.59784298,
      "memory(GiB)": 97.17,
      "step": 16325,
      "train_speed(iter/s)": 1.630426
    },
    {
      "acc": 0.65299778,
      "epoch": 0.41425672247590056,
      "grad_norm": 5.6875,
      "learning_rate": 9.340054539832095e-06,
      "loss": 1.5794241,
      "memory(GiB)": 97.17,
      "step": 16330,
      "train_speed(iter/s)": 1.630477
    },
    {
      "acc": 0.64211025,
      "epoch": 0.4143835616438356,
      "grad_norm": 6.78125,
      "learning_rate": 9.339533755409828e-06,
      "loss": 1.68089371,
      "memory(GiB)": 97.17,
      "step": 16335,
      "train_speed(iter/s)": 1.630532
    },
    {
      "acc": 0.64001141,
      "epoch": 0.41451040081177065,
      "grad_norm": 5.5,
      "learning_rate": 9.339012780115492e-06,
      "loss": 1.6290781,
      "memory(GiB)": 97.17,
      "step": 16340,
      "train_speed(iter/s)": 1.630583
    },
    {
      "acc": 0.6527792,
      "epoch": 0.41463723997970575,
      "grad_norm": 5.03125,
      "learning_rate": 9.338491613972002e-06,
      "loss": 1.59525042,
      "memory(GiB)": 97.17,
      "step": 16345,
      "train_speed(iter/s)": 1.630632
    },
    {
      "acc": 0.67494831,
      "epoch": 0.4147640791476408,
      "grad_norm": 5.46875,
      "learning_rate": 9.337970257002282e-06,
      "loss": 1.60703545,
      "memory(GiB)": 97.17,
      "step": 16350,
      "train_speed(iter/s)": 1.630682
    },
    {
      "acc": 0.64884949,
      "epoch": 0.41489091831557584,
      "grad_norm": 6.53125,
      "learning_rate": 9.337448709229261e-06,
      "loss": 1.70602894,
      "memory(GiB)": 97.17,
      "step": 16355,
      "train_speed(iter/s)": 1.630739
    },
    {
      "acc": 0.66629572,
      "epoch": 0.4150177574835109,
      "grad_norm": 5.96875,
      "learning_rate": 9.336926970675883e-06,
      "loss": 1.62642403,
      "memory(GiB)": 97.17,
      "step": 16360,
      "train_speed(iter/s)": 1.630795
    },
    {
      "acc": 0.6303978,
      "epoch": 0.415144596651446,
      "grad_norm": 5.34375,
      "learning_rate": 9.33640504136509e-06,
      "loss": 1.71301994,
      "memory(GiB)": 97.17,
      "step": 16365,
      "train_speed(iter/s)": 1.630847
    },
    {
      "acc": 0.65087194,
      "epoch": 0.41527143581938103,
      "grad_norm": 5.84375,
      "learning_rate": 9.335882921319845e-06,
      "loss": 1.61233215,
      "memory(GiB)": 97.17,
      "step": 16370,
      "train_speed(iter/s)": 1.630902
    },
    {
      "acc": 0.66915307,
      "epoch": 0.4153982749873161,
      "grad_norm": 5.21875,
      "learning_rate": 9.335360610563111e-06,
      "loss": 1.62349396,
      "memory(GiB)": 97.17,
      "step": 16375,
      "train_speed(iter/s)": 1.630954
    },
    {
      "acc": 0.65744815,
      "epoch": 0.4155251141552511,
      "grad_norm": 7.40625,
      "learning_rate": 9.33483810911786e-06,
      "loss": 1.6356617,
      "memory(GiB)": 97.17,
      "step": 16380,
      "train_speed(iter/s)": 1.631004
    },
    {
      "acc": 0.63689585,
      "epoch": 0.4156519533231862,
      "grad_norm": 5.71875,
      "learning_rate": 9.334315417007079e-06,
      "loss": 1.69260674,
      "memory(GiB)": 97.17,
      "step": 16385,
      "train_speed(iter/s)": 1.631052
    },
    {
      "acc": 0.64266686,
      "epoch": 0.41577879249112126,
      "grad_norm": 4.96875,
      "learning_rate": 9.333792534253751e-06,
      "loss": 1.63932495,
      "memory(GiB)": 97.17,
      "step": 16390,
      "train_speed(iter/s)": 1.631106
    },
    {
      "acc": 0.64490991,
      "epoch": 0.4159056316590563,
      "grad_norm": 5.40625,
      "learning_rate": 9.333269460880879e-06,
      "loss": 1.62875023,
      "memory(GiB)": 97.17,
      "step": 16395,
      "train_speed(iter/s)": 1.631158
    },
    {
      "acc": 0.65804968,
      "epoch": 0.41603247082699135,
      "grad_norm": 7.375,
      "learning_rate": 9.33274619691147e-06,
      "loss": 1.5533864,
      "memory(GiB)": 97.17,
      "step": 16400,
      "train_speed(iter/s)": 1.631211
    },
    {
      "acc": 0.649788,
      "epoch": 0.41615930999492645,
      "grad_norm": 6.125,
      "learning_rate": 9.332222742368537e-06,
      "loss": 1.58417892,
      "memory(GiB)": 97.17,
      "step": 16405,
      "train_speed(iter/s)": 1.631264
    },
    {
      "acc": 0.64279757,
      "epoch": 0.4162861491628615,
      "grad_norm": 5.25,
      "learning_rate": 9.331699097275108e-06,
      "loss": 1.66128044,
      "memory(GiB)": 97.17,
      "step": 16410,
      "train_speed(iter/s)": 1.631319
    },
    {
      "acc": 0.65956573,
      "epoch": 0.41641298833079654,
      "grad_norm": 7.875,
      "learning_rate": 9.331175261654213e-06,
      "loss": 1.61678085,
      "memory(GiB)": 97.17,
      "step": 16415,
      "train_speed(iter/s)": 1.631374
    },
    {
      "acc": 0.64912167,
      "epoch": 0.4165398274987316,
      "grad_norm": 5.4375,
      "learning_rate": 9.330651235528891e-06,
      "loss": 1.64932976,
      "memory(GiB)": 97.17,
      "step": 16420,
      "train_speed(iter/s)": 1.631424
    },
    {
      "acc": 0.64392099,
      "epoch": 0.4166666666666667,
      "grad_norm": 5.375,
      "learning_rate": 9.330127018922195e-06,
      "loss": 1.66445293,
      "memory(GiB)": 97.17,
      "step": 16425,
      "train_speed(iter/s)": 1.631475
    },
    {
      "acc": 0.66941986,
      "epoch": 0.41679350583460173,
      "grad_norm": 5.5,
      "learning_rate": 9.329602611857179e-06,
      "loss": 1.53443508,
      "memory(GiB)": 97.17,
      "step": 16430,
      "train_speed(iter/s)": 1.631525
    },
    {
      "acc": 0.65015574,
      "epoch": 0.4169203450025368,
      "grad_norm": 5.875,
      "learning_rate": 9.329078014356909e-06,
      "loss": 1.59227047,
      "memory(GiB)": 97.17,
      "step": 16435,
      "train_speed(iter/s)": 1.631578
    },
    {
      "acc": 0.67195711,
      "epoch": 0.4170471841704718,
      "grad_norm": 6.21875,
      "learning_rate": 9.32855322644446e-06,
      "loss": 1.52868013,
      "memory(GiB)": 97.17,
      "step": 16440,
      "train_speed(iter/s)": 1.63163
    },
    {
      "acc": 0.65971193,
      "epoch": 0.4171740233384069,
      "grad_norm": 5.78125,
      "learning_rate": 9.328028248142916e-06,
      "loss": 1.55929604,
      "memory(GiB)": 97.17,
      "step": 16445,
      "train_speed(iter/s)": 1.63168
    },
    {
      "acc": 0.64351778,
      "epoch": 0.41730086250634196,
      "grad_norm": 4.71875,
      "learning_rate": 9.327503079475365e-06,
      "loss": 1.67029533,
      "memory(GiB)": 97.17,
      "step": 16450,
      "train_speed(iter/s)": 1.631732
    },
    {
      "acc": 0.64811516,
      "epoch": 0.417427701674277,
      "grad_norm": 6.125,
      "learning_rate": 9.326977720464908e-06,
      "loss": 1.61634636,
      "memory(GiB)": 97.17,
      "step": 16455,
      "train_speed(iter/s)": 1.631787
    },
    {
      "acc": 0.65875635,
      "epoch": 0.41755454084221205,
      "grad_norm": 6.1875,
      "learning_rate": 9.326452171134652e-06,
      "loss": 1.61829872,
      "memory(GiB)": 97.17,
      "step": 16460,
      "train_speed(iter/s)": 1.631836
    },
    {
      "acc": 0.63802261,
      "epoch": 0.41768138001014715,
      "grad_norm": 6.5,
      "learning_rate": 9.325926431507714e-06,
      "loss": 1.64991493,
      "memory(GiB)": 97.17,
      "step": 16465,
      "train_speed(iter/s)": 1.631886
    },
    {
      "acc": 0.65828586,
      "epoch": 0.4178082191780822,
      "grad_norm": 5.625,
      "learning_rate": 9.325400501607218e-06,
      "loss": 1.61011219,
      "memory(GiB)": 97.17,
      "step": 16470,
      "train_speed(iter/s)": 1.631935
    },
    {
      "acc": 0.64187222,
      "epoch": 0.41793505834601724,
      "grad_norm": 6.125,
      "learning_rate": 9.324874381456295e-06,
      "loss": 1.66432953,
      "memory(GiB)": 97.17,
      "step": 16475,
      "train_speed(iter/s)": 1.631986
    },
    {
      "acc": 0.65641108,
      "epoch": 0.4180618975139523,
      "grad_norm": 7.1875,
      "learning_rate": 9.324348071078088e-06,
      "loss": 1.60426331,
      "memory(GiB)": 97.17,
      "step": 16480,
      "train_speed(iter/s)": 1.632041
    },
    {
      "acc": 0.64060287,
      "epoch": 0.4181887366818874,
      "grad_norm": 5.15625,
      "learning_rate": 9.323821570495748e-06,
      "loss": 1.66343651,
      "memory(GiB)": 97.17,
      "step": 16485,
      "train_speed(iter/s)": 1.63209
    },
    {
      "acc": 0.64553232,
      "epoch": 0.41831557584982243,
      "grad_norm": 5.75,
      "learning_rate": 9.32329487973243e-06,
      "loss": 1.63939247,
      "memory(GiB)": 97.17,
      "step": 16490,
      "train_speed(iter/s)": 1.63214
    },
    {
      "acc": 0.66361637,
      "epoch": 0.4184424150177575,
      "grad_norm": 5.5,
      "learning_rate": 9.3227679988113e-06,
      "loss": 1.63096085,
      "memory(GiB)": 97.17,
      "step": 16495,
      "train_speed(iter/s)": 1.632191
    },
    {
      "acc": 0.66125531,
      "epoch": 0.4185692541856925,
      "grad_norm": 6.5625,
      "learning_rate": 9.322240927755534e-06,
      "loss": 1.60955887,
      "memory(GiB)": 97.17,
      "step": 16500,
      "train_speed(iter/s)": 1.632239
    },
    {
      "acc": 0.64777756,
      "epoch": 0.4186960933536276,
      "grad_norm": 7.4375,
      "learning_rate": 9.321713666588314e-06,
      "loss": 1.62595806,
      "memory(GiB)": 97.17,
      "step": 16505,
      "train_speed(iter/s)": 1.632289
    },
    {
      "acc": 0.65439548,
      "epoch": 0.41882293252156266,
      "grad_norm": 5.1875,
      "learning_rate": 9.321186215332833e-06,
      "loss": 1.67922421,
      "memory(GiB)": 97.17,
      "step": 16510,
      "train_speed(iter/s)": 1.632342
    },
    {
      "acc": 0.65504904,
      "epoch": 0.4189497716894977,
      "grad_norm": 5.25,
      "learning_rate": 9.320658574012289e-06,
      "loss": 1.63084793,
      "memory(GiB)": 97.17,
      "step": 16515,
      "train_speed(iter/s)": 1.632397
    },
    {
      "acc": 0.65674686,
      "epoch": 0.41907661085743275,
      "grad_norm": 5.5,
      "learning_rate": 9.32013074264989e-06,
      "loss": 1.61073532,
      "memory(GiB)": 97.17,
      "step": 16520,
      "train_speed(iter/s)": 1.632445
    },
    {
      "acc": 0.65759077,
      "epoch": 0.41920345002536785,
      "grad_norm": 6.09375,
      "learning_rate": 9.319602721268853e-06,
      "loss": 1.59918833,
      "memory(GiB)": 97.17,
      "step": 16525,
      "train_speed(iter/s)": 1.632498
    },
    {
      "acc": 0.6814713,
      "epoch": 0.4193302891933029,
      "grad_norm": 5.15625,
      "learning_rate": 9.319074509892403e-06,
      "loss": 1.53227873,
      "memory(GiB)": 97.17,
      "step": 16530,
      "train_speed(iter/s)": 1.632549
    },
    {
      "acc": 0.65880942,
      "epoch": 0.41945712836123794,
      "grad_norm": 4.9375,
      "learning_rate": 9.318546108543774e-06,
      "loss": 1.55095558,
      "memory(GiB)": 97.17,
      "step": 16535,
      "train_speed(iter/s)": 1.632603
    },
    {
      "acc": 0.63179932,
      "epoch": 0.419583967529173,
      "grad_norm": 5.5,
      "learning_rate": 9.318017517246205e-06,
      "loss": 1.68021927,
      "memory(GiB)": 97.17,
      "step": 16540,
      "train_speed(iter/s)": 1.632653
    },
    {
      "acc": 0.6292191,
      "epoch": 0.4197108066971081,
      "grad_norm": 6.53125,
      "learning_rate": 9.317488736022948e-06,
      "loss": 1.69453812,
      "memory(GiB)": 97.17,
      "step": 16545,
      "train_speed(iter/s)": 1.632705
    },
    {
      "acc": 0.65382562,
      "epoch": 0.41983764586504313,
      "grad_norm": 6.75,
      "learning_rate": 9.316959764897259e-06,
      "loss": 1.64304657,
      "memory(GiB)": 97.17,
      "step": 16550,
      "train_speed(iter/s)": 1.632759
    },
    {
      "acc": 0.64357138,
      "epoch": 0.4199644850329782,
      "grad_norm": 5.875,
      "learning_rate": 9.316430603892406e-06,
      "loss": 1.61357231,
      "memory(GiB)": 97.17,
      "step": 16555,
      "train_speed(iter/s)": 1.632814
    },
    {
      "acc": 0.64302549,
      "epoch": 0.4200913242009132,
      "grad_norm": 4.9375,
      "learning_rate": 9.315901253031663e-06,
      "loss": 1.64227886,
      "memory(GiB)": 97.17,
      "step": 16560,
      "train_speed(iter/s)": 1.632866
    },
    {
      "acc": 0.64157295,
      "epoch": 0.4202181633688483,
      "grad_norm": 6.6875,
      "learning_rate": 9.315371712338315e-06,
      "loss": 1.7676487,
      "memory(GiB)": 97.17,
      "step": 16565,
      "train_speed(iter/s)": 1.632919
    },
    {
      "acc": 0.65801296,
      "epoch": 0.42034500253678336,
      "grad_norm": 5.40625,
      "learning_rate": 9.314841981835652e-06,
      "loss": 1.59332809,
      "memory(GiB)": 97.17,
      "step": 16570,
      "train_speed(iter/s)": 1.632974
    },
    {
      "acc": 0.64057236,
      "epoch": 0.4204718417047184,
      "grad_norm": 5.75,
      "learning_rate": 9.314312061546974e-06,
      "loss": 1.6918026,
      "memory(GiB)": 97.17,
      "step": 16575,
      "train_speed(iter/s)": 1.633024
    },
    {
      "acc": 0.65233259,
      "epoch": 0.42059868087265345,
      "grad_norm": 4.5625,
      "learning_rate": 9.313781951495588e-06,
      "loss": 1.66234646,
      "memory(GiB)": 97.17,
      "step": 16580,
      "train_speed(iter/s)": 1.633077
    },
    {
      "acc": 0.65132189,
      "epoch": 0.42072552004058855,
      "grad_norm": 5.46875,
      "learning_rate": 9.313251651704816e-06,
      "loss": 1.64941978,
      "memory(GiB)": 97.17,
      "step": 16585,
      "train_speed(iter/s)": 1.63313
    },
    {
      "acc": 0.6595892,
      "epoch": 0.4208523592085236,
      "grad_norm": 5.0625,
      "learning_rate": 9.312721162197975e-06,
      "loss": 1.60508728,
      "memory(GiB)": 97.17,
      "step": 16590,
      "train_speed(iter/s)": 1.633182
    },
    {
      "acc": 0.66196313,
      "epoch": 0.42097919837645864,
      "grad_norm": 5.25,
      "learning_rate": 9.312190482998405e-06,
      "loss": 1.59453812,
      "memory(GiB)": 97.17,
      "step": 16595,
      "train_speed(iter/s)": 1.633241
    },
    {
      "acc": 0.64996533,
      "epoch": 0.4211060375443937,
      "grad_norm": 5.65625,
      "learning_rate": 9.311659614129443e-06,
      "loss": 1.66943264,
      "memory(GiB)": 97.17,
      "step": 16600,
      "train_speed(iter/s)": 1.633295
    },
    {
      "acc": 0.64626307,
      "epoch": 0.4212328767123288,
      "grad_norm": 6.125,
      "learning_rate": 9.311128555614443e-06,
      "loss": 1.65645447,
      "memory(GiB)": 97.17,
      "step": 16605,
      "train_speed(iter/s)": 1.633353
    },
    {
      "acc": 0.65463581,
      "epoch": 0.42135971588026383,
      "grad_norm": 5.0,
      "learning_rate": 9.31059730747676e-06,
      "loss": 1.55136127,
      "memory(GiB)": 97.17,
      "step": 16610,
      "train_speed(iter/s)": 1.633404
    },
    {
      "acc": 0.65890212,
      "epoch": 0.4214865550481989,
      "grad_norm": 5.5,
      "learning_rate": 9.310065869739763e-06,
      "loss": 1.6205616,
      "memory(GiB)": 97.17,
      "step": 16615,
      "train_speed(iter/s)": 1.633459
    },
    {
      "acc": 0.6442235,
      "epoch": 0.4216133942161339,
      "grad_norm": 5.34375,
      "learning_rate": 9.309534242426826e-06,
      "loss": 1.59124451,
      "memory(GiB)": 97.17,
      "step": 16620,
      "train_speed(iter/s)": 1.633512
    },
    {
      "acc": 0.65964456,
      "epoch": 0.421740233384069,
      "grad_norm": 4.78125,
      "learning_rate": 9.30900242556133e-06,
      "loss": 1.58672934,
      "memory(GiB)": 97.17,
      "step": 16625,
      "train_speed(iter/s)": 1.633566
    },
    {
      "acc": 0.65840259,
      "epoch": 0.42186707255200406,
      "grad_norm": 6.15625,
      "learning_rate": 9.308470419166672e-06,
      "loss": 1.55596886,
      "memory(GiB)": 97.17,
      "step": 16630,
      "train_speed(iter/s)": 1.633617
    },
    {
      "acc": 0.64595361,
      "epoch": 0.4219939117199391,
      "grad_norm": 5.28125,
      "learning_rate": 9.307938223266247e-06,
      "loss": 1.63031864,
      "memory(GiB)": 97.17,
      "step": 16635,
      "train_speed(iter/s)": 1.633672
    },
    {
      "acc": 0.63774099,
      "epoch": 0.42212075088787415,
      "grad_norm": 4.59375,
      "learning_rate": 9.307405837883467e-06,
      "loss": 1.70536747,
      "memory(GiB)": 97.17,
      "step": 16640,
      "train_speed(iter/s)": 1.633725
    },
    {
      "acc": 0.64718413,
      "epoch": 0.42224759005580925,
      "grad_norm": 5.25,
      "learning_rate": 9.306873263041745e-06,
      "loss": 1.62035789,
      "memory(GiB)": 97.17,
      "step": 16645,
      "train_speed(iter/s)": 1.633774
    },
    {
      "acc": 0.65107799,
      "epoch": 0.4223744292237443,
      "grad_norm": 6.625,
      "learning_rate": 9.30634049876451e-06,
      "loss": 1.6117794,
      "memory(GiB)": 97.17,
      "step": 16650,
      "train_speed(iter/s)": 1.633829
    },
    {
      "acc": 0.64570847,
      "epoch": 0.42250126839167934,
      "grad_norm": 5.96875,
      "learning_rate": 9.305807545075194e-06,
      "loss": 1.60762367,
      "memory(GiB)": 97.17,
      "step": 16655,
      "train_speed(iter/s)": 1.633881
    },
    {
      "acc": 0.65536709,
      "epoch": 0.4226281075596144,
      "grad_norm": 4.78125,
      "learning_rate": 9.305274401997237e-06,
      "loss": 1.62804165,
      "memory(GiB)": 97.17,
      "step": 16660,
      "train_speed(iter/s)": 1.633935
    },
    {
      "acc": 0.65016775,
      "epoch": 0.4227549467275495,
      "grad_norm": 5.3125,
      "learning_rate": 9.304741069554088e-06,
      "loss": 1.59783401,
      "memory(GiB)": 97.17,
      "step": 16665,
      "train_speed(iter/s)": 1.633985
    },
    {
      "acc": 0.66279917,
      "epoch": 0.42288178589548453,
      "grad_norm": 5.28125,
      "learning_rate": 9.304207547769211e-06,
      "loss": 1.57599201,
      "memory(GiB)": 97.17,
      "step": 16670,
      "train_speed(iter/s)": 1.634035
    },
    {
      "acc": 0.66747637,
      "epoch": 0.4230086250634196,
      "grad_norm": 6.46875,
      "learning_rate": 9.303673836666067e-06,
      "loss": 1.61142273,
      "memory(GiB)": 97.17,
      "step": 16675,
      "train_speed(iter/s)": 1.634087
    },
    {
      "acc": 0.64322491,
      "epoch": 0.4231354642313546,
      "grad_norm": 5.25,
      "learning_rate": 9.303139936268133e-06,
      "loss": 1.66456013,
      "memory(GiB)": 97.17,
      "step": 16680,
      "train_speed(iter/s)": 1.634144
    },
    {
      "acc": 0.63724728,
      "epoch": 0.4232623033992897,
      "grad_norm": 5.25,
      "learning_rate": 9.302605846598894e-06,
      "loss": 1.6882061,
      "memory(GiB)": 97.17,
      "step": 16685,
      "train_speed(iter/s)": 1.634198
    },
    {
      "acc": 0.64323635,
      "epoch": 0.42338914256722476,
      "grad_norm": 6.6875,
      "learning_rate": 9.30207156768184e-06,
      "loss": 1.62905502,
      "memory(GiB)": 97.17,
      "step": 16690,
      "train_speed(iter/s)": 1.634252
    },
    {
      "acc": 0.65594854,
      "epoch": 0.4235159817351598,
      "grad_norm": 6.09375,
      "learning_rate": 9.30153709954047e-06,
      "loss": 1.56245403,
      "memory(GiB)": 97.17,
      "step": 16695,
      "train_speed(iter/s)": 1.634306
    },
    {
      "acc": 0.65969977,
      "epoch": 0.42364282090309485,
      "grad_norm": 5.0,
      "learning_rate": 9.301002442198294e-06,
      "loss": 1.55928516,
      "memory(GiB)": 97.17,
      "step": 16700,
      "train_speed(iter/s)": 1.634358
    },
    {
      "acc": 0.65157671,
      "epoch": 0.42376966007102995,
      "grad_norm": 5.8125,
      "learning_rate": 9.300467595678829e-06,
      "loss": 1.6262455,
      "memory(GiB)": 97.17,
      "step": 16705,
      "train_speed(iter/s)": 1.634412
    },
    {
      "acc": 0.64952097,
      "epoch": 0.423896499238965,
      "grad_norm": 7.09375,
      "learning_rate": 9.299932560005596e-06,
      "loss": 1.65484352,
      "memory(GiB)": 97.17,
      "step": 16710,
      "train_speed(iter/s)": 1.634467
    },
    {
      "acc": 0.65549059,
      "epoch": 0.42402333840690004,
      "grad_norm": 5.90625,
      "learning_rate": 9.299397335202133e-06,
      "loss": 1.64444237,
      "memory(GiB)": 97.17,
      "step": 16715,
      "train_speed(iter/s)": 1.634521
    },
    {
      "acc": 0.66014671,
      "epoch": 0.4241501775748351,
      "grad_norm": 5.90625,
      "learning_rate": 9.29886192129198e-06,
      "loss": 1.61765747,
      "memory(GiB)": 97.17,
      "step": 16720,
      "train_speed(iter/s)": 1.634571
    },
    {
      "acc": 0.63968897,
      "epoch": 0.4242770167427702,
      "grad_norm": 6.15625,
      "learning_rate": 9.298326318298688e-06,
      "loss": 1.656534,
      "memory(GiB)": 97.17,
      "step": 16725,
      "train_speed(iter/s)": 1.634623
    },
    {
      "acc": 0.647507,
      "epoch": 0.42440385591070523,
      "grad_norm": 5.46875,
      "learning_rate": 9.29779052624581e-06,
      "loss": 1.58004341,
      "memory(GiB)": 97.17,
      "step": 16730,
      "train_speed(iter/s)": 1.634678
    },
    {
      "acc": 0.65747089,
      "epoch": 0.4245306950786403,
      "grad_norm": 6.8125,
      "learning_rate": 9.29725454515692e-06,
      "loss": 1.62379761,
      "memory(GiB)": 97.17,
      "step": 16735,
      "train_speed(iter/s)": 1.634732
    },
    {
      "acc": 0.65892248,
      "epoch": 0.4246575342465753,
      "grad_norm": 5.625,
      "learning_rate": 9.296718375055587e-06,
      "loss": 1.54168205,
      "memory(GiB)": 97.17,
      "step": 16740,
      "train_speed(iter/s)": 1.634786
    },
    {
      "acc": 0.65943336,
      "epoch": 0.4247843734145104,
      "grad_norm": 5.5,
      "learning_rate": 9.296182015965399e-06,
      "loss": 1.63737831,
      "memory(GiB)": 97.17,
      "step": 16745,
      "train_speed(iter/s)": 1.634839
    },
    {
      "acc": 0.65610476,
      "epoch": 0.42491121258244546,
      "grad_norm": 7.71875,
      "learning_rate": 9.295645467909942e-06,
      "loss": 1.60306702,
      "memory(GiB)": 97.17,
      "step": 16750,
      "train_speed(iter/s)": 1.634894
    },
    {
      "acc": 0.64639149,
      "epoch": 0.4250380517503805,
      "grad_norm": 5.25,
      "learning_rate": 9.29510873091282e-06,
      "loss": 1.55421333,
      "memory(GiB)": 97.17,
      "step": 16755,
      "train_speed(iter/s)": 1.634946
    },
    {
      "acc": 0.64904027,
      "epoch": 0.42516489091831555,
      "grad_norm": 6.28125,
      "learning_rate": 9.29457180499764e-06,
      "loss": 1.66660004,
      "memory(GiB)": 97.17,
      "step": 16760,
      "train_speed(iter/s)": 1.635001
    },
    {
      "acc": 0.64320278,
      "epoch": 0.42529173008625065,
      "grad_norm": 5.71875,
      "learning_rate": 9.294034690188016e-06,
      "loss": 1.63236961,
      "memory(GiB)": 97.17,
      "step": 16765,
      "train_speed(iter/s)": 1.635054
    },
    {
      "acc": 0.64747686,
      "epoch": 0.4254185692541857,
      "grad_norm": 5.0,
      "learning_rate": 9.293497386507577e-06,
      "loss": 1.66572227,
      "memory(GiB)": 97.17,
      "step": 16770,
      "train_speed(iter/s)": 1.635108
    },
    {
      "acc": 0.66596174,
      "epoch": 0.42554540842212074,
      "grad_norm": 6.28125,
      "learning_rate": 9.292959893979953e-06,
      "loss": 1.60522346,
      "memory(GiB)": 97.17,
      "step": 16775,
      "train_speed(iter/s)": 1.635163
    },
    {
      "acc": 0.65274973,
      "epoch": 0.4256722475900558,
      "grad_norm": 5.25,
      "learning_rate": 9.292422212628786e-06,
      "loss": 1.65038452,
      "memory(GiB)": 97.17,
      "step": 16780,
      "train_speed(iter/s)": 1.635218
    },
    {
      "acc": 0.65924911,
      "epoch": 0.4257990867579909,
      "grad_norm": 5.40625,
      "learning_rate": 9.291884342477728e-06,
      "loss": 1.62604218,
      "memory(GiB)": 97.17,
      "step": 16785,
      "train_speed(iter/s)": 1.635269
    },
    {
      "acc": 0.65498648,
      "epoch": 0.42592592592592593,
      "grad_norm": 5.21875,
      "learning_rate": 9.291346283550433e-06,
      "loss": 1.58802414,
      "memory(GiB)": 97.17,
      "step": 16790,
      "train_speed(iter/s)": 1.635325
    },
    {
      "acc": 0.64843564,
      "epoch": 0.426052765093861,
      "grad_norm": 6.40625,
      "learning_rate": 9.290808035870569e-06,
      "loss": 1.67216606,
      "memory(GiB)": 97.17,
      "step": 16795,
      "train_speed(iter/s)": 1.635381
    },
    {
      "acc": 0.64500399,
      "epoch": 0.426179604261796,
      "grad_norm": 6.09375,
      "learning_rate": 9.29026959946181e-06,
      "loss": 1.68752384,
      "memory(GiB)": 97.17,
      "step": 16800,
      "train_speed(iter/s)": 1.635432
    },
    {
      "acc": 0.6609199,
      "epoch": 0.4263064434297311,
      "grad_norm": 4.875,
      "learning_rate": 9.289730974347841e-06,
      "loss": 1.60022697,
      "memory(GiB)": 97.17,
      "step": 16805,
      "train_speed(iter/s)": 1.635486
    },
    {
      "acc": 0.67437744,
      "epoch": 0.42643328259766616,
      "grad_norm": 5.5625,
      "learning_rate": 9.28919216055235e-06,
      "loss": 1.56150808,
      "memory(GiB)": 97.17,
      "step": 16810,
      "train_speed(iter/s)": 1.635539
    },
    {
      "acc": 0.66691504,
      "epoch": 0.4265601217656012,
      "grad_norm": 8.0625,
      "learning_rate": 9.288653158099038e-06,
      "loss": 1.54448881,
      "memory(GiB)": 97.17,
      "step": 16815,
      "train_speed(iter/s)": 1.635592
    },
    {
      "acc": 0.65150204,
      "epoch": 0.42668696093353625,
      "grad_norm": 5.375,
      "learning_rate": 9.288113967011612e-06,
      "loss": 1.65020218,
      "memory(GiB)": 97.17,
      "step": 16820,
      "train_speed(iter/s)": 1.635642
    },
    {
      "acc": 0.66859961,
      "epoch": 0.42681380010147135,
      "grad_norm": 5.78125,
      "learning_rate": 9.28757458731379e-06,
      "loss": 1.57901583,
      "memory(GiB)": 97.17,
      "step": 16825,
      "train_speed(iter/s)": 1.635694
    },
    {
      "acc": 0.64868131,
      "epoch": 0.4269406392694064,
      "grad_norm": 5.59375,
      "learning_rate": 9.287035019029295e-06,
      "loss": 1.67384911,
      "memory(GiB)": 97.17,
      "step": 16830,
      "train_speed(iter/s)": 1.635746
    },
    {
      "acc": 0.65630088,
      "epoch": 0.42706747843734144,
      "grad_norm": 5.625,
      "learning_rate": 9.286495262181859e-06,
      "loss": 1.53766232,
      "memory(GiB)": 97.17,
      "step": 16835,
      "train_speed(iter/s)": 1.6358
    },
    {
      "acc": 0.64981813,
      "epoch": 0.4271943176052765,
      "grad_norm": 7.4375,
      "learning_rate": 9.285955316795224e-06,
      "loss": 1.66251774,
      "memory(GiB)": 97.17,
      "step": 16840,
      "train_speed(iter/s)": 1.635856
    },
    {
      "acc": 0.65139971,
      "epoch": 0.4273211567732116,
      "grad_norm": 7.84375,
      "learning_rate": 9.285415182893138e-06,
      "loss": 1.6517189,
      "memory(GiB)": 97.17,
      "step": 16845,
      "train_speed(iter/s)": 1.635909
    },
    {
      "acc": 0.65064836,
      "epoch": 0.42744799594114663,
      "grad_norm": 6.4375,
      "learning_rate": 9.28487486049936e-06,
      "loss": 1.62671642,
      "memory(GiB)": 97.17,
      "step": 16850,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.65457125,
      "epoch": 0.4275748351090817,
      "grad_norm": 7.90625,
      "learning_rate": 9.284334349637655e-06,
      "loss": 1.68887558,
      "memory(GiB)": 97.17,
      "step": 16855,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.66219091,
      "epoch": 0.4277016742770167,
      "grad_norm": 5.46875,
      "learning_rate": 9.283793650331798e-06,
      "loss": 1.56025734,
      "memory(GiB)": 97.17,
      "step": 16860,
      "train_speed(iter/s)": 1.636073
    },
    {
      "acc": 0.66432199,
      "epoch": 0.4278285134449518,
      "grad_norm": 5.625,
      "learning_rate": 9.283252762605568e-06,
      "loss": 1.6029665,
      "memory(GiB)": 97.17,
      "step": 16865,
      "train_speed(iter/s)": 1.636124
    },
    {
      "acc": 0.64984684,
      "epoch": 0.42795535261288686,
      "grad_norm": 5.375,
      "learning_rate": 9.28271168648276e-06,
      "loss": 1.64032516,
      "memory(GiB)": 97.17,
      "step": 16870,
      "train_speed(iter/s)": 1.636177
    },
    {
      "acc": 0.65118694,
      "epoch": 0.4280821917808219,
      "grad_norm": 6.9375,
      "learning_rate": 9.282170421987171e-06,
      "loss": 1.65879517,
      "memory(GiB)": 97.17,
      "step": 16875,
      "train_speed(iter/s)": 1.636232
    },
    {
      "acc": 0.64793534,
      "epoch": 0.42820903094875695,
      "grad_norm": 6.15625,
      "learning_rate": 9.281628969142609e-06,
      "loss": 1.73152161,
      "memory(GiB)": 97.17,
      "step": 16880,
      "train_speed(iter/s)": 1.63629
    },
    {
      "acc": 0.6323853,
      "epoch": 0.42833587011669205,
      "grad_norm": 5.90625,
      "learning_rate": 9.281087327972886e-06,
      "loss": 1.70044098,
      "memory(GiB)": 97.17,
      "step": 16885,
      "train_speed(iter/s)": 1.636345
    },
    {
      "acc": 0.65641103,
      "epoch": 0.4284627092846271,
      "grad_norm": 5.875,
      "learning_rate": 9.280545498501832e-06,
      "loss": 1.56100769,
      "memory(GiB)": 97.17,
      "step": 16890,
      "train_speed(iter/s)": 1.636401
    },
    {
      "acc": 0.64588432,
      "epoch": 0.42858954845256214,
      "grad_norm": 5.46875,
      "learning_rate": 9.280003480753274e-06,
      "loss": 1.68182926,
      "memory(GiB)": 97.17,
      "step": 16895,
      "train_speed(iter/s)": 1.636457
    },
    {
      "acc": 0.63259392,
      "epoch": 0.4287163876204972,
      "grad_norm": 5.53125,
      "learning_rate": 9.279461274751054e-06,
      "loss": 1.72251854,
      "memory(GiB)": 97.17,
      "step": 16900,
      "train_speed(iter/s)": 1.636513
    },
    {
      "acc": 0.64388452,
      "epoch": 0.4288432267884323,
      "grad_norm": 5.4375,
      "learning_rate": 9.27891888051902e-06,
      "loss": 1.63054733,
      "memory(GiB)": 97.17,
      "step": 16905,
      "train_speed(iter/s)": 1.636564
    },
    {
      "acc": 0.65210323,
      "epoch": 0.42897006595636733,
      "grad_norm": 5.90625,
      "learning_rate": 9.278376298081032e-06,
      "loss": 1.6496582,
      "memory(GiB)": 97.17,
      "step": 16910,
      "train_speed(iter/s)": 1.636621
    },
    {
      "acc": 0.65076361,
      "epoch": 0.4290969051243024,
      "grad_norm": 5.78125,
      "learning_rate": 9.277833527460952e-06,
      "loss": 1.6364645,
      "memory(GiB)": 97.17,
      "step": 16915,
      "train_speed(iter/s)": 1.636679
    },
    {
      "acc": 0.66106238,
      "epoch": 0.4292237442922374,
      "grad_norm": 5.8125,
      "learning_rate": 9.277290568682653e-06,
      "loss": 1.63757401,
      "memory(GiB)": 97.17,
      "step": 16920,
      "train_speed(iter/s)": 1.636733
    },
    {
      "acc": 0.63203897,
      "epoch": 0.4293505834601725,
      "grad_norm": 5.59375,
      "learning_rate": 9.27674742177002e-06,
      "loss": 1.70032616,
      "memory(GiB)": 97.17,
      "step": 16925,
      "train_speed(iter/s)": 1.636791
    },
    {
      "acc": 0.6439805,
      "epoch": 0.42947742262810756,
      "grad_norm": 5.21875,
      "learning_rate": 9.27620408674694e-06,
      "loss": 1.59769163,
      "memory(GiB)": 97.17,
      "step": 16930,
      "train_speed(iter/s)": 1.636845
    },
    {
      "acc": 0.64007516,
      "epoch": 0.4296042617960426,
      "grad_norm": 10.0625,
      "learning_rate": 9.275660563637313e-06,
      "loss": 1.62902184,
      "memory(GiB)": 97.17,
      "step": 16935,
      "train_speed(iter/s)": 1.636897
    },
    {
      "acc": 0.65972977,
      "epoch": 0.42973110096397765,
      "grad_norm": 5.5625,
      "learning_rate": 9.275116852465043e-06,
      "loss": 1.61088524,
      "memory(GiB)": 97.17,
      "step": 16940,
      "train_speed(iter/s)": 1.636954
    },
    {
      "acc": 0.6521903,
      "epoch": 0.42985794013191275,
      "grad_norm": 6.375,
      "learning_rate": 9.274572953254048e-06,
      "loss": 1.58857517,
      "memory(GiB)": 97.17,
      "step": 16945,
      "train_speed(iter/s)": 1.637008
    },
    {
      "acc": 0.6719595,
      "epoch": 0.4299847792998478,
      "grad_norm": 7.125,
      "learning_rate": 9.27402886602825e-06,
      "loss": 1.49877234,
      "memory(GiB)": 97.17,
      "step": 16950,
      "train_speed(iter/s)": 1.637063
    },
    {
      "acc": 0.64124508,
      "epoch": 0.43011161846778284,
      "grad_norm": 6.125,
      "learning_rate": 9.27348459081158e-06,
      "loss": 1.65098991,
      "memory(GiB)": 97.17,
      "step": 16955,
      "train_speed(iter/s)": 1.637116
    },
    {
      "acc": 0.65731564,
      "epoch": 0.4302384576357179,
      "grad_norm": 5.0,
      "learning_rate": 9.272940127627979e-06,
      "loss": 1.60240383,
      "memory(GiB)": 97.17,
      "step": 16960,
      "train_speed(iter/s)": 1.637172
    },
    {
      "acc": 0.65151868,
      "epoch": 0.430365296803653,
      "grad_norm": 6.53125,
      "learning_rate": 9.272395476501392e-06,
      "loss": 1.6307972,
      "memory(GiB)": 97.17,
      "step": 16965,
      "train_speed(iter/s)": 1.637231
    },
    {
      "acc": 0.63347569,
      "epoch": 0.43049213597158803,
      "grad_norm": 4.59375,
      "learning_rate": 9.27185063745578e-06,
      "loss": 1.71275272,
      "memory(GiB)": 97.17,
      "step": 16970,
      "train_speed(iter/s)": 1.637282
    },
    {
      "acc": 0.65819225,
      "epoch": 0.4306189751395231,
      "grad_norm": 5.3125,
      "learning_rate": 9.271305610515103e-06,
      "loss": 1.58094749,
      "memory(GiB)": 97.17,
      "step": 16975,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.65439162,
      "epoch": 0.4307458143074581,
      "grad_norm": 5.09375,
      "learning_rate": 9.270760395703334e-06,
      "loss": 1.65464859,
      "memory(GiB)": 97.17,
      "step": 16980,
      "train_speed(iter/s)": 1.637385
    },
    {
      "acc": 0.64985428,
      "epoch": 0.4308726534753932,
      "grad_norm": 5.625,
      "learning_rate": 9.270214993044456e-06,
      "loss": 1.67589989,
      "memory(GiB)": 97.17,
      "step": 16985,
      "train_speed(iter/s)": 1.637438
    },
    {
      "acc": 0.64864516,
      "epoch": 0.43099949264332826,
      "grad_norm": 5.53125,
      "learning_rate": 9.269669402562458e-06,
      "loss": 1.62970772,
      "memory(GiB)": 97.17,
      "step": 16990,
      "train_speed(iter/s)": 1.637491
    },
    {
      "acc": 0.66255465,
      "epoch": 0.4311263318112633,
      "grad_norm": 5.59375,
      "learning_rate": 9.269123624281336e-06,
      "loss": 1.60183716,
      "memory(GiB)": 97.17,
      "step": 16995,
      "train_speed(iter/s)": 1.637544
    },
    {
      "acc": 0.67056694,
      "epoch": 0.43125317097919835,
      "grad_norm": 4.96875,
      "learning_rate": 9.268577658225097e-06,
      "loss": 1.50912457,
      "memory(GiB)": 97.17,
      "step": 17000,
      "train_speed(iter/s)": 1.637596
    },
    {
      "epoch": 0.43125317097919835,
      "eval_acc": 0.6428900485948832,
      "eval_loss": 1.5926954746246338,
      "eval_runtime": 58.4038,
      "eval_samples_per_second": 109.068,
      "eval_steps_per_second": 27.276,
      "step": 17000
    },
    {
      "acc": 0.66105733,
      "epoch": 0.43138001014713345,
      "grad_norm": 5.96875,
      "learning_rate": 9.268031504417756e-06,
      "loss": 1.60161285,
      "memory(GiB)": 97.17,
      "step": 17005,
      "train_speed(iter/s)": 1.627824
    },
    {
      "acc": 0.65026212,
      "epoch": 0.4315068493150685,
      "grad_norm": 6.59375,
      "learning_rate": 9.267485162883334e-06,
      "loss": 1.62716045,
      "memory(GiB)": 97.17,
      "step": 17010,
      "train_speed(iter/s)": 1.627868
    },
    {
      "acc": 0.63613806,
      "epoch": 0.43163368848300354,
      "grad_norm": 6.0,
      "learning_rate": 9.266938633645861e-06,
      "loss": 1.71146259,
      "memory(GiB)": 97.17,
      "step": 17015,
      "train_speed(iter/s)": 1.627925
    },
    {
      "acc": 0.66908288,
      "epoch": 0.4317605276509386,
      "grad_norm": 8.0625,
      "learning_rate": 9.266391916729376e-06,
      "loss": 1.5233408,
      "memory(GiB)": 97.17,
      "step": 17020,
      "train_speed(iter/s)": 1.62798
    },
    {
      "acc": 0.64691744,
      "epoch": 0.4318873668188737,
      "grad_norm": 5.59375,
      "learning_rate": 9.265845012157926e-06,
      "loss": 1.65524063,
      "memory(GiB)": 97.17,
      "step": 17025,
      "train_speed(iter/s)": 1.628033
    },
    {
      "acc": 0.64269838,
      "epoch": 0.43201420598680873,
      "grad_norm": 6.6875,
      "learning_rate": 9.265297919955566e-06,
      "loss": 1.61516838,
      "memory(GiB)": 97.17,
      "step": 17030,
      "train_speed(iter/s)": 1.628084
    },
    {
      "acc": 0.64917049,
      "epoch": 0.4321410451547438,
      "grad_norm": 5.875,
      "learning_rate": 9.264750640146363e-06,
      "loss": 1.64190254,
      "memory(GiB)": 97.17,
      "step": 17035,
      "train_speed(iter/s)": 1.628141
    },
    {
      "acc": 0.65181785,
      "epoch": 0.4322678843226788,
      "grad_norm": 4.375,
      "learning_rate": 9.264203172754384e-06,
      "loss": 1.61965599,
      "memory(GiB)": 97.17,
      "step": 17040,
      "train_speed(iter/s)": 1.628193
    },
    {
      "acc": 0.65951385,
      "epoch": 0.4323947234906139,
      "grad_norm": 5.875,
      "learning_rate": 9.263655517803713e-06,
      "loss": 1.64219055,
      "memory(GiB)": 97.17,
      "step": 17045,
      "train_speed(iter/s)": 1.628244
    },
    {
      "acc": 0.64553347,
      "epoch": 0.43252156265854896,
      "grad_norm": 4.8125,
      "learning_rate": 9.263107675318434e-06,
      "loss": 1.68419914,
      "memory(GiB)": 97.17,
      "step": 17050,
      "train_speed(iter/s)": 1.628294
    },
    {
      "acc": 0.66039534,
      "epoch": 0.432648401826484,
      "grad_norm": 5.125,
      "learning_rate": 9.262559645322648e-06,
      "loss": 1.5950983,
      "memory(GiB)": 97.17,
      "step": 17055,
      "train_speed(iter/s)": 1.628347
    },
    {
      "acc": 0.65618653,
      "epoch": 0.43277524099441905,
      "grad_norm": 5.78125,
      "learning_rate": 9.262011427840459e-06,
      "loss": 1.61158066,
      "memory(GiB)": 97.17,
      "step": 17060,
      "train_speed(iter/s)": 1.628398
    },
    {
      "acc": 0.66023817,
      "epoch": 0.43290208016235415,
      "grad_norm": 6.375,
      "learning_rate": 9.261463022895976e-06,
      "loss": 1.5364913,
      "memory(GiB)": 97.17,
      "step": 17065,
      "train_speed(iter/s)": 1.628454
    },
    {
      "acc": 0.66091561,
      "epoch": 0.4330289193302892,
      "grad_norm": 6.125,
      "learning_rate": 9.260914430513325e-06,
      "loss": 1.6025919,
      "memory(GiB)": 97.17,
      "step": 17070,
      "train_speed(iter/s)": 1.628506
    },
    {
      "acc": 0.63900504,
      "epoch": 0.43315575849822424,
      "grad_norm": 5.4375,
      "learning_rate": 9.260365650716632e-06,
      "loss": 1.6923912,
      "memory(GiB)": 97.17,
      "step": 17075,
      "train_speed(iter/s)": 1.628559
    },
    {
      "acc": 0.65349422,
      "epoch": 0.4332825976661593,
      "grad_norm": 6.1875,
      "learning_rate": 9.259816683530038e-06,
      "loss": 1.65538883,
      "memory(GiB)": 97.17,
      "step": 17080,
      "train_speed(iter/s)": 1.628612
    },
    {
      "acc": 0.66349411,
      "epoch": 0.4334094368340944,
      "grad_norm": 5.78125,
      "learning_rate": 9.259267528977687e-06,
      "loss": 1.5489666,
      "memory(GiB)": 97.17,
      "step": 17085,
      "train_speed(iter/s)": 1.628666
    },
    {
      "acc": 0.65586414,
      "epoch": 0.43353627600202943,
      "grad_norm": 5.53125,
      "learning_rate": 9.258718187083734e-06,
      "loss": 1.61791191,
      "memory(GiB)": 97.17,
      "step": 17090,
      "train_speed(iter/s)": 1.628716
    },
    {
      "acc": 0.660045,
      "epoch": 0.4336631151699645,
      "grad_norm": 6.125,
      "learning_rate": 9.258168657872341e-06,
      "loss": 1.59707642,
      "memory(GiB)": 97.17,
      "step": 17095,
      "train_speed(iter/s)": 1.628769
    },
    {
      "acc": 0.65517187,
      "epoch": 0.4337899543378995,
      "grad_norm": 5.6875,
      "learning_rate": 9.25761894136768e-06,
      "loss": 1.65323601,
      "memory(GiB)": 97.17,
      "step": 17100,
      "train_speed(iter/s)": 1.628824
    },
    {
      "acc": 0.64673848,
      "epoch": 0.4339167935058346,
      "grad_norm": 4.6875,
      "learning_rate": 9.25706903759393e-06,
      "loss": 1.64696388,
      "memory(GiB)": 97.17,
      "step": 17105,
      "train_speed(iter/s)": 1.628881
    },
    {
      "acc": 0.6620564,
      "epoch": 0.43404363267376966,
      "grad_norm": 5.84375,
      "learning_rate": 9.256518946575274e-06,
      "loss": 1.57205925,
      "memory(GiB)": 97.17,
      "step": 17110,
      "train_speed(iter/s)": 1.628932
    },
    {
      "acc": 0.65434246,
      "epoch": 0.4341704718417047,
      "grad_norm": 7.0,
      "learning_rate": 9.255968668335912e-06,
      "loss": 1.65409088,
      "memory(GiB)": 97.17,
      "step": 17115,
      "train_speed(iter/s)": 1.628986
    },
    {
      "acc": 0.65264697,
      "epoch": 0.43429731100963975,
      "grad_norm": 7.59375,
      "learning_rate": 9.255418202900048e-06,
      "loss": 1.59731989,
      "memory(GiB)": 97.17,
      "step": 17120,
      "train_speed(iter/s)": 1.629038
    },
    {
      "acc": 0.64172454,
      "epoch": 0.43442415017757485,
      "grad_norm": 5.34375,
      "learning_rate": 9.254867550291891e-06,
      "loss": 1.65201607,
      "memory(GiB)": 97.17,
      "step": 17125,
      "train_speed(iter/s)": 1.629095
    },
    {
      "acc": 0.65509071,
      "epoch": 0.4345509893455099,
      "grad_norm": 5.71875,
      "learning_rate": 9.254316710535662e-06,
      "loss": 1.55219383,
      "memory(GiB)": 97.17,
      "step": 17130,
      "train_speed(iter/s)": 1.629151
    },
    {
      "acc": 0.6626318,
      "epoch": 0.43467782851344494,
      "grad_norm": 6.65625,
      "learning_rate": 9.253765683655591e-06,
      "loss": 1.61080475,
      "memory(GiB)": 97.17,
      "step": 17135,
      "train_speed(iter/s)": 1.629193
    },
    {
      "acc": 0.6753911,
      "epoch": 0.43480466768138,
      "grad_norm": 6.65625,
      "learning_rate": 9.253214469675913e-06,
      "loss": 1.54352493,
      "memory(GiB)": 97.17,
      "step": 17140,
      "train_speed(iter/s)": 1.629248
    },
    {
      "acc": 0.66376657,
      "epoch": 0.4349315068493151,
      "grad_norm": 5.84375,
      "learning_rate": 9.252663068620874e-06,
      "loss": 1.59380894,
      "memory(GiB)": 97.17,
      "step": 17145,
      "train_speed(iter/s)": 1.629304
    },
    {
      "acc": 0.64371247,
      "epoch": 0.43505834601725013,
      "grad_norm": 4.78125,
      "learning_rate": 9.252111480514726e-06,
      "loss": 1.66468639,
      "memory(GiB)": 97.17,
      "step": 17150,
      "train_speed(iter/s)": 1.629355
    },
    {
      "acc": 0.65283666,
      "epoch": 0.4351851851851852,
      "grad_norm": 5.9375,
      "learning_rate": 9.251559705381731e-06,
      "loss": 1.66845894,
      "memory(GiB)": 97.17,
      "step": 17155,
      "train_speed(iter/s)": 1.629409
    },
    {
      "acc": 0.62987647,
      "epoch": 0.4353120243531202,
      "grad_norm": 6.625,
      "learning_rate": 9.251007743246159e-06,
      "loss": 1.68390102,
      "memory(GiB)": 97.17,
      "step": 17160,
      "train_speed(iter/s)": 1.629466
    },
    {
      "acc": 0.66484098,
      "epoch": 0.4354388635210553,
      "grad_norm": 7.375,
      "learning_rate": 9.250455594132286e-06,
      "loss": 1.6065155,
      "memory(GiB)": 97.17,
      "step": 17165,
      "train_speed(iter/s)": 1.629522
    },
    {
      "acc": 0.65618491,
      "epoch": 0.43556570268899036,
      "grad_norm": 6.0,
      "learning_rate": 9.249903258064399e-06,
      "loss": 1.66336746,
      "memory(GiB)": 97.17,
      "step": 17170,
      "train_speed(iter/s)": 1.629576
    },
    {
      "acc": 0.66518464,
      "epoch": 0.4356925418569254,
      "grad_norm": 5.65625,
      "learning_rate": 9.249350735066792e-06,
      "loss": 1.55276699,
      "memory(GiB)": 97.17,
      "step": 17175,
      "train_speed(iter/s)": 1.629631
    },
    {
      "acc": 0.63644385,
      "epoch": 0.43581938102486045,
      "grad_norm": 5.875,
      "learning_rate": 9.248798025163768e-06,
      "loss": 1.63725357,
      "memory(GiB)": 97.17,
      "step": 17180,
      "train_speed(iter/s)": 1.629687
    },
    {
      "acc": 0.66316242,
      "epoch": 0.43594622019279555,
      "grad_norm": 7.9375,
      "learning_rate": 9.248245128379638e-06,
      "loss": 1.56775265,
      "memory(GiB)": 97.17,
      "step": 17185,
      "train_speed(iter/s)": 1.629742
    },
    {
      "acc": 0.65070972,
      "epoch": 0.4360730593607306,
      "grad_norm": 5.4375,
      "learning_rate": 9.24769204473872e-06,
      "loss": 1.60765514,
      "memory(GiB)": 97.17,
      "step": 17190,
      "train_speed(iter/s)": 1.629794
    },
    {
      "acc": 0.65628014,
      "epoch": 0.43619989852866564,
      "grad_norm": 5.3125,
      "learning_rate": 9.24713877426534e-06,
      "loss": 1.58805046,
      "memory(GiB)": 97.17,
      "step": 17195,
      "train_speed(iter/s)": 1.629845
    },
    {
      "acc": 0.66126528,
      "epoch": 0.4363267376966007,
      "grad_norm": 5.09375,
      "learning_rate": 9.246585316983837e-06,
      "loss": 1.61996746,
      "memory(GiB)": 97.17,
      "step": 17200,
      "train_speed(iter/s)": 1.629898
    },
    {
      "acc": 0.63899622,
      "epoch": 0.4364535768645358,
      "grad_norm": 5.0625,
      "learning_rate": 9.24603167291855e-06,
      "loss": 1.69389324,
      "memory(GiB)": 97.17,
      "step": 17205,
      "train_speed(iter/s)": 1.629951
    },
    {
      "acc": 0.6473176,
      "epoch": 0.43658041603247083,
      "grad_norm": 5.65625,
      "learning_rate": 9.245477842093832e-06,
      "loss": 1.62571335,
      "memory(GiB)": 97.17,
      "step": 17210,
      "train_speed(iter/s)": 1.630003
    },
    {
      "acc": 0.64487653,
      "epoch": 0.4367072552004059,
      "grad_norm": 5.8125,
      "learning_rate": 9.244923824534046e-06,
      "loss": 1.62782249,
      "memory(GiB)": 97.17,
      "step": 17215,
      "train_speed(iter/s)": 1.630059
    },
    {
      "acc": 0.64285207,
      "epoch": 0.4368340943683409,
      "grad_norm": 6.34375,
      "learning_rate": 9.244369620263558e-06,
      "loss": 1.64927273,
      "memory(GiB)": 97.17,
      "step": 17220,
      "train_speed(iter/s)": 1.630113
    },
    {
      "acc": 0.65320253,
      "epoch": 0.436960933536276,
      "grad_norm": 5.78125,
      "learning_rate": 9.243815229306746e-06,
      "loss": 1.57669182,
      "memory(GiB)": 97.17,
      "step": 17225,
      "train_speed(iter/s)": 1.630164
    },
    {
      "acc": 0.64941335,
      "epoch": 0.43708777270421106,
      "grad_norm": 6.34375,
      "learning_rate": 9.243260651687989e-06,
      "loss": 1.64794197,
      "memory(GiB)": 97.17,
      "step": 17230,
      "train_speed(iter/s)": 1.630217
    },
    {
      "acc": 0.64359121,
      "epoch": 0.4372146118721461,
      "grad_norm": 5.34375,
      "learning_rate": 9.242705887431685e-06,
      "loss": 1.66660576,
      "memory(GiB)": 97.17,
      "step": 17235,
      "train_speed(iter/s)": 1.630271
    },
    {
      "acc": 0.65126886,
      "epoch": 0.43734145104008115,
      "grad_norm": 6.78125,
      "learning_rate": 9.242150936562235e-06,
      "loss": 1.6832901,
      "memory(GiB)": 97.17,
      "step": 17240,
      "train_speed(iter/s)": 1.630324
    },
    {
      "acc": 0.65410557,
      "epoch": 0.43746829020801625,
      "grad_norm": 5.0625,
      "learning_rate": 9.241595799104046e-06,
      "loss": 1.58133001,
      "memory(GiB)": 97.17,
      "step": 17245,
      "train_speed(iter/s)": 1.630376
    },
    {
      "acc": 0.65065088,
      "epoch": 0.4375951293759513,
      "grad_norm": 5.3125,
      "learning_rate": 9.241040475081537e-06,
      "loss": 1.63557606,
      "memory(GiB)": 97.17,
      "step": 17250,
      "train_speed(iter/s)": 1.630427
    },
    {
      "acc": 0.64462276,
      "epoch": 0.43772196854388634,
      "grad_norm": 5.28125,
      "learning_rate": 9.240484964519131e-06,
      "loss": 1.65170746,
      "memory(GiB)": 97.17,
      "step": 17255,
      "train_speed(iter/s)": 1.630479
    },
    {
      "acc": 0.65563655,
      "epoch": 0.4378488077118214,
      "grad_norm": 6.625,
      "learning_rate": 9.239929267441267e-06,
      "loss": 1.59051075,
      "memory(GiB)": 97.17,
      "step": 17260,
      "train_speed(iter/s)": 1.630528
    },
    {
      "acc": 0.66627102,
      "epoch": 0.4379756468797565,
      "grad_norm": 6.03125,
      "learning_rate": 9.239373383872382e-06,
      "loss": 1.6809948,
      "memory(GiB)": 97.17,
      "step": 17265,
      "train_speed(iter/s)": 1.630583
    },
    {
      "acc": 0.63808298,
      "epoch": 0.43810248604769153,
      "grad_norm": 5.0,
      "learning_rate": 9.238817313836927e-06,
      "loss": 1.59716835,
      "memory(GiB)": 97.17,
      "step": 17270,
      "train_speed(iter/s)": 1.630633
    },
    {
      "acc": 0.65963092,
      "epoch": 0.4382293252156266,
      "grad_norm": 6.53125,
      "learning_rate": 9.238261057359365e-06,
      "loss": 1.59871511,
      "memory(GiB)": 97.17,
      "step": 17275,
      "train_speed(iter/s)": 1.630685
    },
    {
      "acc": 0.6534812,
      "epoch": 0.4383561643835616,
      "grad_norm": 5.15625,
      "learning_rate": 9.237704614464157e-06,
      "loss": 1.63023701,
      "memory(GiB)": 97.17,
      "step": 17280,
      "train_speed(iter/s)": 1.630739
    },
    {
      "acc": 0.64379129,
      "epoch": 0.4384830035514967,
      "grad_norm": 4.96875,
      "learning_rate": 9.237147985175781e-06,
      "loss": 1.67299042,
      "memory(GiB)": 97.17,
      "step": 17285,
      "train_speed(iter/s)": 1.630792
    },
    {
      "acc": 0.65906954,
      "epoch": 0.43860984271943176,
      "grad_norm": 4.96875,
      "learning_rate": 9.236591169518717e-06,
      "loss": 1.61831512,
      "memory(GiB)": 97.17,
      "step": 17290,
      "train_speed(iter/s)": 1.630841
    },
    {
      "acc": 0.65493207,
      "epoch": 0.4387366818873668,
      "grad_norm": 5.75,
      "learning_rate": 9.236034167517461e-06,
      "loss": 1.63975716,
      "memory(GiB)": 97.17,
      "step": 17295,
      "train_speed(iter/s)": 1.630893
    },
    {
      "acc": 0.65858755,
      "epoch": 0.43886352105530185,
      "grad_norm": 7.75,
      "learning_rate": 9.235476979196507e-06,
      "loss": 1.61957054,
      "memory(GiB)": 97.17,
      "step": 17300,
      "train_speed(iter/s)": 1.630945
    },
    {
      "acc": 0.66521482,
      "epoch": 0.43899036022323695,
      "grad_norm": 5.78125,
      "learning_rate": 9.234919604580368e-06,
      "loss": 1.62613564,
      "memory(GiB)": 97.17,
      "step": 17305,
      "train_speed(iter/s)": 1.630997
    },
    {
      "acc": 0.66867867,
      "epoch": 0.439117199391172,
      "grad_norm": 5.28125,
      "learning_rate": 9.234362043693556e-06,
      "loss": 1.54498463,
      "memory(GiB)": 97.17,
      "step": 17310,
      "train_speed(iter/s)": 1.631051
    },
    {
      "acc": 0.64163628,
      "epoch": 0.43924403855910704,
      "grad_norm": 5.03125,
      "learning_rate": 9.233804296560596e-06,
      "loss": 1.63727856,
      "memory(GiB)": 97.17,
      "step": 17315,
      "train_speed(iter/s)": 1.631102
    },
    {
      "acc": 0.63664804,
      "epoch": 0.4393708777270421,
      "grad_norm": 5.875,
      "learning_rate": 9.233246363206021e-06,
      "loss": 1.70027809,
      "memory(GiB)": 97.17,
      "step": 17320,
      "train_speed(iter/s)": 1.631157
    },
    {
      "acc": 0.6491395,
      "epoch": 0.4394977168949772,
      "grad_norm": 5.53125,
      "learning_rate": 9.232688243654371e-06,
      "loss": 1.68130474,
      "memory(GiB)": 97.17,
      "step": 17325,
      "train_speed(iter/s)": 1.631211
    },
    {
      "acc": 0.64025698,
      "epoch": 0.43962455606291223,
      "grad_norm": 5.53125,
      "learning_rate": 9.232129937930194e-06,
      "loss": 1.6935976,
      "memory(GiB)": 97.17,
      "step": 17330,
      "train_speed(iter/s)": 1.631263
    },
    {
      "acc": 0.65890064,
      "epoch": 0.4397513952308473,
      "grad_norm": 4.96875,
      "learning_rate": 9.231571446058047e-06,
      "loss": 1.66459961,
      "memory(GiB)": 97.17,
      "step": 17335,
      "train_speed(iter/s)": 1.631316
    },
    {
      "acc": 0.65500379,
      "epoch": 0.4398782343987823,
      "grad_norm": 4.84375,
      "learning_rate": 9.231012768062497e-06,
      "loss": 1.61756439,
      "memory(GiB)": 97.17,
      "step": 17340,
      "train_speed(iter/s)": 1.631368
    },
    {
      "acc": 0.65212064,
      "epoch": 0.4400050735667174,
      "grad_norm": 5.0625,
      "learning_rate": 9.230453903968112e-06,
      "loss": 1.64284668,
      "memory(GiB)": 97.17,
      "step": 17345,
      "train_speed(iter/s)": 1.63142
    },
    {
      "acc": 0.66191378,
      "epoch": 0.44013191273465246,
      "grad_norm": 6.75,
      "learning_rate": 9.22989485379948e-06,
      "loss": 1.58716145,
      "memory(GiB)": 97.17,
      "step": 17350,
      "train_speed(iter/s)": 1.631473
    },
    {
      "acc": 0.65656586,
      "epoch": 0.4402587519025875,
      "grad_norm": 5.15625,
      "learning_rate": 9.229335617581187e-06,
      "loss": 1.52349739,
      "memory(GiB)": 97.17,
      "step": 17355,
      "train_speed(iter/s)": 1.631523
    },
    {
      "acc": 0.6607399,
      "epoch": 0.44038559107052255,
      "grad_norm": 5.09375,
      "learning_rate": 9.22877619533783e-06,
      "loss": 1.58022556,
      "memory(GiB)": 97.17,
      "step": 17360,
      "train_speed(iter/s)": 1.631576
    },
    {
      "acc": 0.65099888,
      "epoch": 0.44051243023845765,
      "grad_norm": 4.78125,
      "learning_rate": 9.228216587094014e-06,
      "loss": 1.59342184,
      "memory(GiB)": 97.17,
      "step": 17365,
      "train_speed(iter/s)": 1.631625
    },
    {
      "acc": 0.66549187,
      "epoch": 0.4406392694063927,
      "grad_norm": 7.0625,
      "learning_rate": 9.227656792874358e-06,
      "loss": 1.53481331,
      "memory(GiB)": 97.17,
      "step": 17370,
      "train_speed(iter/s)": 1.631675
    },
    {
      "acc": 0.65343962,
      "epoch": 0.44076610857432774,
      "grad_norm": 6.0,
      "learning_rate": 9.227096812703479e-06,
      "loss": 1.65888443,
      "memory(GiB)": 97.17,
      "step": 17375,
      "train_speed(iter/s)": 1.631723
    },
    {
      "acc": 0.65020285,
      "epoch": 0.4408929477422628,
      "grad_norm": 6.53125,
      "learning_rate": 9.22653664660601e-06,
      "loss": 1.62792549,
      "memory(GiB)": 97.17,
      "step": 17380,
      "train_speed(iter/s)": 1.631774
    },
    {
      "acc": 0.6598793,
      "epoch": 0.4410197869101979,
      "grad_norm": 4.9375,
      "learning_rate": 9.225976294606589e-06,
      "loss": 1.62354546,
      "memory(GiB)": 97.17,
      "step": 17385,
      "train_speed(iter/s)": 1.631824
    },
    {
      "acc": 0.64714398,
      "epoch": 0.44114662607813293,
      "grad_norm": 6.78125,
      "learning_rate": 9.225415756729863e-06,
      "loss": 1.62764759,
      "memory(GiB)": 97.17,
      "step": 17390,
      "train_speed(iter/s)": 1.631874
    },
    {
      "acc": 0.63898187,
      "epoch": 0.441273465246068,
      "grad_norm": 6.4375,
      "learning_rate": 9.224855033000489e-06,
      "loss": 1.64272556,
      "memory(GiB)": 97.17,
      "step": 17395,
      "train_speed(iter/s)": 1.631924
    },
    {
      "acc": 0.66961508,
      "epoch": 0.441400304414003,
      "grad_norm": 4.96875,
      "learning_rate": 9.224294123443125e-06,
      "loss": 1.55725889,
      "memory(GiB)": 97.17,
      "step": 17400,
      "train_speed(iter/s)": 1.631965
    },
    {
      "acc": 0.66209025,
      "epoch": 0.4415271435819381,
      "grad_norm": 5.09375,
      "learning_rate": 9.223733028082447e-06,
      "loss": 1.58789091,
      "memory(GiB)": 97.17,
      "step": 17405,
      "train_speed(iter/s)": 1.632016
    },
    {
      "acc": 0.65001025,
      "epoch": 0.44165398274987316,
      "grad_norm": 5.84375,
      "learning_rate": 9.223171746943132e-06,
      "loss": 1.61291885,
      "memory(GiB)": 97.17,
      "step": 17410,
      "train_speed(iter/s)": 1.632067
    },
    {
      "acc": 0.66293631,
      "epoch": 0.4417808219178082,
      "grad_norm": 5.375,
      "learning_rate": 9.222610280049868e-06,
      "loss": 1.65044289,
      "memory(GiB)": 97.17,
      "step": 17415,
      "train_speed(iter/s)": 1.632117
    },
    {
      "acc": 0.64517002,
      "epoch": 0.44190766108574325,
      "grad_norm": 5.84375,
      "learning_rate": 9.222048627427352e-06,
      "loss": 1.67645073,
      "memory(GiB)": 97.17,
      "step": 17420,
      "train_speed(iter/s)": 1.63217
    },
    {
      "acc": 0.6596745,
      "epoch": 0.44203450025367835,
      "grad_norm": 5.59375,
      "learning_rate": 9.221486789100288e-06,
      "loss": 1.60908546,
      "memory(GiB)": 97.17,
      "step": 17425,
      "train_speed(iter/s)": 1.632218
    },
    {
      "acc": 0.64544168,
      "epoch": 0.4421613394216134,
      "grad_norm": 6.65625,
      "learning_rate": 9.220924765093386e-06,
      "loss": 1.62117519,
      "memory(GiB)": 97.17,
      "step": 17430,
      "train_speed(iter/s)": 1.632268
    },
    {
      "acc": 0.63961639,
      "epoch": 0.44228817858954844,
      "grad_norm": 5.0,
      "learning_rate": 9.220362555431369e-06,
      "loss": 1.6463419,
      "memory(GiB)": 97.17,
      "step": 17435,
      "train_speed(iter/s)": 1.632321
    },
    {
      "acc": 0.64414148,
      "epoch": 0.4424150177574835,
      "grad_norm": 5.03125,
      "learning_rate": 9.219800160138964e-06,
      "loss": 1.58510227,
      "memory(GiB)": 97.17,
      "step": 17440,
      "train_speed(iter/s)": 1.632374
    },
    {
      "acc": 0.64963026,
      "epoch": 0.4425418569254186,
      "grad_norm": 6.28125,
      "learning_rate": 9.219237579240907e-06,
      "loss": 1.61883717,
      "memory(GiB)": 97.17,
      "step": 17445,
      "train_speed(iter/s)": 1.632424
    },
    {
      "acc": 0.6608654,
      "epoch": 0.44266869609335363,
      "grad_norm": 5.5,
      "learning_rate": 9.218674812761946e-06,
      "loss": 1.58943939,
      "memory(GiB)": 97.17,
      "step": 17450,
      "train_speed(iter/s)": 1.632472
    },
    {
      "acc": 0.64905262,
      "epoch": 0.4427955352612887,
      "grad_norm": 6.03125,
      "learning_rate": 9.21811186072683e-06,
      "loss": 1.59389238,
      "memory(GiB)": 97.17,
      "step": 17455,
      "train_speed(iter/s)": 1.632521
    },
    {
      "acc": 0.66097002,
      "epoch": 0.4429223744292237,
      "grad_norm": 6.0625,
      "learning_rate": 9.21754872316032e-06,
      "loss": 1.61188049,
      "memory(GiB)": 97.17,
      "step": 17460,
      "train_speed(iter/s)": 1.63257
    },
    {
      "acc": 0.67582827,
      "epoch": 0.4430492135971588,
      "grad_norm": 6.46875,
      "learning_rate": 9.21698540008719e-06,
      "loss": 1.54886074,
      "memory(GiB)": 97.17,
      "step": 17465,
      "train_speed(iter/s)": 1.63262
    },
    {
      "acc": 0.64947634,
      "epoch": 0.44317605276509386,
      "grad_norm": 6.03125,
      "learning_rate": 9.216421891532214e-06,
      "loss": 1.61587067,
      "memory(GiB)": 97.17,
      "step": 17470,
      "train_speed(iter/s)": 1.632669
    },
    {
      "acc": 0.66888046,
      "epoch": 0.4433028919330289,
      "grad_norm": 6.46875,
      "learning_rate": 9.215858197520178e-06,
      "loss": 1.6213028,
      "memory(GiB)": 97.17,
      "step": 17475,
      "train_speed(iter/s)": 1.632719
    },
    {
      "acc": 0.65724154,
      "epoch": 0.44342973110096395,
      "grad_norm": 5.71875,
      "learning_rate": 9.215294318075876e-06,
      "loss": 1.62744942,
      "memory(GiB)": 97.17,
      "step": 17480,
      "train_speed(iter/s)": 1.632774
    },
    {
      "acc": 0.66109142,
      "epoch": 0.44355657026889905,
      "grad_norm": 7.71875,
      "learning_rate": 9.21473025322411e-06,
      "loss": 1.55315933,
      "memory(GiB)": 97.17,
      "step": 17485,
      "train_speed(iter/s)": 1.632825
    },
    {
      "acc": 0.65149584,
      "epoch": 0.4436834094368341,
      "grad_norm": 5.0625,
      "learning_rate": 9.21416600298969e-06,
      "loss": 1.65997868,
      "memory(GiB)": 97.17,
      "step": 17490,
      "train_speed(iter/s)": 1.632877
    },
    {
      "acc": 0.64868312,
      "epoch": 0.44381024860476914,
      "grad_norm": 5.1875,
      "learning_rate": 9.213601567397434e-06,
      "loss": 1.55312653,
      "memory(GiB)": 97.17,
      "step": 17495,
      "train_speed(iter/s)": 1.632927
    },
    {
      "acc": 0.64261456,
      "epoch": 0.4439370877727042,
      "grad_norm": 6.8125,
      "learning_rate": 9.213036946472169e-06,
      "loss": 1.70766525,
      "memory(GiB)": 97.17,
      "step": 17500,
      "train_speed(iter/s)": 1.632976
    },
    {
      "acc": 0.63818736,
      "epoch": 0.4440639269406393,
      "grad_norm": 5.625,
      "learning_rate": 9.212472140238729e-06,
      "loss": 1.63124352,
      "memory(GiB)": 97.17,
      "step": 17505,
      "train_speed(iter/s)": 1.633028
    },
    {
      "acc": 0.6509151,
      "epoch": 0.44419076610857433,
      "grad_norm": 6.25,
      "learning_rate": 9.211907148721958e-06,
      "loss": 1.62345543,
      "memory(GiB)": 97.17,
      "step": 17510,
      "train_speed(iter/s)": 1.63308
    },
    {
      "acc": 0.65588999,
      "epoch": 0.4443176052765094,
      "grad_norm": 5.0,
      "learning_rate": 9.211341971946705e-06,
      "loss": 1.62372894,
      "memory(GiB)": 97.17,
      "step": 17515,
      "train_speed(iter/s)": 1.633127
    },
    {
      "acc": 0.66911678,
      "epoch": 0.4444444444444444,
      "grad_norm": 5.15625,
      "learning_rate": 9.21077660993783e-06,
      "loss": 1.605439,
      "memory(GiB)": 97.17,
      "step": 17520,
      "train_speed(iter/s)": 1.633178
    },
    {
      "acc": 0.65478215,
      "epoch": 0.4445712836123795,
      "grad_norm": 6.53125,
      "learning_rate": 9.210211062720198e-06,
      "loss": 1.64492722,
      "memory(GiB)": 97.17,
      "step": 17525,
      "train_speed(iter/s)": 1.633225
    },
    {
      "acc": 0.63903246,
      "epoch": 0.44469812278031456,
      "grad_norm": 4.875,
      "learning_rate": 9.209645330318689e-06,
      "loss": 1.68473778,
      "memory(GiB)": 97.17,
      "step": 17530,
      "train_speed(iter/s)": 1.633277
    },
    {
      "acc": 0.65429549,
      "epoch": 0.4448249619482496,
      "grad_norm": 6.875,
      "learning_rate": 9.209079412758183e-06,
      "loss": 1.6484724,
      "memory(GiB)": 97.17,
      "step": 17535,
      "train_speed(iter/s)": 1.633324
    },
    {
      "acc": 0.65037231,
      "epoch": 0.44495180111618465,
      "grad_norm": 7.9375,
      "learning_rate": 9.208513310063572e-06,
      "loss": 1.65741959,
      "memory(GiB)": 97.17,
      "step": 17540,
      "train_speed(iter/s)": 1.633375
    },
    {
      "acc": 0.66384783,
      "epoch": 0.44507864028411975,
      "grad_norm": 6.5625,
      "learning_rate": 9.207947022259755e-06,
      "loss": 1.6025013,
      "memory(GiB)": 97.17,
      "step": 17545,
      "train_speed(iter/s)": 1.633426
    },
    {
      "acc": 0.6584012,
      "epoch": 0.4452054794520548,
      "grad_norm": 6.90625,
      "learning_rate": 9.207380549371642e-06,
      "loss": 1.56964865,
      "memory(GiB)": 97.17,
      "step": 17550,
      "train_speed(iter/s)": 1.633474
    },
    {
      "acc": 0.64631844,
      "epoch": 0.44533231861998984,
      "grad_norm": 6.40625,
      "learning_rate": 9.206813891424147e-06,
      "loss": 1.65718536,
      "memory(GiB)": 97.17,
      "step": 17555,
      "train_speed(iter/s)": 1.63352
    },
    {
      "acc": 0.64426107,
      "epoch": 0.4454591577879249,
      "grad_norm": 5.3125,
      "learning_rate": 9.206247048442196e-06,
      "loss": 1.68766079,
      "memory(GiB)": 97.17,
      "step": 17560,
      "train_speed(iter/s)": 1.633571
    },
    {
      "acc": 0.65517244,
      "epoch": 0.44558599695586,
      "grad_norm": 5.1875,
      "learning_rate": 9.20568002045072e-06,
      "loss": 1.61405602,
      "memory(GiB)": 97.17,
      "step": 17565,
      "train_speed(iter/s)": 1.633623
    },
    {
      "acc": 0.64193468,
      "epoch": 0.44571283612379503,
      "grad_norm": 4.6875,
      "learning_rate": 9.20511280747466e-06,
      "loss": 1.63794937,
      "memory(GiB)": 97.17,
      "step": 17570,
      "train_speed(iter/s)": 1.633672
    },
    {
      "acc": 0.65987406,
      "epoch": 0.4458396752917301,
      "grad_norm": 4.9375,
      "learning_rate": 9.204545409538962e-06,
      "loss": 1.55015793,
      "memory(GiB)": 97.17,
      "step": 17575,
      "train_speed(iter/s)": 1.633724
    },
    {
      "acc": 0.64122486,
      "epoch": 0.4459665144596651,
      "grad_norm": 5.03125,
      "learning_rate": 9.203977826668587e-06,
      "loss": 1.72442589,
      "memory(GiB)": 97.17,
      "step": 17580,
      "train_speed(iter/s)": 1.633774
    },
    {
      "acc": 0.66856356,
      "epoch": 0.4460933536276002,
      "grad_norm": 5.34375,
      "learning_rate": 9.203410058888498e-06,
      "loss": 1.54310093,
      "memory(GiB)": 97.17,
      "step": 17585,
      "train_speed(iter/s)": 1.633822
    },
    {
      "acc": 0.6521203,
      "epoch": 0.44622019279553526,
      "grad_norm": 5.71875,
      "learning_rate": 9.202842106223667e-06,
      "loss": 1.62952347,
      "memory(GiB)": 97.17,
      "step": 17590,
      "train_speed(iter/s)": 1.63387
    },
    {
      "acc": 0.64071207,
      "epoch": 0.4463470319634703,
      "grad_norm": 5.875,
      "learning_rate": 9.202273968699075e-06,
      "loss": 1.68462486,
      "memory(GiB)": 97.17,
      "step": 17595,
      "train_speed(iter/s)": 1.633923
    },
    {
      "acc": 0.66767244,
      "epoch": 0.44647387113140535,
      "grad_norm": 7.65625,
      "learning_rate": 9.201705646339714e-06,
      "loss": 1.61972485,
      "memory(GiB)": 97.17,
      "step": 17600,
      "train_speed(iter/s)": 1.633974
    },
    {
      "acc": 0.65168958,
      "epoch": 0.44660071029934045,
      "grad_norm": 4.96875,
      "learning_rate": 9.201137139170578e-06,
      "loss": 1.5934412,
      "memory(GiB)": 97.17,
      "step": 17605,
      "train_speed(iter/s)": 1.634024
    },
    {
      "acc": 0.66115465,
      "epoch": 0.4467275494672755,
      "grad_norm": 5.0625,
      "learning_rate": 9.200568447216673e-06,
      "loss": 1.58206282,
      "memory(GiB)": 97.17,
      "step": 17610,
      "train_speed(iter/s)": 1.634073
    },
    {
      "acc": 0.65404058,
      "epoch": 0.44685438863521054,
      "grad_norm": 5.0625,
      "learning_rate": 9.199999570503015e-06,
      "loss": 1.57423792,
      "memory(GiB)": 97.17,
      "step": 17615,
      "train_speed(iter/s)": 1.634124
    },
    {
      "acc": 0.6270751,
      "epoch": 0.4469812278031456,
      "grad_norm": 4.375,
      "learning_rate": 9.199430509054625e-06,
      "loss": 1.74342499,
      "memory(GiB)": 97.17,
      "step": 17620,
      "train_speed(iter/s)": 1.634175
    },
    {
      "acc": 0.65442171,
      "epoch": 0.4471080669710807,
      "grad_norm": 6.0625,
      "learning_rate": 9.19886126289653e-06,
      "loss": 1.61158905,
      "memory(GiB)": 97.17,
      "step": 17625,
      "train_speed(iter/s)": 1.63422
    },
    {
      "acc": 0.6561059,
      "epoch": 0.44723490613901573,
      "grad_norm": 5.46875,
      "learning_rate": 9.198291832053771e-06,
      "loss": 1.65528717,
      "memory(GiB)": 97.17,
      "step": 17630,
      "train_speed(iter/s)": 1.634267
    },
    {
      "acc": 0.65104151,
      "epoch": 0.4473617453069508,
      "grad_norm": 5.8125,
      "learning_rate": 9.197722216551395e-06,
      "loss": 1.63396149,
      "memory(GiB)": 97.17,
      "step": 17635,
      "train_speed(iter/s)": 1.634317
    },
    {
      "acc": 0.6543983,
      "epoch": 0.4474885844748858,
      "grad_norm": 4.59375,
      "learning_rate": 9.197152416414452e-06,
      "loss": 1.58556957,
      "memory(GiB)": 97.17,
      "step": 17640,
      "train_speed(iter/s)": 1.634368
    },
    {
      "acc": 0.64443326,
      "epoch": 0.4476154236428209,
      "grad_norm": 5.3125,
      "learning_rate": 9.196582431668007e-06,
      "loss": 1.71702366,
      "memory(GiB)": 97.17,
      "step": 17645,
      "train_speed(iter/s)": 1.634419
    },
    {
      "acc": 0.66325607,
      "epoch": 0.44774226281075596,
      "grad_norm": 6.0625,
      "learning_rate": 9.196012262337131e-06,
      "loss": 1.59066172,
      "memory(GiB)": 97.17,
      "step": 17650,
      "train_speed(iter/s)": 1.634468
    },
    {
      "acc": 0.64031544,
      "epoch": 0.447869101978691,
      "grad_norm": 5.96875,
      "learning_rate": 9.195441908446902e-06,
      "loss": 1.70693684,
      "memory(GiB)": 97.17,
      "step": 17655,
      "train_speed(iter/s)": 1.634518
    },
    {
      "acc": 0.65541973,
      "epoch": 0.44799594114662605,
      "grad_norm": 5.53125,
      "learning_rate": 9.194871370022407e-06,
      "loss": 1.63178329,
      "memory(GiB)": 97.17,
      "step": 17660,
      "train_speed(iter/s)": 1.634568
    },
    {
      "acc": 0.66145759,
      "epoch": 0.44812278031456115,
      "grad_norm": 5.4375,
      "learning_rate": 9.194300647088739e-06,
      "loss": 1.60694675,
      "memory(GiB)": 97.17,
      "step": 17665,
      "train_speed(iter/s)": 1.634616
    },
    {
      "acc": 0.63905807,
      "epoch": 0.4482496194824962,
      "grad_norm": 5.40625,
      "learning_rate": 9.193729739671002e-06,
      "loss": 1.65575333,
      "memory(GiB)": 97.17,
      "step": 17670,
      "train_speed(iter/s)": 1.634666
    },
    {
      "acc": 0.64483628,
      "epoch": 0.44837645865043124,
      "grad_norm": 4.5,
      "learning_rate": 9.193158647794308e-06,
      "loss": 1.61878014,
      "memory(GiB)": 97.17,
      "step": 17675,
      "train_speed(iter/s)": 1.634713
    },
    {
      "acc": 0.64103479,
      "epoch": 0.4485032978183663,
      "grad_norm": 6.03125,
      "learning_rate": 9.192587371483777e-06,
      "loss": 1.64115047,
      "memory(GiB)": 97.17,
      "step": 17680,
      "train_speed(iter/s)": 1.634763
    },
    {
      "acc": 0.63627243,
      "epoch": 0.4486301369863014,
      "grad_norm": 6.25,
      "learning_rate": 9.192015910764535e-06,
      "loss": 1.69046936,
      "memory(GiB)": 97.17,
      "step": 17685,
      "train_speed(iter/s)": 1.63481
    },
    {
      "acc": 0.6593894,
      "epoch": 0.44875697615423643,
      "grad_norm": 5.96875,
      "learning_rate": 9.191444265661715e-06,
      "loss": 1.64550056,
      "memory(GiB)": 97.17,
      "step": 17690,
      "train_speed(iter/s)": 1.634861
    },
    {
      "acc": 0.64664779,
      "epoch": 0.4488838153221715,
      "grad_norm": 5.125,
      "learning_rate": 9.190872436200464e-06,
      "loss": 1.65807114,
      "memory(GiB)": 97.17,
      "step": 17695,
      "train_speed(iter/s)": 1.63491
    },
    {
      "acc": 0.63693523,
      "epoch": 0.4490106544901065,
      "grad_norm": 6.40625,
      "learning_rate": 9.190300422405933e-06,
      "loss": 1.61289158,
      "memory(GiB)": 97.17,
      "step": 17700,
      "train_speed(iter/s)": 1.634959
    },
    {
      "acc": 0.6512929,
      "epoch": 0.4491374936580416,
      "grad_norm": 5.6875,
      "learning_rate": 9.18972822430328e-06,
      "loss": 1.661796,
      "memory(GiB)": 97.17,
      "step": 17705,
      "train_speed(iter/s)": 1.635012
    },
    {
      "acc": 0.65816598,
      "epoch": 0.44926433282597666,
      "grad_norm": 5.5,
      "learning_rate": 9.189155841917675e-06,
      "loss": 1.55127144,
      "memory(GiB)": 97.17,
      "step": 17710,
      "train_speed(iter/s)": 1.635059
    },
    {
      "acc": 0.64287744,
      "epoch": 0.4493911719939117,
      "grad_norm": 5.6875,
      "learning_rate": 9.18858327527429e-06,
      "loss": 1.6665947,
      "memory(GiB)": 97.17,
      "step": 17715,
      "train_speed(iter/s)": 1.635111
    },
    {
      "acc": 0.63745298,
      "epoch": 0.44951801116184675,
      "grad_norm": 5.40625,
      "learning_rate": 9.188010524398314e-06,
      "loss": 1.62430573,
      "memory(GiB)": 97.17,
      "step": 17720,
      "train_speed(iter/s)": 1.635161
    },
    {
      "acc": 0.68063426,
      "epoch": 0.44964485032978185,
      "grad_norm": 6.4375,
      "learning_rate": 9.187437589314939e-06,
      "loss": 1.51655159,
      "memory(GiB)": 97.17,
      "step": 17725,
      "train_speed(iter/s)": 1.635209
    },
    {
      "acc": 0.65839486,
      "epoch": 0.4497716894977169,
      "grad_norm": 5.6875,
      "learning_rate": 9.186864470049358e-06,
      "loss": 1.5760601,
      "memory(GiB)": 97.17,
      "step": 17730,
      "train_speed(iter/s)": 1.635259
    },
    {
      "acc": 0.66085672,
      "epoch": 0.44989852866565194,
      "grad_norm": 4.78125,
      "learning_rate": 9.186291166626789e-06,
      "loss": 1.59891024,
      "memory(GiB)": 97.17,
      "step": 17735,
      "train_speed(iter/s)": 1.63531
    },
    {
      "acc": 0.6589829,
      "epoch": 0.450025367833587,
      "grad_norm": 5.59375,
      "learning_rate": 9.185717679072444e-06,
      "loss": 1.61997185,
      "memory(GiB)": 97.17,
      "step": 17740,
      "train_speed(iter/s)": 1.635359
    },
    {
      "acc": 0.6514677,
      "epoch": 0.4501522070015221,
      "grad_norm": 5.71875,
      "learning_rate": 9.185144007411547e-06,
      "loss": 1.61622772,
      "memory(GiB)": 97.17,
      "step": 17745,
      "train_speed(iter/s)": 1.63541
    },
    {
      "acc": 0.63201022,
      "epoch": 0.45027904616945713,
      "grad_norm": 4.9375,
      "learning_rate": 9.18457015166933e-06,
      "loss": 1.62467995,
      "memory(GiB)": 97.17,
      "step": 17750,
      "train_speed(iter/s)": 1.635458
    },
    {
      "acc": 0.65665436,
      "epoch": 0.4504058853373922,
      "grad_norm": 6.40625,
      "learning_rate": 9.183996111871034e-06,
      "loss": 1.55665665,
      "memory(GiB)": 97.17,
      "step": 17755,
      "train_speed(iter/s)": 1.635507
    },
    {
      "acc": 0.67177896,
      "epoch": 0.4505327245053272,
      "grad_norm": 5.875,
      "learning_rate": 9.18342188804191e-06,
      "loss": 1.65101967,
      "memory(GiB)": 97.17,
      "step": 17760,
      "train_speed(iter/s)": 1.635557
    },
    {
      "acc": 0.64590473,
      "epoch": 0.4506595636732623,
      "grad_norm": 12.4375,
      "learning_rate": 9.182847480207215e-06,
      "loss": 1.62943039,
      "memory(GiB)": 97.17,
      "step": 17765,
      "train_speed(iter/s)": 1.635606
    },
    {
      "acc": 0.6614017,
      "epoch": 0.45078640284119736,
      "grad_norm": 4.9375,
      "learning_rate": 9.182272888392211e-06,
      "loss": 1.60033531,
      "memory(GiB)": 97.17,
      "step": 17770,
      "train_speed(iter/s)": 1.635654
    },
    {
      "acc": 0.64702106,
      "epoch": 0.4509132420091324,
      "grad_norm": 6.84375,
      "learning_rate": 9.181698112622175e-06,
      "loss": 1.63443604,
      "memory(GiB)": 97.17,
      "step": 17775,
      "train_speed(iter/s)": 1.635706
    },
    {
      "acc": 0.6286911,
      "epoch": 0.45104008117706745,
      "grad_norm": 5.15625,
      "learning_rate": 9.181123152922384e-06,
      "loss": 1.69296875,
      "memory(GiB)": 97.17,
      "step": 17780,
      "train_speed(iter/s)": 1.635752
    },
    {
      "acc": 0.64293375,
      "epoch": 0.45116692034500255,
      "grad_norm": 5.03125,
      "learning_rate": 9.18054800931813e-06,
      "loss": 1.63769188,
      "memory(GiB)": 97.17,
      "step": 17785,
      "train_speed(iter/s)": 1.635799
    },
    {
      "acc": 0.64541492,
      "epoch": 0.4512937595129376,
      "grad_norm": 5.65625,
      "learning_rate": 9.17997268183471e-06,
      "loss": 1.58755579,
      "memory(GiB)": 97.17,
      "step": 17790,
      "train_speed(iter/s)": 1.635848
    },
    {
      "acc": 0.63184462,
      "epoch": 0.45142059868087264,
      "grad_norm": 5.3125,
      "learning_rate": 9.17939717049743e-06,
      "loss": 1.67936535,
      "memory(GiB)": 97.17,
      "step": 17795,
      "train_speed(iter/s)": 1.6359
    },
    {
      "acc": 0.65373926,
      "epoch": 0.4515474378488077,
      "grad_norm": 6.4375,
      "learning_rate": 9.1788214753316e-06,
      "loss": 1.64679909,
      "memory(GiB)": 97.17,
      "step": 17800,
      "train_speed(iter/s)": 1.635947
    },
    {
      "acc": 0.65907717,
      "epoch": 0.4516742770167428,
      "grad_norm": 5.96875,
      "learning_rate": 9.178245596362546e-06,
      "loss": 1.60675106,
      "memory(GiB)": 97.17,
      "step": 17805,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.6508554,
      "epoch": 0.45180111618467783,
      "grad_norm": 5.9375,
      "learning_rate": 9.177669533615599e-06,
      "loss": 1.61277542,
      "memory(GiB)": 97.17,
      "step": 17810,
      "train_speed(iter/s)": 1.636043
    },
    {
      "acc": 0.65542326,
      "epoch": 0.4519279553526129,
      "grad_norm": 5.03125,
      "learning_rate": 9.17709328711609e-06,
      "loss": 1.56318274,
      "memory(GiB)": 97.17,
      "step": 17815,
      "train_speed(iter/s)": 1.636092
    },
    {
      "acc": 0.63743639,
      "epoch": 0.4520547945205479,
      "grad_norm": 5.375,
      "learning_rate": 9.17651685688937e-06,
      "loss": 1.66922951,
      "memory(GiB)": 97.17,
      "step": 17820,
      "train_speed(iter/s)": 1.636138
    },
    {
      "acc": 0.65172262,
      "epoch": 0.452181633688483,
      "grad_norm": 5.59375,
      "learning_rate": 9.175940242960792e-06,
      "loss": 1.66571579,
      "memory(GiB)": 97.17,
      "step": 17825,
      "train_speed(iter/s)": 1.636191
    },
    {
      "acc": 0.65301452,
      "epoch": 0.45230847285641806,
      "grad_norm": 4.90625,
      "learning_rate": 9.175363445355718e-06,
      "loss": 1.6732338,
      "memory(GiB)": 97.17,
      "step": 17830,
      "train_speed(iter/s)": 1.63624
    },
    {
      "acc": 0.66110291,
      "epoch": 0.4524353120243531,
      "grad_norm": 5.53125,
      "learning_rate": 9.174786464099519e-06,
      "loss": 1.60229473,
      "memory(GiB)": 97.17,
      "step": 17835,
      "train_speed(iter/s)": 1.636288
    },
    {
      "acc": 0.64350657,
      "epoch": 0.45256215119228815,
      "grad_norm": 5.65625,
      "learning_rate": 9.17420929921757e-06,
      "loss": 1.64239502,
      "memory(GiB)": 97.17,
      "step": 17840,
      "train_speed(iter/s)": 1.636336
    },
    {
      "acc": 0.66892071,
      "epoch": 0.45268899036022325,
      "grad_norm": 6.65625,
      "learning_rate": 9.173631950735262e-06,
      "loss": 1.56081057,
      "memory(GiB)": 97.17,
      "step": 17845,
      "train_speed(iter/s)": 1.636383
    },
    {
      "acc": 0.65648899,
      "epoch": 0.4528158295281583,
      "grad_norm": 5.84375,
      "learning_rate": 9.173054418677985e-06,
      "loss": 1.64969406,
      "memory(GiB)": 97.17,
      "step": 17850,
      "train_speed(iter/s)": 1.636434
    },
    {
      "acc": 0.6633276,
      "epoch": 0.45294266869609334,
      "grad_norm": 6.53125,
      "learning_rate": 9.172476703071145e-06,
      "loss": 1.58426533,
      "memory(GiB)": 97.17,
      "step": 17855,
      "train_speed(iter/s)": 1.63648
    },
    {
      "acc": 0.66323233,
      "epoch": 0.4530695078640284,
      "grad_norm": 5.78125,
      "learning_rate": 9.171898803940148e-06,
      "loss": 1.56856079,
      "memory(GiB)": 97.17,
      "step": 17860,
      "train_speed(iter/s)": 1.636528
    },
    {
      "acc": 0.66387711,
      "epoch": 0.4531963470319635,
      "grad_norm": 4.84375,
      "learning_rate": 9.17132072131042e-06,
      "loss": 1.56297207,
      "memory(GiB)": 97.17,
      "step": 17865,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.65439901,
      "epoch": 0.45332318619989853,
      "grad_norm": 5.9375,
      "learning_rate": 9.170742455207378e-06,
      "loss": 1.60466576,
      "memory(GiB)": 97.17,
      "step": 17870,
      "train_speed(iter/s)": 1.636621
    },
    {
      "acc": 0.65063219,
      "epoch": 0.4534500253678336,
      "grad_norm": 7.125,
      "learning_rate": 9.170164005656465e-06,
      "loss": 1.61257973,
      "memory(GiB)": 97.17,
      "step": 17875,
      "train_speed(iter/s)": 1.636669
    },
    {
      "acc": 0.67193336,
      "epoch": 0.4535768645357686,
      "grad_norm": 6.46875,
      "learning_rate": 9.169585372683118e-06,
      "loss": 1.49588842,
      "memory(GiB)": 97.17,
      "step": 17880,
      "train_speed(iter/s)": 1.636715
    },
    {
      "acc": 0.66043739,
      "epoch": 0.4537037037037037,
      "grad_norm": 6.6875,
      "learning_rate": 9.169006556312794e-06,
      "loss": 1.61532135,
      "memory(GiB)": 97.17,
      "step": 17885,
      "train_speed(iter/s)": 1.636763
    },
    {
      "acc": 0.64468875,
      "epoch": 0.45383054287163876,
      "grad_norm": 5.65625,
      "learning_rate": 9.168427556570946e-06,
      "loss": 1.66786633,
      "memory(GiB)": 97.17,
      "step": 17890,
      "train_speed(iter/s)": 1.636808
    },
    {
      "acc": 0.66313658,
      "epoch": 0.4539573820395738,
      "grad_norm": 6.1875,
      "learning_rate": 9.167848373483044e-06,
      "loss": 1.61062508,
      "memory(GiB)": 97.17,
      "step": 17895,
      "train_speed(iter/s)": 1.636859
    },
    {
      "acc": 0.65240016,
      "epoch": 0.45408422120750885,
      "grad_norm": 5.34375,
      "learning_rate": 9.167269007074561e-06,
      "loss": 1.63836269,
      "memory(GiB)": 97.17,
      "step": 17900,
      "train_speed(iter/s)": 1.636905
    },
    {
      "acc": 0.65331364,
      "epoch": 0.45421106037544395,
      "grad_norm": 4.15625,
      "learning_rate": 9.166689457370983e-06,
      "loss": 1.6184988,
      "memory(GiB)": 97.17,
      "step": 17905,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.63214364,
      "epoch": 0.454337899543379,
      "grad_norm": 5.40625,
      "learning_rate": 9.166109724397801e-06,
      "loss": 1.65282021,
      "memory(GiB)": 97.17,
      "step": 17910,
      "train_speed(iter/s)": 1.637001
    },
    {
      "acc": 0.65271082,
      "epoch": 0.45446473871131404,
      "grad_norm": 7.84375,
      "learning_rate": 9.165529808180511e-06,
      "loss": 1.61452484,
      "memory(GiB)": 97.17,
      "step": 17915,
      "train_speed(iter/s)": 1.637052
    },
    {
      "acc": 0.6501111,
      "epoch": 0.4545915778792491,
      "grad_norm": 4.65625,
      "learning_rate": 9.164949708744622e-06,
      "loss": 1.62055683,
      "memory(GiB)": 97.17,
      "step": 17920,
      "train_speed(iter/s)": 1.637099
    },
    {
      "acc": 0.6526082,
      "epoch": 0.4547184170471842,
      "grad_norm": 8.25,
      "learning_rate": 9.164369426115652e-06,
      "loss": 1.58178997,
      "memory(GiB)": 97.17,
      "step": 17925,
      "train_speed(iter/s)": 1.637146
    },
    {
      "acc": 0.65028062,
      "epoch": 0.45484525621511923,
      "grad_norm": 5.1875,
      "learning_rate": 9.16378896031912e-06,
      "loss": 1.65476837,
      "memory(GiB)": 97.17,
      "step": 17930,
      "train_speed(iter/s)": 1.637195
    },
    {
      "acc": 0.65379977,
      "epoch": 0.4549720953830543,
      "grad_norm": 5.6875,
      "learning_rate": 9.163208311380561e-06,
      "loss": 1.60218658,
      "memory(GiB)": 97.17,
      "step": 17935,
      "train_speed(iter/s)": 1.637244
    },
    {
      "acc": 0.65739164,
      "epoch": 0.4550989345509893,
      "grad_norm": 4.8125,
      "learning_rate": 9.162627479325511e-06,
      "loss": 1.59945431,
      "memory(GiB)": 97.17,
      "step": 17940,
      "train_speed(iter/s)": 1.637293
    },
    {
      "acc": 0.66322875,
      "epoch": 0.4552257737189244,
      "grad_norm": 6.0,
      "learning_rate": 9.16204646417952e-06,
      "loss": 1.60096436,
      "memory(GiB)": 97.17,
      "step": 17945,
      "train_speed(iter/s)": 1.637342
    },
    {
      "acc": 0.65176544,
      "epoch": 0.45535261288685946,
      "grad_norm": 5.3125,
      "learning_rate": 9.161465265968145e-06,
      "loss": 1.56856689,
      "memory(GiB)": 97.17,
      "step": 17950,
      "train_speed(iter/s)": 1.637389
    },
    {
      "acc": 0.66775045,
      "epoch": 0.4554794520547945,
      "grad_norm": 6.4375,
      "learning_rate": 9.160883884716948e-06,
      "loss": 1.54070654,
      "memory(GiB)": 97.17,
      "step": 17955,
      "train_speed(iter/s)": 1.637436
    },
    {
      "acc": 0.65757189,
      "epoch": 0.45560629122272955,
      "grad_norm": 4.875,
      "learning_rate": 9.1603023204515e-06,
      "loss": 1.65287247,
      "memory(GiB)": 97.17,
      "step": 17960,
      "train_speed(iter/s)": 1.637485
    },
    {
      "acc": 0.64055328,
      "epoch": 0.45573313039066465,
      "grad_norm": 4.78125,
      "learning_rate": 9.15972057319738e-06,
      "loss": 1.62745132,
      "memory(GiB)": 97.17,
      "step": 17965,
      "train_speed(iter/s)": 1.63753
    },
    {
      "acc": 0.66283889,
      "epoch": 0.4558599695585997,
      "grad_norm": 7.59375,
      "learning_rate": 9.159138642980178e-06,
      "loss": 1.63845444,
      "memory(GiB)": 97.17,
      "step": 17970,
      "train_speed(iter/s)": 1.637574
    },
    {
      "acc": 0.64528112,
      "epoch": 0.45598680872653474,
      "grad_norm": 6.78125,
      "learning_rate": 9.15855652982549e-06,
      "loss": 1.66781158,
      "memory(GiB)": 97.17,
      "step": 17975,
      "train_speed(iter/s)": 1.637619
    },
    {
      "acc": 0.66027126,
      "epoch": 0.4561136478944698,
      "grad_norm": 7.3125,
      "learning_rate": 9.15797423375892e-06,
      "loss": 1.62811623,
      "memory(GiB)": 97.17,
      "step": 17980,
      "train_speed(iter/s)": 1.637667
    },
    {
      "acc": 0.65096769,
      "epoch": 0.4562404870624049,
      "grad_norm": 6.0625,
      "learning_rate": 9.157391754806079e-06,
      "loss": 1.65528755,
      "memory(GiB)": 97.17,
      "step": 17985,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.66090035,
      "epoch": 0.45636732623033993,
      "grad_norm": 7.40625,
      "learning_rate": 9.156809092992588e-06,
      "loss": 1.59680519,
      "memory(GiB)": 97.17,
      "step": 17990,
      "train_speed(iter/s)": 1.637757
    },
    {
      "acc": 0.6504416,
      "epoch": 0.456494165398275,
      "grad_norm": 6.625,
      "learning_rate": 9.156226248344072e-06,
      "loss": 1.62218189,
      "memory(GiB)": 97.17,
      "step": 17995,
      "train_speed(iter/s)": 1.6378
    },
    {
      "acc": 0.66787395,
      "epoch": 0.45662100456621,
      "grad_norm": 4.9375,
      "learning_rate": 9.15564322088617e-06,
      "loss": 1.57303944,
      "memory(GiB)": 97.17,
      "step": 18000,
      "train_speed(iter/s)": 1.637847
    },
    {
      "epoch": 0.45662100456621,
      "eval_acc": 0.6431803231444878,
      "eval_loss": 1.5907191038131714,
      "eval_runtime": 58.3705,
      "eval_samples_per_second": 109.131,
      "eval_steps_per_second": 27.291,
      "step": 18000
    },
    {
      "acc": 0.64999857,
      "epoch": 0.4567478437341451,
      "grad_norm": 5.40625,
      "learning_rate": 9.155060010644525e-06,
      "loss": 1.62140446,
      "memory(GiB)": 97.17,
      "step": 18005,
      "train_speed(iter/s)": 1.628615
    },
    {
      "acc": 0.65930071,
      "epoch": 0.45687468290208016,
      "grad_norm": 6.84375,
      "learning_rate": 9.154476617644792e-06,
      "loss": 1.57437725,
      "memory(GiB)": 97.17,
      "step": 18010,
      "train_speed(iter/s)": 1.628661
    },
    {
      "acc": 0.65113735,
      "epoch": 0.4570015220700152,
      "grad_norm": 9.3125,
      "learning_rate": 9.153893041912627e-06,
      "loss": 1.56833649,
      "memory(GiB)": 97.17,
      "step": 18015,
      "train_speed(iter/s)": 1.628707
    },
    {
      "acc": 0.66707897,
      "epoch": 0.45712836123795025,
      "grad_norm": 6.34375,
      "learning_rate": 9.1533092834737e-06,
      "loss": 1.56586494,
      "memory(GiB)": 97.17,
      "step": 18020,
      "train_speed(iter/s)": 1.628755
    },
    {
      "acc": 0.65480509,
      "epoch": 0.45725520040588535,
      "grad_norm": 6.0,
      "learning_rate": 9.152725342353688e-06,
      "loss": 1.62384796,
      "memory(GiB)": 97.17,
      "step": 18025,
      "train_speed(iter/s)": 1.6288
    },
    {
      "acc": 0.63864136,
      "epoch": 0.4573820395738204,
      "grad_norm": 7.21875,
      "learning_rate": 9.152141218578276e-06,
      "loss": 1.69390335,
      "memory(GiB)": 97.17,
      "step": 18030,
      "train_speed(iter/s)": 1.628846
    },
    {
      "acc": 0.64268441,
      "epoch": 0.45750887874175544,
      "grad_norm": 6.71875,
      "learning_rate": 9.151556912173154e-06,
      "loss": 1.70281296,
      "memory(GiB)": 97.17,
      "step": 18035,
      "train_speed(iter/s)": 1.628891
    },
    {
      "acc": 0.65205879,
      "epoch": 0.4576357179096905,
      "grad_norm": 6.75,
      "learning_rate": 9.150972423164024e-06,
      "loss": 1.62303314,
      "memory(GiB)": 97.17,
      "step": 18040,
      "train_speed(iter/s)": 1.628936
    },
    {
      "acc": 0.65466805,
      "epoch": 0.4577625570776256,
      "grad_norm": 6.8125,
      "learning_rate": 9.150387751576594e-06,
      "loss": 1.67525177,
      "memory(GiB)": 97.17,
      "step": 18045,
      "train_speed(iter/s)": 1.628984
    },
    {
      "acc": 0.64855871,
      "epoch": 0.45788939624556063,
      "grad_norm": 4.3125,
      "learning_rate": 9.14980289743658e-06,
      "loss": 1.62913437,
      "memory(GiB)": 97.17,
      "step": 18050,
      "train_speed(iter/s)": 1.629017
    },
    {
      "acc": 0.64263763,
      "epoch": 0.4580162354134957,
      "grad_norm": 5.90625,
      "learning_rate": 9.149217860769708e-06,
      "loss": 1.66544647,
      "memory(GiB)": 97.17,
      "step": 18055,
      "train_speed(iter/s)": 1.629067
    },
    {
      "acc": 0.66130505,
      "epoch": 0.4581430745814307,
      "grad_norm": 6.25,
      "learning_rate": 9.14863264160171e-06,
      "loss": 1.57379818,
      "memory(GiB)": 97.17,
      "step": 18060,
      "train_speed(iter/s)": 1.629116
    },
    {
      "acc": 0.64522581,
      "epoch": 0.4582699137493658,
      "grad_norm": 5.59375,
      "learning_rate": 9.148047239958325e-06,
      "loss": 1.6959671,
      "memory(GiB)": 97.17,
      "step": 18065,
      "train_speed(iter/s)": 1.629168
    },
    {
      "acc": 0.64630213,
      "epoch": 0.45839675291730086,
      "grad_norm": 5.875,
      "learning_rate": 9.147461655865302e-06,
      "loss": 1.62077827,
      "memory(GiB)": 97.17,
      "step": 18070,
      "train_speed(iter/s)": 1.629219
    },
    {
      "acc": 0.66259346,
      "epoch": 0.4585235920852359,
      "grad_norm": 5.71875,
      "learning_rate": 9.1468758893484e-06,
      "loss": 1.65996361,
      "memory(GiB)": 97.17,
      "step": 18075,
      "train_speed(iter/s)": 1.629271
    },
    {
      "acc": 0.66683135,
      "epoch": 0.45865043125317095,
      "grad_norm": 5.03125,
      "learning_rate": 9.14628994043338e-06,
      "loss": 1.64738579,
      "memory(GiB)": 97.17,
      "step": 18080,
      "train_speed(iter/s)": 1.629322
    },
    {
      "acc": 0.65946379,
      "epoch": 0.45877727042110605,
      "grad_norm": 5.1875,
      "learning_rate": 9.145703809146018e-06,
      "loss": 1.63190613,
      "memory(GiB)": 97.17,
      "step": 18085,
      "train_speed(iter/s)": 1.629373
    },
    {
      "acc": 0.63743377,
      "epoch": 0.4589041095890411,
      "grad_norm": 5.875,
      "learning_rate": 9.145117495512092e-06,
      "loss": 1.65002594,
      "memory(GiB)": 97.17,
      "step": 18090,
      "train_speed(iter/s)": 1.62942
    },
    {
      "acc": 0.65873318,
      "epoch": 0.45903094875697614,
      "grad_norm": 5.3125,
      "learning_rate": 9.144530999557393e-06,
      "loss": 1.70841599,
      "memory(GiB)": 97.17,
      "step": 18095,
      "train_speed(iter/s)": 1.629472
    },
    {
      "acc": 0.6582479,
      "epoch": 0.4591577879249112,
      "grad_norm": 7.4375,
      "learning_rate": 9.143944321307718e-06,
      "loss": 1.62365761,
      "memory(GiB)": 97.17,
      "step": 18100,
      "train_speed(iter/s)": 1.629524
    },
    {
      "acc": 0.65006514,
      "epoch": 0.4592846270928463,
      "grad_norm": 5.03125,
      "learning_rate": 9.143357460788868e-06,
      "loss": 1.65395966,
      "memory(GiB)": 97.17,
      "step": 18105,
      "train_speed(iter/s)": 1.629568
    },
    {
      "acc": 0.64196391,
      "epoch": 0.45941146626078133,
      "grad_norm": 6.15625,
      "learning_rate": 9.14277041802666e-06,
      "loss": 1.70032959,
      "memory(GiB)": 97.17,
      "step": 18110,
      "train_speed(iter/s)": 1.629617
    },
    {
      "acc": 0.6500701,
      "epoch": 0.4595383054287164,
      "grad_norm": 5.65625,
      "learning_rate": 9.142183193046912e-06,
      "loss": 1.61194954,
      "memory(GiB)": 97.17,
      "step": 18115,
      "train_speed(iter/s)": 1.629665
    },
    {
      "acc": 0.66789994,
      "epoch": 0.4596651445966514,
      "grad_norm": 5.28125,
      "learning_rate": 9.141595785875453e-06,
      "loss": 1.53755598,
      "memory(GiB)": 97.17,
      "step": 18120,
      "train_speed(iter/s)": 1.629715
    },
    {
      "acc": 0.65315056,
      "epoch": 0.4597919837645865,
      "grad_norm": 4.65625,
      "learning_rate": 9.141008196538122e-06,
      "loss": 1.60123291,
      "memory(GiB)": 97.17,
      "step": 18125,
      "train_speed(iter/s)": 1.629768
    },
    {
      "acc": 0.64036112,
      "epoch": 0.45991882293252156,
      "grad_norm": 5.625,
      "learning_rate": 9.14042042506076e-06,
      "loss": 1.6665062,
      "memory(GiB)": 97.17,
      "step": 18130,
      "train_speed(iter/s)": 1.629821
    },
    {
      "acc": 0.65954399,
      "epoch": 0.4600456621004566,
      "grad_norm": 6.34375,
      "learning_rate": 9.139832471469224e-06,
      "loss": 1.62609215,
      "memory(GiB)": 97.17,
      "step": 18135,
      "train_speed(iter/s)": 1.629867
    },
    {
      "acc": 0.66797457,
      "epoch": 0.46017250126839165,
      "grad_norm": 4.6875,
      "learning_rate": 9.13924433578937e-06,
      "loss": 1.58874931,
      "memory(GiB)": 97.17,
      "step": 18140,
      "train_speed(iter/s)": 1.629916
    },
    {
      "acc": 0.65389252,
      "epoch": 0.46029934043632675,
      "grad_norm": 5.1875,
      "learning_rate": 9.138656018047074e-06,
      "loss": 1.60943718,
      "memory(GiB)": 97.17,
      "step": 18145,
      "train_speed(iter/s)": 1.629967
    },
    {
      "acc": 0.66221528,
      "epoch": 0.4604261796042618,
      "grad_norm": 5.21875,
      "learning_rate": 9.138067518268206e-06,
      "loss": 1.59948177,
      "memory(GiB)": 97.17,
      "step": 18150,
      "train_speed(iter/s)": 1.630015
    },
    {
      "acc": 0.66288638,
      "epoch": 0.46055301877219684,
      "grad_norm": 5.03125,
      "learning_rate": 9.137478836478654e-06,
      "loss": 1.60354652,
      "memory(GiB)": 97.17,
      "step": 18155,
      "train_speed(iter/s)": 1.630066
    },
    {
      "acc": 0.65707946,
      "epoch": 0.4606798579401319,
      "grad_norm": 5.96875,
      "learning_rate": 9.13688997270431e-06,
      "loss": 1.594314,
      "memory(GiB)": 97.17,
      "step": 18160,
      "train_speed(iter/s)": 1.630117
    },
    {
      "acc": 0.67923374,
      "epoch": 0.460806697108067,
      "grad_norm": 5.53125,
      "learning_rate": 9.136300926971076e-06,
      "loss": 1.55084715,
      "memory(GiB)": 97.17,
      "step": 18165,
      "train_speed(iter/s)": 1.630169
    },
    {
      "acc": 0.65642824,
      "epoch": 0.46093353627600203,
      "grad_norm": 4.84375,
      "learning_rate": 9.135711699304858e-06,
      "loss": 1.66619377,
      "memory(GiB)": 97.17,
      "step": 18170,
      "train_speed(iter/s)": 1.630222
    },
    {
      "acc": 0.65295868,
      "epoch": 0.4610603754439371,
      "grad_norm": 5.40625,
      "learning_rate": 9.135122289731575e-06,
      "loss": 1.6186924,
      "memory(GiB)": 97.17,
      "step": 18175,
      "train_speed(iter/s)": 1.630275
    },
    {
      "acc": 0.65446663,
      "epoch": 0.4611872146118721,
      "grad_norm": 6.3125,
      "learning_rate": 9.134532698277154e-06,
      "loss": 1.57828321,
      "memory(GiB)": 97.17,
      "step": 18180,
      "train_speed(iter/s)": 1.630329
    },
    {
      "acc": 0.65571055,
      "epoch": 0.4613140537798072,
      "grad_norm": 7.5,
      "learning_rate": 9.133942924967524e-06,
      "loss": 1.63390808,
      "memory(GiB)": 97.17,
      "step": 18185,
      "train_speed(iter/s)": 1.630381
    },
    {
      "acc": 0.6724884,
      "epoch": 0.46144089294774226,
      "grad_norm": 7.9375,
      "learning_rate": 9.133352969828628e-06,
      "loss": 1.55879192,
      "memory(GiB)": 97.17,
      "step": 18190,
      "train_speed(iter/s)": 1.630433
    },
    {
      "acc": 0.64934139,
      "epoch": 0.4615677321156773,
      "grad_norm": 5.34375,
      "learning_rate": 9.132762832886416e-06,
      "loss": 1.57403498,
      "memory(GiB)": 97.17,
      "step": 18195,
      "train_speed(iter/s)": 1.630485
    },
    {
      "acc": 0.64723287,
      "epoch": 0.46169457128361235,
      "grad_norm": 6.53125,
      "learning_rate": 9.13217251416684e-06,
      "loss": 1.69276886,
      "memory(GiB)": 97.17,
      "step": 18200,
      "train_speed(iter/s)": 1.630539
    },
    {
      "acc": 0.64785681,
      "epoch": 0.46182141045154745,
      "grad_norm": 6.5,
      "learning_rate": 9.131582013695867e-06,
      "loss": 1.63392258,
      "memory(GiB)": 97.17,
      "step": 18205,
      "train_speed(iter/s)": 1.630589
    },
    {
      "acc": 0.66743007,
      "epoch": 0.4619482496194825,
      "grad_norm": 6.25,
      "learning_rate": 9.130991331499474e-06,
      "loss": 1.67314148,
      "memory(GiB)": 97.17,
      "step": 18210,
      "train_speed(iter/s)": 1.630644
    },
    {
      "acc": 0.67090573,
      "epoch": 0.46207508878741754,
      "grad_norm": 6.21875,
      "learning_rate": 9.130400467603637e-06,
      "loss": 1.55846739,
      "memory(GiB)": 97.17,
      "step": 18215,
      "train_speed(iter/s)": 1.630696
    },
    {
      "acc": 0.63479166,
      "epoch": 0.4622019279553526,
      "grad_norm": 6.40625,
      "learning_rate": 9.129809422034349e-06,
      "loss": 1.66420631,
      "memory(GiB)": 97.17,
      "step": 18220,
      "train_speed(iter/s)": 1.630752
    },
    {
      "acc": 0.65594912,
      "epoch": 0.4623287671232877,
      "grad_norm": 6.3125,
      "learning_rate": 9.129218194817601e-06,
      "loss": 1.6711628,
      "memory(GiB)": 97.17,
      "step": 18225,
      "train_speed(iter/s)": 1.630805
    },
    {
      "acc": 0.64859838,
      "epoch": 0.46245560629122273,
      "grad_norm": 5.6875,
      "learning_rate": 9.128626785979404e-06,
      "loss": 1.67346287,
      "memory(GiB)": 97.17,
      "step": 18230,
      "train_speed(iter/s)": 1.630857
    },
    {
      "acc": 0.64274359,
      "epoch": 0.4625824454591578,
      "grad_norm": 5.625,
      "learning_rate": 9.128035195545766e-06,
      "loss": 1.64927006,
      "memory(GiB)": 97.17,
      "step": 18235,
      "train_speed(iter/s)": 1.630911
    },
    {
      "acc": 0.63747911,
      "epoch": 0.4627092846270928,
      "grad_norm": 5.59375,
      "learning_rate": 9.12744342354271e-06,
      "loss": 1.74260864,
      "memory(GiB)": 97.17,
      "step": 18240,
      "train_speed(iter/s)": 1.630961
    },
    {
      "acc": 0.65783386,
      "epoch": 0.4628361237950279,
      "grad_norm": 6.03125,
      "learning_rate": 9.126851469996265e-06,
      "loss": 1.60981178,
      "memory(GiB)": 97.17,
      "step": 18245,
      "train_speed(iter/s)": 1.631013
    },
    {
      "acc": 0.64982357,
      "epoch": 0.46296296296296297,
      "grad_norm": 6.5,
      "learning_rate": 9.126259334932467e-06,
      "loss": 1.62623215,
      "memory(GiB)": 97.17,
      "step": 18250,
      "train_speed(iter/s)": 1.631068
    },
    {
      "acc": 0.64053478,
      "epoch": 0.463089802130898,
      "grad_norm": 7.15625,
      "learning_rate": 9.125667018377362e-06,
      "loss": 1.71250153,
      "memory(GiB)": 97.17,
      "step": 18255,
      "train_speed(iter/s)": 1.631122
    },
    {
      "acc": 0.65603752,
      "epoch": 0.46321664129883305,
      "grad_norm": 6.9375,
      "learning_rate": 9.125074520357002e-06,
      "loss": 1.6324482,
      "memory(GiB)": 97.17,
      "step": 18260,
      "train_speed(iter/s)": 1.631173
    },
    {
      "acc": 0.65509129,
      "epoch": 0.46334348046676815,
      "grad_norm": 5.25,
      "learning_rate": 9.124481840897446e-06,
      "loss": 1.5972024,
      "memory(GiB)": 97.17,
      "step": 18265,
      "train_speed(iter/s)": 1.631223
    },
    {
      "acc": 0.64845858,
      "epoch": 0.4634703196347032,
      "grad_norm": 5.0625,
      "learning_rate": 9.123888980024765e-06,
      "loss": 1.58163042,
      "memory(GiB)": 97.17,
      "step": 18270,
      "train_speed(iter/s)": 1.631272
    },
    {
      "acc": 0.6682056,
      "epoch": 0.46359715880263824,
      "grad_norm": 4.625,
      "learning_rate": 9.123295937765034e-06,
      "loss": 1.57059727,
      "memory(GiB)": 97.17,
      "step": 18275,
      "train_speed(iter/s)": 1.631325
    },
    {
      "acc": 0.67809019,
      "epoch": 0.4637239979705733,
      "grad_norm": 6.25,
      "learning_rate": 9.122702714144339e-06,
      "loss": 1.53272963,
      "memory(GiB)": 97.17,
      "step": 18280,
      "train_speed(iter/s)": 1.631376
    },
    {
      "acc": 0.65309391,
      "epoch": 0.4638508371385084,
      "grad_norm": 5.375,
      "learning_rate": 9.122109309188772e-06,
      "loss": 1.61041451,
      "memory(GiB)": 97.17,
      "step": 18285,
      "train_speed(iter/s)": 1.631429
    },
    {
      "acc": 0.67439466,
      "epoch": 0.46397767630644343,
      "grad_norm": 7.375,
      "learning_rate": 9.121515722924435e-06,
      "loss": 1.58415031,
      "memory(GiB)": 97.17,
      "step": 18290,
      "train_speed(iter/s)": 1.631482
    },
    {
      "acc": 0.66338997,
      "epoch": 0.4641045154743785,
      "grad_norm": 5.375,
      "learning_rate": 9.120921955377433e-06,
      "loss": 1.57285643,
      "memory(GiB)": 97.17,
      "step": 18295,
      "train_speed(iter/s)": 1.631535
    },
    {
      "acc": 0.6486599,
      "epoch": 0.4642313546423135,
      "grad_norm": 5.5,
      "learning_rate": 9.120328006573887e-06,
      "loss": 1.62669716,
      "memory(GiB)": 97.17,
      "step": 18300,
      "train_speed(iter/s)": 1.631584
    },
    {
      "acc": 0.65634398,
      "epoch": 0.4643581938102486,
      "grad_norm": 5.46875,
      "learning_rate": 9.119733876539916e-06,
      "loss": 1.55204182,
      "memory(GiB)": 97.17,
      "step": 18305,
      "train_speed(iter/s)": 1.631635
    },
    {
      "acc": 0.65420036,
      "epoch": 0.46448503297818367,
      "grad_norm": 6.125,
      "learning_rate": 9.119139565301658e-06,
      "loss": 1.59699821,
      "memory(GiB)": 97.17,
      "step": 18310,
      "train_speed(iter/s)": 1.631687
    },
    {
      "acc": 0.65560145,
      "epoch": 0.4646118721461187,
      "grad_norm": 5.71875,
      "learning_rate": 9.118545072885253e-06,
      "loss": 1.6689415,
      "memory(GiB)": 97.17,
      "step": 18315,
      "train_speed(iter/s)": 1.631737
    },
    {
      "acc": 0.66651087,
      "epoch": 0.46473871131405375,
      "grad_norm": 8.4375,
      "learning_rate": 9.117950399316845e-06,
      "loss": 1.57884626,
      "memory(GiB)": 97.17,
      "step": 18320,
      "train_speed(iter/s)": 1.631786
    },
    {
      "acc": 0.66842513,
      "epoch": 0.46486555048198885,
      "grad_norm": 6.0625,
      "learning_rate": 9.117355544622595e-06,
      "loss": 1.58536873,
      "memory(GiB)": 97.17,
      "step": 18325,
      "train_speed(iter/s)": 1.631835
    },
    {
      "acc": 0.64968004,
      "epoch": 0.4649923896499239,
      "grad_norm": 5.59375,
      "learning_rate": 9.116760508828664e-06,
      "loss": 1.55995865,
      "memory(GiB)": 97.17,
      "step": 18330,
      "train_speed(iter/s)": 1.631884
    },
    {
      "acc": 0.66877708,
      "epoch": 0.46511922881785894,
      "grad_norm": 5.34375,
      "learning_rate": 9.116165291961225e-06,
      "loss": 1.59292078,
      "memory(GiB)": 97.17,
      "step": 18335,
      "train_speed(iter/s)": 1.631931
    },
    {
      "acc": 0.64873347,
      "epoch": 0.465246067985794,
      "grad_norm": 5.0625,
      "learning_rate": 9.11556989404646e-06,
      "loss": 1.60118408,
      "memory(GiB)": 97.17,
      "step": 18340,
      "train_speed(iter/s)": 1.631979
    },
    {
      "acc": 0.64336104,
      "epoch": 0.4653729071537291,
      "grad_norm": 5.84375,
      "learning_rate": 9.114974315110558e-06,
      "loss": 1.68564701,
      "memory(GiB)": 97.17,
      "step": 18345,
      "train_speed(iter/s)": 1.632027
    },
    {
      "acc": 0.66673598,
      "epoch": 0.46549974632166413,
      "grad_norm": 6.3125,
      "learning_rate": 9.114378555179712e-06,
      "loss": 1.61895199,
      "memory(GiB)": 97.17,
      "step": 18350,
      "train_speed(iter/s)": 1.632076
    },
    {
      "acc": 0.66504173,
      "epoch": 0.4656265854895992,
      "grad_norm": 5.78125,
      "learning_rate": 9.11378261428013e-06,
      "loss": 1.56538124,
      "memory(GiB)": 97.17,
      "step": 18355,
      "train_speed(iter/s)": 1.632125
    },
    {
      "acc": 0.65559959,
      "epoch": 0.4657534246575342,
      "grad_norm": 5.71875,
      "learning_rate": 9.11318649243802e-06,
      "loss": 1.62557507,
      "memory(GiB)": 97.17,
      "step": 18360,
      "train_speed(iter/s)": 1.632174
    },
    {
      "acc": 0.65828705,
      "epoch": 0.4658802638254693,
      "grad_norm": 5.75,
      "learning_rate": 9.112590189679604e-06,
      "loss": 1.61647129,
      "memory(GiB)": 97.17,
      "step": 18365,
      "train_speed(iter/s)": 1.632228
    },
    {
      "acc": 0.64962778,
      "epoch": 0.46600710299340437,
      "grad_norm": 4.6875,
      "learning_rate": 9.111993706031109e-06,
      "loss": 1.60085125,
      "memory(GiB)": 97.17,
      "step": 18370,
      "train_speed(iter/s)": 1.632275
    },
    {
      "acc": 0.66088781,
      "epoch": 0.4661339421613394,
      "grad_norm": 5.53125,
      "learning_rate": 9.111397041518774e-06,
      "loss": 1.61294212,
      "memory(GiB)": 97.17,
      "step": 18375,
      "train_speed(iter/s)": 1.632322
    },
    {
      "acc": 0.65508494,
      "epoch": 0.46626078132927445,
      "grad_norm": 5.5625,
      "learning_rate": 9.11080019616884e-06,
      "loss": 1.59683514,
      "memory(GiB)": 97.17,
      "step": 18380,
      "train_speed(iter/s)": 1.632373
    },
    {
      "acc": 0.66044416,
      "epoch": 0.46638762049720955,
      "grad_norm": 6.03125,
      "learning_rate": 9.11020317000756e-06,
      "loss": 1.63567581,
      "memory(GiB)": 97.17,
      "step": 18385,
      "train_speed(iter/s)": 1.632422
    },
    {
      "acc": 0.63134689,
      "epoch": 0.4665144596651446,
      "grad_norm": 5.71875,
      "learning_rate": 9.109605963061194e-06,
      "loss": 1.73026295,
      "memory(GiB)": 97.17,
      "step": 18390,
      "train_speed(iter/s)": 1.632468
    },
    {
      "acc": 0.66334658,
      "epoch": 0.46664129883307964,
      "grad_norm": 6.375,
      "learning_rate": 9.10900857535601e-06,
      "loss": 1.573347,
      "memory(GiB)": 97.17,
      "step": 18395,
      "train_speed(iter/s)": 1.632518
    },
    {
      "acc": 0.63498793,
      "epoch": 0.4667681380010147,
      "grad_norm": 6.125,
      "learning_rate": 9.108411006918283e-06,
      "loss": 1.68694935,
      "memory(GiB)": 97.17,
      "step": 18400,
      "train_speed(iter/s)": 1.632569
    },
    {
      "acc": 0.65896225,
      "epoch": 0.4668949771689498,
      "grad_norm": 6.6875,
      "learning_rate": 9.107813257774298e-06,
      "loss": 1.63279266,
      "memory(GiB)": 97.17,
      "step": 18405,
      "train_speed(iter/s)": 1.632621
    },
    {
      "acc": 0.64386454,
      "epoch": 0.46702181633688483,
      "grad_norm": 5.9375,
      "learning_rate": 9.107215327950345e-06,
      "loss": 1.62228317,
      "memory(GiB)": 97.17,
      "step": 18410,
      "train_speed(iter/s)": 1.632669
    },
    {
      "acc": 0.65005245,
      "epoch": 0.4671486555048199,
      "grad_norm": 6.125,
      "learning_rate": 9.106617217472724e-06,
      "loss": 1.6490984,
      "memory(GiB)": 97.17,
      "step": 18415,
      "train_speed(iter/s)": 1.63272
    },
    {
      "acc": 0.64562855,
      "epoch": 0.4672754946727549,
      "grad_norm": 5.34375,
      "learning_rate": 9.106018926367744e-06,
      "loss": 1.63344383,
      "memory(GiB)": 97.17,
      "step": 18420,
      "train_speed(iter/s)": 1.63277
    },
    {
      "acc": 0.65251913,
      "epoch": 0.46740233384069,
      "grad_norm": 5.46875,
      "learning_rate": 9.10542045466172e-06,
      "loss": 1.63472099,
      "memory(GiB)": 97.17,
      "step": 18425,
      "train_speed(iter/s)": 1.632817
    },
    {
      "acc": 0.65662622,
      "epoch": 0.46752917300862507,
      "grad_norm": 5.78125,
      "learning_rate": 9.104821802380974e-06,
      "loss": 1.66220341,
      "memory(GiB)": 97.17,
      "step": 18430,
      "train_speed(iter/s)": 1.632866
    },
    {
      "acc": 0.65970116,
      "epoch": 0.4676560121765601,
      "grad_norm": 5.25,
      "learning_rate": 9.104222969551838e-06,
      "loss": 1.53581562,
      "memory(GiB)": 97.17,
      "step": 18435,
      "train_speed(iter/s)": 1.632914
    },
    {
      "acc": 0.66083355,
      "epoch": 0.46778285134449515,
      "grad_norm": 5.9375,
      "learning_rate": 9.103623956200654e-06,
      "loss": 1.63388538,
      "memory(GiB)": 97.17,
      "step": 18440,
      "train_speed(iter/s)": 1.632962
    },
    {
      "acc": 0.648207,
      "epoch": 0.46790969051243025,
      "grad_norm": 6.0,
      "learning_rate": 9.103024762353766e-06,
      "loss": 1.69622955,
      "memory(GiB)": 97.17,
      "step": 18445,
      "train_speed(iter/s)": 1.633007
    },
    {
      "acc": 0.66191196,
      "epoch": 0.4680365296803653,
      "grad_norm": 4.9375,
      "learning_rate": 9.102425388037527e-06,
      "loss": 1.58526535,
      "memory(GiB)": 97.17,
      "step": 18450,
      "train_speed(iter/s)": 1.633057
    },
    {
      "acc": 0.65277543,
      "epoch": 0.46816336884830034,
      "grad_norm": 5.6875,
      "learning_rate": 9.101825833278308e-06,
      "loss": 1.62531395,
      "memory(GiB)": 97.17,
      "step": 18455,
      "train_speed(iter/s)": 1.633106
    },
    {
      "acc": 0.65889072,
      "epoch": 0.4682902080162354,
      "grad_norm": 5.5625,
      "learning_rate": 9.101226098102473e-06,
      "loss": 1.56215038,
      "memory(GiB)": 97.17,
      "step": 18460,
      "train_speed(iter/s)": 1.633156
    },
    {
      "acc": 0.65938935,
      "epoch": 0.4684170471841705,
      "grad_norm": 8.0,
      "learning_rate": 9.100626182536405e-06,
      "loss": 1.64778728,
      "memory(GiB)": 97.17,
      "step": 18465,
      "train_speed(iter/s)": 1.633204
    },
    {
      "acc": 0.65993805,
      "epoch": 0.46854388635210553,
      "grad_norm": 5.84375,
      "learning_rate": 9.100026086606488e-06,
      "loss": 1.58330822,
      "memory(GiB)": 97.17,
      "step": 18470,
      "train_speed(iter/s)": 1.633255
    },
    {
      "acc": 0.67940564,
      "epoch": 0.4686707255200406,
      "grad_norm": 5.90625,
      "learning_rate": 9.09942581033912e-06,
      "loss": 1.5305685,
      "memory(GiB)": 97.17,
      "step": 18475,
      "train_speed(iter/s)": 1.633303
    },
    {
      "acc": 0.66762815,
      "epoch": 0.4687975646879756,
      "grad_norm": 5.90625,
      "learning_rate": 9.0988253537607e-06,
      "loss": 1.52450085,
      "memory(GiB)": 97.17,
      "step": 18480,
      "train_speed(iter/s)": 1.633352
    },
    {
      "acc": 0.66718826,
      "epoch": 0.4689244038559107,
      "grad_norm": 4.90625,
      "learning_rate": 9.098224716897644e-06,
      "loss": 1.59068661,
      "memory(GiB)": 97.17,
      "step": 18485,
      "train_speed(iter/s)": 1.633399
    },
    {
      "acc": 0.66398401,
      "epoch": 0.46905124302384577,
      "grad_norm": 5.6875,
      "learning_rate": 9.097623899776366e-06,
      "loss": 1.58663921,
      "memory(GiB)": 97.17,
      "step": 18490,
      "train_speed(iter/s)": 1.633448
    },
    {
      "acc": 0.64091969,
      "epoch": 0.4691780821917808,
      "grad_norm": 6.25,
      "learning_rate": 9.097022902423294e-06,
      "loss": 1.59719944,
      "memory(GiB)": 97.17,
      "step": 18495,
      "train_speed(iter/s)": 1.633498
    },
    {
      "acc": 0.65153608,
      "epoch": 0.46930492135971585,
      "grad_norm": 5.21875,
      "learning_rate": 9.096421724864864e-06,
      "loss": 1.63566704,
      "memory(GiB)": 97.17,
      "step": 18500,
      "train_speed(iter/s)": 1.633546
    },
    {
      "acc": 0.65336642,
      "epoch": 0.46943176052765095,
      "grad_norm": 5.28125,
      "learning_rate": 9.095820367127517e-06,
      "loss": 1.64004707,
      "memory(GiB)": 97.17,
      "step": 18505,
      "train_speed(iter/s)": 1.633593
    },
    {
      "acc": 0.64430351,
      "epoch": 0.469558599695586,
      "grad_norm": 5.5,
      "learning_rate": 9.095218829237703e-06,
      "loss": 1.64820118,
      "memory(GiB)": 97.17,
      "step": 18510,
      "train_speed(iter/s)": 1.633642
    },
    {
      "acc": 0.63728056,
      "epoch": 0.46968543886352104,
      "grad_norm": 5.5625,
      "learning_rate": 9.094617111221881e-06,
      "loss": 1.6593441,
      "memory(GiB)": 97.17,
      "step": 18515,
      "train_speed(iter/s)": 1.63369
    },
    {
      "acc": 0.64452839,
      "epoch": 0.4698122780314561,
      "grad_norm": 4.65625,
      "learning_rate": 9.09401521310652e-06,
      "loss": 1.58186817,
      "memory(GiB)": 97.17,
      "step": 18520,
      "train_speed(iter/s)": 1.633741
    },
    {
      "acc": 0.67075787,
      "epoch": 0.4699391171993912,
      "grad_norm": 5.1875,
      "learning_rate": 9.093413134918088e-06,
      "loss": 1.55130606,
      "memory(GiB)": 97.17,
      "step": 18525,
      "train_speed(iter/s)": 1.633792
    },
    {
      "acc": 0.66500759,
      "epoch": 0.47006595636732623,
      "grad_norm": 5.21875,
      "learning_rate": 9.09281087668307e-06,
      "loss": 1.60559177,
      "memory(GiB)": 97.17,
      "step": 18530,
      "train_speed(iter/s)": 1.63384
    },
    {
      "acc": 0.6658442,
      "epoch": 0.4701927955352613,
      "grad_norm": 6.09375,
      "learning_rate": 9.09220843842796e-06,
      "loss": 1.58783302,
      "memory(GiB)": 97.17,
      "step": 18535,
      "train_speed(iter/s)": 1.63389
    },
    {
      "acc": 0.64500914,
      "epoch": 0.4703196347031963,
      "grad_norm": 5.09375,
      "learning_rate": 9.09160582017925e-06,
      "loss": 1.60298882,
      "memory(GiB)": 97.17,
      "step": 18540,
      "train_speed(iter/s)": 1.63394
    },
    {
      "acc": 0.63264341,
      "epoch": 0.4704464738711314,
      "grad_norm": 5.25,
      "learning_rate": 9.091003021963449e-06,
      "loss": 1.68952332,
      "memory(GiB)": 97.17,
      "step": 18545,
      "train_speed(iter/s)": 1.633991
    },
    {
      "acc": 0.65284557,
      "epoch": 0.47057331303906647,
      "grad_norm": 5.5,
      "learning_rate": 9.09040004380707e-06,
      "loss": 1.6526825,
      "memory(GiB)": 97.17,
      "step": 18550,
      "train_speed(iter/s)": 1.634037
    },
    {
      "acc": 0.64369583,
      "epoch": 0.4707001522070015,
      "grad_norm": 5.46875,
      "learning_rate": 9.089796885736637e-06,
      "loss": 1.62356949,
      "memory(GiB)": 97.17,
      "step": 18555,
      "train_speed(iter/s)": 1.634085
    },
    {
      "acc": 0.66493397,
      "epoch": 0.47082699137493655,
      "grad_norm": 4.71875,
      "learning_rate": 9.089193547778674e-06,
      "loss": 1.5830575,
      "memory(GiB)": 97.17,
      "step": 18560,
      "train_speed(iter/s)": 1.634132
    },
    {
      "acc": 0.64886351,
      "epoch": 0.47095383054287165,
      "grad_norm": 5.375,
      "learning_rate": 9.088590029959724e-06,
      "loss": 1.604776,
      "memory(GiB)": 97.17,
      "step": 18565,
      "train_speed(iter/s)": 1.63418
    },
    {
      "acc": 0.64658508,
      "epoch": 0.4710806697108067,
      "grad_norm": 6.125,
      "learning_rate": 9.08798633230633e-06,
      "loss": 1.70517406,
      "memory(GiB)": 97.17,
      "step": 18570,
      "train_speed(iter/s)": 1.634228
    },
    {
      "acc": 0.66426773,
      "epoch": 0.47120750887874174,
      "grad_norm": 4.875,
      "learning_rate": 9.087382454845044e-06,
      "loss": 1.66563606,
      "memory(GiB)": 97.17,
      "step": 18575,
      "train_speed(iter/s)": 1.634276
    },
    {
      "acc": 0.64232955,
      "epoch": 0.4713343480466768,
      "grad_norm": 5.15625,
      "learning_rate": 9.08677839760243e-06,
      "loss": 1.67427292,
      "memory(GiB)": 97.17,
      "step": 18580,
      "train_speed(iter/s)": 1.634327
    },
    {
      "acc": 0.63752437,
      "epoch": 0.4714611872146119,
      "grad_norm": 5.5625,
      "learning_rate": 9.086174160605055e-06,
      "loss": 1.62534065,
      "memory(GiB)": 97.17,
      "step": 18585,
      "train_speed(iter/s)": 1.634379
    },
    {
      "acc": 0.64555354,
      "epoch": 0.47158802638254693,
      "grad_norm": 4.84375,
      "learning_rate": 9.085569743879498e-06,
      "loss": 1.65529881,
      "memory(GiB)": 97.17,
      "step": 18590,
      "train_speed(iter/s)": 1.634424
    },
    {
      "acc": 0.65081263,
      "epoch": 0.471714865550482,
      "grad_norm": 5.0,
      "learning_rate": 9.084965147452342e-06,
      "loss": 1.61330452,
      "memory(GiB)": 97.17,
      "step": 18595,
      "train_speed(iter/s)": 1.634474
    },
    {
      "acc": 0.65549164,
      "epoch": 0.471841704718417,
      "grad_norm": 5.15625,
      "learning_rate": 9.084360371350182e-06,
      "loss": 1.60776005,
      "memory(GiB)": 97.17,
      "step": 18600,
      "train_speed(iter/s)": 1.63452
    },
    {
      "acc": 0.65038986,
      "epoch": 0.4719685438863521,
      "grad_norm": 5.09375,
      "learning_rate": 9.083755415599617e-06,
      "loss": 1.61372547,
      "memory(GiB)": 97.17,
      "step": 18605,
      "train_speed(iter/s)": 1.63457
    },
    {
      "acc": 0.65071692,
      "epoch": 0.47209538305428717,
      "grad_norm": 5.65625,
      "learning_rate": 9.083150280227255e-06,
      "loss": 1.58352623,
      "memory(GiB)": 97.17,
      "step": 18610,
      "train_speed(iter/s)": 1.63462
    },
    {
      "acc": 0.65439391,
      "epoch": 0.4722222222222222,
      "grad_norm": 5.28125,
      "learning_rate": 9.082544965259716e-06,
      "loss": 1.59898672,
      "memory(GiB)": 97.17,
      "step": 18615,
      "train_speed(iter/s)": 1.634667
    },
    {
      "acc": 0.66267424,
      "epoch": 0.47234906139015725,
      "grad_norm": 5.34375,
      "learning_rate": 9.081939470723619e-06,
      "loss": 1.61046295,
      "memory(GiB)": 97.17,
      "step": 18620,
      "train_speed(iter/s)": 1.634718
    },
    {
      "acc": 0.63989754,
      "epoch": 0.47247590055809235,
      "grad_norm": 6.5,
      "learning_rate": 9.081333796645603e-06,
      "loss": 1.62868729,
      "memory(GiB)": 97.17,
      "step": 18625,
      "train_speed(iter/s)": 1.63477
    },
    {
      "acc": 0.65537705,
      "epoch": 0.4726027397260274,
      "grad_norm": 6.0625,
      "learning_rate": 9.080727943052304e-06,
      "loss": 1.59164009,
      "memory(GiB)": 97.17,
      "step": 18630,
      "train_speed(iter/s)": 1.634822
    },
    {
      "acc": 0.64680939,
      "epoch": 0.47272957889396244,
      "grad_norm": 5.71875,
      "learning_rate": 9.080121909970369e-06,
      "loss": 1.63066101,
      "memory(GiB)": 97.17,
      "step": 18635,
      "train_speed(iter/s)": 1.634869
    },
    {
      "acc": 0.65633183,
      "epoch": 0.4728564180618975,
      "grad_norm": 6.4375,
      "learning_rate": 9.079515697426457e-06,
      "loss": 1.57321014,
      "memory(GiB)": 97.17,
      "step": 18640,
      "train_speed(iter/s)": 1.634918
    },
    {
      "acc": 0.65313835,
      "epoch": 0.4729832572298326,
      "grad_norm": 5.84375,
      "learning_rate": 9.078909305447231e-06,
      "loss": 1.62922554,
      "memory(GiB)": 97.17,
      "step": 18645,
      "train_speed(iter/s)": 1.634968
    },
    {
      "acc": 0.66686578,
      "epoch": 0.47311009639776763,
      "grad_norm": 5.1875,
      "learning_rate": 9.078302734059363e-06,
      "loss": 1.5349041,
      "memory(GiB)": 97.17,
      "step": 18650,
      "train_speed(iter/s)": 1.635018
    },
    {
      "acc": 0.66314716,
      "epoch": 0.4732369355657027,
      "grad_norm": 6.1875,
      "learning_rate": 9.077695983289531e-06,
      "loss": 1.60294647,
      "memory(GiB)": 97.17,
      "step": 18655,
      "train_speed(iter/s)": 1.635065
    },
    {
      "acc": 0.64946804,
      "epoch": 0.4733637747336377,
      "grad_norm": 4.90625,
      "learning_rate": 9.077089053164426e-06,
      "loss": 1.62236404,
      "memory(GiB)": 97.17,
      "step": 18660,
      "train_speed(iter/s)": 1.635116
    },
    {
      "acc": 0.65981016,
      "epoch": 0.4734906139015728,
      "grad_norm": 4.875,
      "learning_rate": 9.076481943710742e-06,
      "loss": 1.56549664,
      "memory(GiB)": 97.17,
      "step": 18665,
      "train_speed(iter/s)": 1.635163
    },
    {
      "acc": 0.65033021,
      "epoch": 0.47361745306950787,
      "grad_norm": 6.3125,
      "learning_rate": 9.075874654955178e-06,
      "loss": 1.61561012,
      "memory(GiB)": 97.17,
      "step": 18670,
      "train_speed(iter/s)": 1.635214
    },
    {
      "acc": 0.64855671,
      "epoch": 0.4737442922374429,
      "grad_norm": 4.6875,
      "learning_rate": 9.075267186924453e-06,
      "loss": 1.65048275,
      "memory(GiB)": 97.17,
      "step": 18675,
      "train_speed(iter/s)": 1.635263
    },
    {
      "acc": 0.64401731,
      "epoch": 0.47387113140537795,
      "grad_norm": 5.53125,
      "learning_rate": 9.074659539645281e-06,
      "loss": 1.60564995,
      "memory(GiB)": 97.17,
      "step": 18680,
      "train_speed(iter/s)": 1.63531
    },
    {
      "acc": 0.64505358,
      "epoch": 0.47399797057331305,
      "grad_norm": 5.8125,
      "learning_rate": 9.074051713144392e-06,
      "loss": 1.69649239,
      "memory(GiB)": 97.17,
      "step": 18685,
      "train_speed(iter/s)": 1.635362
    },
    {
      "acc": 0.65669184,
      "epoch": 0.4741248097412481,
      "grad_norm": 6.03125,
      "learning_rate": 9.073443707448517e-06,
      "loss": 1.63959579,
      "memory(GiB)": 97.17,
      "step": 18690,
      "train_speed(iter/s)": 1.635412
    },
    {
      "acc": 0.64006977,
      "epoch": 0.47425164890918314,
      "grad_norm": 6.625,
      "learning_rate": 9.072835522584402e-06,
      "loss": 1.67402325,
      "memory(GiB)": 97.17,
      "step": 18695,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.66570463,
      "epoch": 0.4743784880771182,
      "grad_norm": 6.25,
      "learning_rate": 9.072227158578798e-06,
      "loss": 1.56113482,
      "memory(GiB)": 97.17,
      "step": 18700,
      "train_speed(iter/s)": 1.635509
    },
    {
      "acc": 0.64029751,
      "epoch": 0.4745053272450533,
      "grad_norm": 6.59375,
      "learning_rate": 9.07161861545846e-06,
      "loss": 1.61435833,
      "memory(GiB)": 97.17,
      "step": 18705,
      "train_speed(iter/s)": 1.635557
    },
    {
      "acc": 0.64389362,
      "epoch": 0.47463216641298833,
      "grad_norm": 7.09375,
      "learning_rate": 9.071009893250158e-06,
      "loss": 1.70898781,
      "memory(GiB)": 97.17,
      "step": 18710,
      "train_speed(iter/s)": 1.635608
    },
    {
      "acc": 0.64417443,
      "epoch": 0.4747590055809234,
      "grad_norm": 6.15625,
      "learning_rate": 9.070400991980666e-06,
      "loss": 1.64150505,
      "memory(GiB)": 97.17,
      "step": 18715,
      "train_speed(iter/s)": 1.635656
    },
    {
      "acc": 0.66944103,
      "epoch": 0.4748858447488584,
      "grad_norm": 5.28125,
      "learning_rate": 9.069791911676765e-06,
      "loss": 1.60371284,
      "memory(GiB)": 97.17,
      "step": 18720,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.669384,
      "epoch": 0.4750126839167935,
      "grad_norm": 5.59375,
      "learning_rate": 9.069182652365245e-06,
      "loss": 1.57409821,
      "memory(GiB)": 97.17,
      "step": 18725,
      "train_speed(iter/s)": 1.635755
    },
    {
      "acc": 0.6405951,
      "epoch": 0.47513952308472857,
      "grad_norm": 6.03125,
      "learning_rate": 9.068573214072904e-06,
      "loss": 1.71492043,
      "memory(GiB)": 97.17,
      "step": 18730,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.64446325,
      "epoch": 0.4752663622526636,
      "grad_norm": 4.6875,
      "learning_rate": 9.067963596826547e-06,
      "loss": 1.65332108,
      "memory(GiB)": 97.17,
      "step": 18735,
      "train_speed(iter/s)": 1.635854
    },
    {
      "acc": 0.65595675,
      "epoch": 0.47539320142059865,
      "grad_norm": 5.25,
      "learning_rate": 9.067353800652991e-06,
      "loss": 1.64717388,
      "memory(GiB)": 97.17,
      "step": 18740,
      "train_speed(iter/s)": 1.635903
    },
    {
      "acc": 0.65872364,
      "epoch": 0.47552004058853375,
      "grad_norm": 5.75,
      "learning_rate": 9.066743825579056e-06,
      "loss": 1.61537857,
      "memory(GiB)": 97.17,
      "step": 18745,
      "train_speed(iter/s)": 1.635951
    },
    {
      "acc": 0.66051412,
      "epoch": 0.4756468797564688,
      "grad_norm": 6.21875,
      "learning_rate": 9.06613367163157e-06,
      "loss": 1.54608269,
      "memory(GiB)": 97.17,
      "step": 18750,
      "train_speed(iter/s)": 1.636001
    },
    {
      "acc": 0.6519444,
      "epoch": 0.47577371892440384,
      "grad_norm": 5.75,
      "learning_rate": 9.06552333883737e-06,
      "loss": 1.61317806,
      "memory(GiB)": 97.17,
      "step": 18755,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.65443516,
      "epoch": 0.4759005580923389,
      "grad_norm": 7.0,
      "learning_rate": 9.064912827223303e-06,
      "loss": 1.62242889,
      "memory(GiB)": 97.17,
      "step": 18760,
      "train_speed(iter/s)": 1.636098
    },
    {
      "acc": 0.65507555,
      "epoch": 0.476027397260274,
      "grad_norm": 5.21875,
      "learning_rate": 9.06430213681622e-06,
      "loss": 1.61009979,
      "memory(GiB)": 97.17,
      "step": 18765,
      "train_speed(iter/s)": 1.636147
    },
    {
      "acc": 0.63461132,
      "epoch": 0.47615423642820903,
      "grad_norm": 5.1875,
      "learning_rate": 9.063691267642987e-06,
      "loss": 1.70338039,
      "memory(GiB)": 97.17,
      "step": 18770,
      "train_speed(iter/s)": 1.636198
    },
    {
      "acc": 0.65015697,
      "epoch": 0.4762810755961441,
      "grad_norm": 5.625,
      "learning_rate": 9.063080219730467e-06,
      "loss": 1.64285851,
      "memory(GiB)": 97.17,
      "step": 18775,
      "train_speed(iter/s)": 1.636251
    },
    {
      "acc": 0.64973965,
      "epoch": 0.4764079147640791,
      "grad_norm": 5.84375,
      "learning_rate": 9.062468993105538e-06,
      "loss": 1.61367493,
      "memory(GiB)": 97.17,
      "step": 18780,
      "train_speed(iter/s)": 1.636299
    },
    {
      "acc": 0.67192078,
      "epoch": 0.4765347539320142,
      "grad_norm": 5.75,
      "learning_rate": 9.061857587795084e-06,
      "loss": 1.52915649,
      "memory(GiB)": 97.17,
      "step": 18785,
      "train_speed(iter/s)": 1.636349
    },
    {
      "acc": 0.65035758,
      "epoch": 0.47666159309994927,
      "grad_norm": 6.0,
      "learning_rate": 9.061246003826e-06,
      "loss": 1.71583366,
      "memory(GiB)": 97.17,
      "step": 18790,
      "train_speed(iter/s)": 1.636399
    },
    {
      "acc": 0.64987903,
      "epoch": 0.4767884322678843,
      "grad_norm": 6.125,
      "learning_rate": 9.060634241225184e-06,
      "loss": 1.64932785,
      "memory(GiB)": 97.17,
      "step": 18795,
      "train_speed(iter/s)": 1.636446
    },
    {
      "acc": 0.65373144,
      "epoch": 0.47691527143581935,
      "grad_norm": 5.65625,
      "learning_rate": 9.060022300019546e-06,
      "loss": 1.61680775,
      "memory(GiB)": 97.17,
      "step": 18800,
      "train_speed(iter/s)": 1.636499
    },
    {
      "acc": 0.65803919,
      "epoch": 0.47704211060375445,
      "grad_norm": 5.875,
      "learning_rate": 9.059410180236e-06,
      "loss": 1.57195053,
      "memory(GiB)": 97.17,
      "step": 18805,
      "train_speed(iter/s)": 1.636546
    },
    {
      "acc": 0.65733013,
      "epoch": 0.4771689497716895,
      "grad_norm": 4.90625,
      "learning_rate": 9.058797881901469e-06,
      "loss": 1.5844697,
      "memory(GiB)": 97.17,
      "step": 18810,
      "train_speed(iter/s)": 1.636592
    },
    {
      "acc": 0.6689486,
      "epoch": 0.47729578893962454,
      "grad_norm": 6.34375,
      "learning_rate": 9.058185405042886e-06,
      "loss": 1.5118207,
      "memory(GiB)": 97.17,
      "step": 18815,
      "train_speed(iter/s)": 1.636642
    },
    {
      "acc": 0.65490046,
      "epoch": 0.4774226281075596,
      "grad_norm": 5.21875,
      "learning_rate": 9.05757274968719e-06,
      "loss": 1.69743958,
      "memory(GiB)": 97.17,
      "step": 18820,
      "train_speed(iter/s)": 1.636688
    },
    {
      "acc": 0.65074377,
      "epoch": 0.4775494672754947,
      "grad_norm": 6.09375,
      "learning_rate": 9.056959915861331e-06,
      "loss": 1.58289957,
      "memory(GiB)": 97.17,
      "step": 18825,
      "train_speed(iter/s)": 1.636736
    },
    {
      "acc": 0.64664888,
      "epoch": 0.47767630644342973,
      "grad_norm": 6.4375,
      "learning_rate": 9.056346903592262e-06,
      "loss": 1.6505188,
      "memory(GiB)": 97.17,
      "step": 18830,
      "train_speed(iter/s)": 1.636784
    },
    {
      "acc": 0.64677014,
      "epoch": 0.4778031456113648,
      "grad_norm": 4.84375,
      "learning_rate": 9.055733712906943e-06,
      "loss": 1.61426468,
      "memory(GiB)": 97.17,
      "step": 18835,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.64425516,
      "epoch": 0.4779299847792998,
      "grad_norm": 5.84375,
      "learning_rate": 9.05512034383235e-06,
      "loss": 1.62231998,
      "memory(GiB)": 97.17,
      "step": 18840,
      "train_speed(iter/s)": 1.63688
    },
    {
      "acc": 0.63516402,
      "epoch": 0.4780568239472349,
      "grad_norm": 5.53125,
      "learning_rate": 9.054506796395458e-06,
      "loss": 1.64320946,
      "memory(GiB)": 97.17,
      "step": 18845,
      "train_speed(iter/s)": 1.636929
    },
    {
      "acc": 0.65735517,
      "epoch": 0.47818366311516997,
      "grad_norm": 5.75,
      "learning_rate": 9.053893070623256e-06,
      "loss": 1.65515213,
      "memory(GiB)": 97.17,
      "step": 18850,
      "train_speed(iter/s)": 1.636978
    },
    {
      "acc": 0.66131325,
      "epoch": 0.478310502283105,
      "grad_norm": 6.09375,
      "learning_rate": 9.053279166542738e-06,
      "loss": 1.63779106,
      "memory(GiB)": 97.17,
      "step": 18855,
      "train_speed(iter/s)": 1.637027
    },
    {
      "acc": 0.66859837,
      "epoch": 0.47843734145104005,
      "grad_norm": 5.84375,
      "learning_rate": 9.052665084180906e-06,
      "loss": 1.58122597,
      "memory(GiB)": 97.17,
      "step": 18860,
      "train_speed(iter/s)": 1.637074
    },
    {
      "acc": 0.64981456,
      "epoch": 0.47856418061897515,
      "grad_norm": 5.5,
      "learning_rate": 9.052050823564767e-06,
      "loss": 1.63787308,
      "memory(GiB)": 97.17,
      "step": 18865,
      "train_speed(iter/s)": 1.637123
    },
    {
      "acc": 0.66255226,
      "epoch": 0.4786910197869102,
      "grad_norm": 4.9375,
      "learning_rate": 9.051436384721344e-06,
      "loss": 1.59685802,
      "memory(GiB)": 97.17,
      "step": 18870,
      "train_speed(iter/s)": 1.637172
    },
    {
      "acc": 0.67633996,
      "epoch": 0.47881785895484524,
      "grad_norm": 5.4375,
      "learning_rate": 9.05082176767766e-06,
      "loss": 1.52323971,
      "memory(GiB)": 97.17,
      "step": 18875,
      "train_speed(iter/s)": 1.637223
    },
    {
      "acc": 0.6529479,
      "epoch": 0.4789446981227803,
      "grad_norm": 4.5625,
      "learning_rate": 9.050206972460749e-06,
      "loss": 1.61393414,
      "memory(GiB)": 97.17,
      "step": 18880,
      "train_speed(iter/s)": 1.637273
    },
    {
      "acc": 0.65651464,
      "epoch": 0.4790715372907154,
      "grad_norm": 6.09375,
      "learning_rate": 9.049591999097651e-06,
      "loss": 1.65935745,
      "memory(GiB)": 97.17,
      "step": 18885,
      "train_speed(iter/s)": 1.637323
    },
    {
      "acc": 0.63807549,
      "epoch": 0.47919837645865043,
      "grad_norm": 5.96875,
      "learning_rate": 9.048976847615418e-06,
      "loss": 1.65334721,
      "memory(GiB)": 97.17,
      "step": 18890,
      "train_speed(iter/s)": 1.637373
    },
    {
      "acc": 0.64474735,
      "epoch": 0.4793252156265855,
      "grad_norm": 5.09375,
      "learning_rate": 9.048361518041107e-06,
      "loss": 1.64946785,
      "memory(GiB)": 97.17,
      "step": 18895,
      "train_speed(iter/s)": 1.63742
    },
    {
      "acc": 0.66650367,
      "epoch": 0.4794520547945205,
      "grad_norm": 4.8125,
      "learning_rate": 9.04774601040178e-06,
      "loss": 1.53804979,
      "memory(GiB)": 97.17,
      "step": 18900,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.64678864,
      "epoch": 0.4795788939624556,
      "grad_norm": 7.34375,
      "learning_rate": 9.047130324724513e-06,
      "loss": 1.61122131,
      "memory(GiB)": 97.17,
      "step": 18905,
      "train_speed(iter/s)": 1.637519
    },
    {
      "acc": 0.64797907,
      "epoch": 0.47970573313039067,
      "grad_norm": 8.5,
      "learning_rate": 9.046514461036385e-06,
      "loss": 1.63961678,
      "memory(GiB)": 97.17,
      "step": 18910,
      "train_speed(iter/s)": 1.637561
    },
    {
      "acc": 0.64692249,
      "epoch": 0.4798325722983257,
      "grad_norm": 4.90625,
      "learning_rate": 9.045898419364483e-06,
      "loss": 1.6343441,
      "memory(GiB)": 97.17,
      "step": 18915,
      "train_speed(iter/s)": 1.637611
    },
    {
      "acc": 0.6598197,
      "epoch": 0.47995941146626075,
      "grad_norm": 4.5,
      "learning_rate": 9.045282199735906e-06,
      "loss": 1.58556423,
      "memory(GiB)": 97.17,
      "step": 18920,
      "train_speed(iter/s)": 1.63766
    },
    {
      "acc": 0.63780127,
      "epoch": 0.48008625063419585,
      "grad_norm": 5.90625,
      "learning_rate": 9.044665802177756e-06,
      "loss": 1.75557442,
      "memory(GiB)": 97.17,
      "step": 18925,
      "train_speed(iter/s)": 1.637705
    },
    {
      "acc": 0.65323682,
      "epoch": 0.4802130898021309,
      "grad_norm": 9.0625,
      "learning_rate": 9.044049226717148e-06,
      "loss": 1.60862904,
      "memory(GiB)": 97.17,
      "step": 18930,
      "train_speed(iter/s)": 1.637752
    },
    {
      "acc": 0.6711524,
      "epoch": 0.48033992897006594,
      "grad_norm": 5.375,
      "learning_rate": 9.043432473381198e-06,
      "loss": 1.58349142,
      "memory(GiB)": 97.17,
      "step": 18935,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.6529037,
      "epoch": 0.480466768138001,
      "grad_norm": 5.78125,
      "learning_rate": 9.042815542197037e-06,
      "loss": 1.64929562,
      "memory(GiB)": 97.17,
      "step": 18940,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.64694366,
      "epoch": 0.4805936073059361,
      "grad_norm": 5.4375,
      "learning_rate": 9.042198433191796e-06,
      "loss": 1.57790012,
      "memory(GiB)": 97.17,
      "step": 18945,
      "train_speed(iter/s)": 1.637896
    },
    {
      "acc": 0.64720192,
      "epoch": 0.48072044647387113,
      "grad_norm": 6.5625,
      "learning_rate": 9.041581146392621e-06,
      "loss": 1.65945187,
      "memory(GiB)": 97.17,
      "step": 18950,
      "train_speed(iter/s)": 1.637943
    },
    {
      "acc": 0.64622927,
      "epoch": 0.4808472856418062,
      "grad_norm": 6.21875,
      "learning_rate": 9.040963681826665e-06,
      "loss": 1.5980423,
      "memory(GiB)": 97.17,
      "step": 18955,
      "train_speed(iter/s)": 1.637996
    },
    {
      "acc": 0.65423183,
      "epoch": 0.4809741248097412,
      "grad_norm": 5.71875,
      "learning_rate": 9.040346039521085e-06,
      "loss": 1.63318958,
      "memory(GiB)": 97.17,
      "step": 18960,
      "train_speed(iter/s)": 1.638046
    },
    {
      "acc": 0.66483846,
      "epoch": 0.4811009639776763,
      "grad_norm": 5.0625,
      "learning_rate": 9.039728219503044e-06,
      "loss": 1.56115227,
      "memory(GiB)": 97.17,
      "step": 18965,
      "train_speed(iter/s)": 1.638095
    },
    {
      "acc": 0.66229806,
      "epoch": 0.48122780314561137,
      "grad_norm": 4.9375,
      "learning_rate": 9.039110221799721e-06,
      "loss": 1.57973385,
      "memory(GiB)": 97.17,
      "step": 18970,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.65636435,
      "epoch": 0.4813546423135464,
      "grad_norm": 5.6875,
      "learning_rate": 9.038492046438298e-06,
      "loss": 1.62687855,
      "memory(GiB)": 97.17,
      "step": 18975,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.65604401,
      "epoch": 0.48148148148148145,
      "grad_norm": 6.4375,
      "learning_rate": 9.037873693445965e-06,
      "loss": 1.63516674,
      "memory(GiB)": 97.17,
      "step": 18980,
      "train_speed(iter/s)": 1.63824
    },
    {
      "acc": 0.65956035,
      "epoch": 0.48160832064941655,
      "grad_norm": 5.125,
      "learning_rate": 9.037255162849918e-06,
      "loss": 1.61233826,
      "memory(GiB)": 97.17,
      "step": 18985,
      "train_speed(iter/s)": 1.63829
    },
    {
      "acc": 0.65382061,
      "epoch": 0.4817351598173516,
      "grad_norm": 5.90625,
      "learning_rate": 9.036636454677363e-06,
      "loss": 1.60012722,
      "memory(GiB)": 97.17,
      "step": 18990,
      "train_speed(iter/s)": 1.63834
    },
    {
      "acc": 0.64359808,
      "epoch": 0.48186199898528664,
      "grad_norm": 6.0625,
      "learning_rate": 9.036017568955516e-06,
      "loss": 1.63891335,
      "memory(GiB)": 97.17,
      "step": 18995,
      "train_speed(iter/s)": 1.638387
    },
    {
      "acc": 0.65544543,
      "epoch": 0.4819888381532217,
      "grad_norm": 4.5,
      "learning_rate": 9.035398505711597e-06,
      "loss": 1.64487362,
      "memory(GiB)": 97.17,
      "step": 19000,
      "train_speed(iter/s)": 1.638436
    },
    {
      "epoch": 0.4819888381532217,
      "eval_acc": 0.6435002516408865,
      "eval_loss": 1.5891553163528442,
      "eval_runtime": 58.1883,
      "eval_samples_per_second": 109.472,
      "eval_steps_per_second": 27.377,
      "step": 19000
    },
    {
      "acc": 0.65876441,
      "epoch": 0.4821156773211568,
      "grad_norm": 6.125,
      "learning_rate": 9.034779264972834e-06,
      "loss": 1.55358868,
      "memory(GiB)": 97.17,
      "step": 19005,
      "train_speed(iter/s)": 1.629709
    },
    {
      "acc": 0.66507697,
      "epoch": 0.48224251648909183,
      "grad_norm": 5.15625,
      "learning_rate": 9.034159846766464e-06,
      "loss": 1.59579582,
      "memory(GiB)": 97.17,
      "step": 19010,
      "train_speed(iter/s)": 1.62976
    },
    {
      "acc": 0.65994821,
      "epoch": 0.4823693556570269,
      "grad_norm": 5.8125,
      "learning_rate": 9.033540251119734e-06,
      "loss": 1.5259758,
      "memory(GiB)": 97.17,
      "step": 19015,
      "train_speed(iter/s)": 1.629813
    },
    {
      "acc": 0.6565865,
      "epoch": 0.4824961948249619,
      "grad_norm": 5.125,
      "learning_rate": 9.032920478059897e-06,
      "loss": 1.57405233,
      "memory(GiB)": 97.17,
      "step": 19020,
      "train_speed(iter/s)": 1.629862
    },
    {
      "acc": 0.65229812,
      "epoch": 0.482623033992897,
      "grad_norm": 5.25,
      "learning_rate": 9.032300527614209e-06,
      "loss": 1.67209702,
      "memory(GiB)": 97.17,
      "step": 19025,
      "train_speed(iter/s)": 1.629915
    },
    {
      "acc": 0.64681187,
      "epoch": 0.48274987316083207,
      "grad_norm": 5.875,
      "learning_rate": 9.031680399809941e-06,
      "loss": 1.66641884,
      "memory(GiB)": 97.17,
      "step": 19030,
      "train_speed(iter/s)": 1.629967
    },
    {
      "acc": 0.63678856,
      "epoch": 0.4828767123287671,
      "grad_norm": 5.03125,
      "learning_rate": 9.031060094674371e-06,
      "loss": 1.70992813,
      "memory(GiB)": 97.17,
      "step": 19035,
      "train_speed(iter/s)": 1.630018
    },
    {
      "acc": 0.64104605,
      "epoch": 0.48300355149670215,
      "grad_norm": 7.21875,
      "learning_rate": 9.030439612234778e-06,
      "loss": 1.60441246,
      "memory(GiB)": 97.17,
      "step": 19040,
      "train_speed(iter/s)": 1.630072
    },
    {
      "acc": 0.66000285,
      "epoch": 0.48313039066463725,
      "grad_norm": 5.03125,
      "learning_rate": 9.029818952518458e-06,
      "loss": 1.59520874,
      "memory(GiB)": 97.17,
      "step": 19045,
      "train_speed(iter/s)": 1.630118
    },
    {
      "acc": 0.67154431,
      "epoch": 0.4832572298325723,
      "grad_norm": 5.34375,
      "learning_rate": 9.029198115552708e-06,
      "loss": 1.54623356,
      "memory(GiB)": 97.17,
      "step": 19050,
      "train_speed(iter/s)": 1.630168
    },
    {
      "acc": 0.64939351,
      "epoch": 0.48338406900050734,
      "grad_norm": 5.4375,
      "learning_rate": 9.028577101364837e-06,
      "loss": 1.59926214,
      "memory(GiB)": 97.17,
      "step": 19055,
      "train_speed(iter/s)": 1.630218
    },
    {
      "acc": 0.66098413,
      "epoch": 0.4835109081684424,
      "grad_norm": 5.78125,
      "learning_rate": 9.027955909982157e-06,
      "loss": 1.58970289,
      "memory(GiB)": 97.17,
      "step": 19060,
      "train_speed(iter/s)": 1.630268
    },
    {
      "acc": 0.68105259,
      "epoch": 0.4836377473363775,
      "grad_norm": 5.0625,
      "learning_rate": 9.027334541431993e-06,
      "loss": 1.49056139,
      "memory(GiB)": 97.17,
      "step": 19065,
      "train_speed(iter/s)": 1.630313
    },
    {
      "acc": 0.65410118,
      "epoch": 0.48376458650431253,
      "grad_norm": 7.65625,
      "learning_rate": 9.026712995741676e-06,
      "loss": 1.59510307,
      "memory(GiB)": 97.17,
      "step": 19070,
      "train_speed(iter/s)": 1.630361
    },
    {
      "acc": 0.63339963,
      "epoch": 0.4838914256722476,
      "grad_norm": 5.71875,
      "learning_rate": 9.026091272938543e-06,
      "loss": 1.66985054,
      "memory(GiB)": 97.17,
      "step": 19075,
      "train_speed(iter/s)": 1.630409
    },
    {
      "acc": 0.65780506,
      "epoch": 0.4840182648401826,
      "grad_norm": 5.09375,
      "learning_rate": 9.02546937304994e-06,
      "loss": 1.59855032,
      "memory(GiB)": 97.17,
      "step": 19080,
      "train_speed(iter/s)": 1.630461
    },
    {
      "acc": 0.65102997,
      "epoch": 0.4841451040081177,
      "grad_norm": 5.5625,
      "learning_rate": 9.02484729610322e-06,
      "loss": 1.64531765,
      "memory(GiB)": 97.17,
      "step": 19085,
      "train_speed(iter/s)": 1.630509
    },
    {
      "acc": 0.64340091,
      "epoch": 0.48427194317605277,
      "grad_norm": 5.65625,
      "learning_rate": 9.02422504212575e-06,
      "loss": 1.68798866,
      "memory(GiB)": 97.17,
      "step": 19090,
      "train_speed(iter/s)": 1.63056
    },
    {
      "acc": 0.66295605,
      "epoch": 0.4843987823439878,
      "grad_norm": 4.90625,
      "learning_rate": 9.023602611144893e-06,
      "loss": 1.59011669,
      "memory(GiB)": 97.17,
      "step": 19095,
      "train_speed(iter/s)": 1.63061
    },
    {
      "acc": 0.64069386,
      "epoch": 0.48452562151192285,
      "grad_norm": 5.0625,
      "learning_rate": 9.02298000318803e-06,
      "loss": 1.6068161,
      "memory(GiB)": 97.17,
      "step": 19100,
      "train_speed(iter/s)": 1.630656
    },
    {
      "acc": 0.64794445,
      "epoch": 0.48465246067985795,
      "grad_norm": 5.75,
      "learning_rate": 9.022357218282546e-06,
      "loss": 1.60515022,
      "memory(GiB)": 97.17,
      "step": 19105,
      "train_speed(iter/s)": 1.630703
    },
    {
      "acc": 0.64888821,
      "epoch": 0.484779299847793,
      "grad_norm": 5.625,
      "learning_rate": 9.021734256455832e-06,
      "loss": 1.65507488,
      "memory(GiB)": 97.17,
      "step": 19110,
      "train_speed(iter/s)": 1.63075
    },
    {
      "acc": 0.68182416,
      "epoch": 0.48490613901572804,
      "grad_norm": 5.21875,
      "learning_rate": 9.02111111773529e-06,
      "loss": 1.537393,
      "memory(GiB)": 97.17,
      "step": 19115,
      "train_speed(iter/s)": 1.630797
    },
    {
      "acc": 0.65830941,
      "epoch": 0.4850329781836631,
      "grad_norm": 5.6875,
      "learning_rate": 9.020487802148328e-06,
      "loss": 1.54871597,
      "memory(GiB)": 97.17,
      "step": 19120,
      "train_speed(iter/s)": 1.630849
    },
    {
      "acc": 0.64492812,
      "epoch": 0.4851598173515982,
      "grad_norm": 5.46875,
      "learning_rate": 9.019864309722362e-06,
      "loss": 1.6327282,
      "memory(GiB)": 97.17,
      "step": 19125,
      "train_speed(iter/s)": 1.630897
    },
    {
      "acc": 0.65784407,
      "epoch": 0.48528665651953323,
      "grad_norm": 5.125,
      "learning_rate": 9.019240640484816e-06,
      "loss": 1.58626938,
      "memory(GiB)": 97.17,
      "step": 19130,
      "train_speed(iter/s)": 1.630944
    },
    {
      "acc": 0.65162892,
      "epoch": 0.4854134956874683,
      "grad_norm": 5.875,
      "learning_rate": 9.018616794463124e-06,
      "loss": 1.57989788,
      "memory(GiB)": 97.17,
      "step": 19135,
      "train_speed(iter/s)": 1.630993
    },
    {
      "acc": 0.64719133,
      "epoch": 0.4855403348554033,
      "grad_norm": 5.21875,
      "learning_rate": 9.017992771684722e-06,
      "loss": 1.65698051,
      "memory(GiB)": 97.17,
      "step": 19140,
      "train_speed(iter/s)": 1.631041
    },
    {
      "acc": 0.63881626,
      "epoch": 0.4856671740233384,
      "grad_norm": 8.0625,
      "learning_rate": 9.017368572177058e-06,
      "loss": 1.61152592,
      "memory(GiB)": 97.17,
      "step": 19145,
      "train_speed(iter/s)": 1.631085
    },
    {
      "acc": 0.64969001,
      "epoch": 0.48579401319127347,
      "grad_norm": 6.125,
      "learning_rate": 9.016744195967588e-06,
      "loss": 1.6037796,
      "memory(GiB)": 97.17,
      "step": 19150,
      "train_speed(iter/s)": 1.63113
    },
    {
      "acc": 0.66961908,
      "epoch": 0.4859208523592085,
      "grad_norm": 6.15625,
      "learning_rate": 9.016119643083777e-06,
      "loss": 1.64942417,
      "memory(GiB)": 97.17,
      "step": 19155,
      "train_speed(iter/s)": 1.631177
    },
    {
      "acc": 0.65805769,
      "epoch": 0.48604769152714355,
      "grad_norm": 6.09375,
      "learning_rate": 9.015494913553091e-06,
      "loss": 1.61547852,
      "memory(GiB)": 97.17,
      "step": 19160,
      "train_speed(iter/s)": 1.631228
    },
    {
      "acc": 0.64842701,
      "epoch": 0.48617453069507865,
      "grad_norm": 5.09375,
      "learning_rate": 9.014870007403012e-06,
      "loss": 1.61685257,
      "memory(GiB)": 97.17,
      "step": 19165,
      "train_speed(iter/s)": 1.631279
    },
    {
      "acc": 0.65189915,
      "epoch": 0.4863013698630137,
      "grad_norm": 6.53125,
      "learning_rate": 9.014244924661026e-06,
      "loss": 1.56398201,
      "memory(GiB)": 97.17,
      "step": 19170,
      "train_speed(iter/s)": 1.631329
    },
    {
      "acc": 0.65040245,
      "epoch": 0.48642820903094874,
      "grad_norm": 5.375,
      "learning_rate": 9.013619665354626e-06,
      "loss": 1.63303699,
      "memory(GiB)": 97.17,
      "step": 19175,
      "train_speed(iter/s)": 1.631376
    },
    {
      "acc": 0.64636779,
      "epoch": 0.4865550481988838,
      "grad_norm": 5.875,
      "learning_rate": 9.01299422951131e-06,
      "loss": 1.68049507,
      "memory(GiB)": 97.17,
      "step": 19180,
      "train_speed(iter/s)": 1.631428
    },
    {
      "acc": 0.64332356,
      "epoch": 0.4866818873668189,
      "grad_norm": 5.84375,
      "learning_rate": 9.012368617158593e-06,
      "loss": 1.65196228,
      "memory(GiB)": 97.17,
      "step": 19185,
      "train_speed(iter/s)": 1.63148
    },
    {
      "acc": 0.64770594,
      "epoch": 0.48680872653475393,
      "grad_norm": 5.3125,
      "learning_rate": 9.01174282832399e-06,
      "loss": 1.60900192,
      "memory(GiB)": 97.17,
      "step": 19190,
      "train_speed(iter/s)": 1.631528
    },
    {
      "acc": 0.66381068,
      "epoch": 0.486935565702689,
      "grad_norm": 5.65625,
      "learning_rate": 9.011116863035027e-06,
      "loss": 1.57273817,
      "memory(GiB)": 97.17,
      "step": 19195,
      "train_speed(iter/s)": 1.631575
    },
    {
      "acc": 0.64555984,
      "epoch": 0.487062404870624,
      "grad_norm": 5.59375,
      "learning_rate": 9.010490721319237e-06,
      "loss": 1.64356728,
      "memory(GiB)": 97.17,
      "step": 19200,
      "train_speed(iter/s)": 1.631623
    },
    {
      "acc": 0.64967284,
      "epoch": 0.4871892440385591,
      "grad_norm": 6.0625,
      "learning_rate": 9.009864403204157e-06,
      "loss": 1.60948715,
      "memory(GiB)": 97.17,
      "step": 19205,
      "train_speed(iter/s)": 1.631668
    },
    {
      "acc": 0.65512981,
      "epoch": 0.48731608320649417,
      "grad_norm": 5.0,
      "learning_rate": 9.00923790871734e-06,
      "loss": 1.57831182,
      "memory(GiB)": 97.17,
      "step": 19210,
      "train_speed(iter/s)": 1.631715
    },
    {
      "acc": 0.65080533,
      "epoch": 0.4874429223744292,
      "grad_norm": 6.375,
      "learning_rate": 9.008611237886339e-06,
      "loss": 1.65522251,
      "memory(GiB)": 97.17,
      "step": 19215,
      "train_speed(iter/s)": 1.631763
    },
    {
      "acc": 0.65541615,
      "epoch": 0.48756976154236426,
      "grad_norm": 6.53125,
      "learning_rate": 9.00798439073872e-06,
      "loss": 1.62413025,
      "memory(GiB)": 97.17,
      "step": 19220,
      "train_speed(iter/s)": 1.631808
    },
    {
      "acc": 0.65070553,
      "epoch": 0.48769660071029935,
      "grad_norm": 7.3125,
      "learning_rate": 9.007357367302052e-06,
      "loss": 1.62216053,
      "memory(GiB)": 97.17,
      "step": 19225,
      "train_speed(iter/s)": 1.631855
    },
    {
      "acc": 0.64507093,
      "epoch": 0.4878234398782344,
      "grad_norm": 5.46875,
      "learning_rate": 9.006730167603914e-06,
      "loss": 1.67657089,
      "memory(GiB)": 97.17,
      "step": 19230,
      "train_speed(iter/s)": 1.631902
    },
    {
      "acc": 0.65183091,
      "epoch": 0.48795027904616944,
      "grad_norm": 6.03125,
      "learning_rate": 9.006102791671896e-06,
      "loss": 1.63866215,
      "memory(GiB)": 97.17,
      "step": 19235,
      "train_speed(iter/s)": 1.631937
    },
    {
      "acc": 0.66255231,
      "epoch": 0.4880771182141045,
      "grad_norm": 5.96875,
      "learning_rate": 9.005475239533591e-06,
      "loss": 1.60908585,
      "memory(GiB)": 97.17,
      "step": 19240,
      "train_speed(iter/s)": 1.631985
    },
    {
      "acc": 0.64312086,
      "epoch": 0.4882039573820396,
      "grad_norm": 6.96875,
      "learning_rate": 9.0048475112166e-06,
      "loss": 1.62154655,
      "memory(GiB)": 97.17,
      "step": 19245,
      "train_speed(iter/s)": 1.632031
    },
    {
      "acc": 0.64979849,
      "epoch": 0.48833079654997463,
      "grad_norm": 5.5625,
      "learning_rate": 9.00421960674854e-06,
      "loss": 1.63101864,
      "memory(GiB)": 97.17,
      "step": 19250,
      "train_speed(iter/s)": 1.632079
    },
    {
      "acc": 0.64713917,
      "epoch": 0.4884576357179097,
      "grad_norm": 5.5625,
      "learning_rate": 9.003591526157021e-06,
      "loss": 1.65120621,
      "memory(GiB)": 97.17,
      "step": 19255,
      "train_speed(iter/s)": 1.632122
    },
    {
      "acc": 0.64968815,
      "epoch": 0.4885844748858447,
      "grad_norm": 6.15625,
      "learning_rate": 9.002963269469672e-06,
      "loss": 1.65066643,
      "memory(GiB)": 97.17,
      "step": 19260,
      "train_speed(iter/s)": 1.632165
    },
    {
      "acc": 0.66717105,
      "epoch": 0.4887113140537798,
      "grad_norm": 5.125,
      "learning_rate": 9.002334836714126e-06,
      "loss": 1.55717783,
      "memory(GiB)": 97.17,
      "step": 19265,
      "train_speed(iter/s)": 1.632207
    },
    {
      "acc": 0.65099764,
      "epoch": 0.48883815322171487,
      "grad_norm": 5.21875,
      "learning_rate": 9.001706227918023e-06,
      "loss": 1.60708771,
      "memory(GiB)": 97.17,
      "step": 19270,
      "train_speed(iter/s)": 1.632247
    },
    {
      "acc": 0.64583077,
      "epoch": 0.4889649923896499,
      "grad_norm": 5.21875,
      "learning_rate": 9.001077443109016e-06,
      "loss": 1.68178291,
      "memory(GiB)": 97.17,
      "step": 19275,
      "train_speed(iter/s)": 1.632295
    },
    {
      "acc": 0.64101601,
      "epoch": 0.48909183155758496,
      "grad_norm": 4.8125,
      "learning_rate": 9.00044848231476e-06,
      "loss": 1.63091698,
      "memory(GiB)": 97.17,
      "step": 19280,
      "train_speed(iter/s)": 1.63234
    },
    {
      "acc": 0.67217836,
      "epoch": 0.48921867072552006,
      "grad_norm": 4.90625,
      "learning_rate": 8.999819345562919e-06,
      "loss": 1.55919819,
      "memory(GiB)": 97.17,
      "step": 19285,
      "train_speed(iter/s)": 1.632386
    },
    {
      "acc": 0.65007591,
      "epoch": 0.4893455098934551,
      "grad_norm": 5.625,
      "learning_rate": 8.999190032881165e-06,
      "loss": 1.65904083,
      "memory(GiB)": 97.17,
      "step": 19290,
      "train_speed(iter/s)": 1.632427
    },
    {
      "acc": 0.67074599,
      "epoch": 0.48947234906139014,
      "grad_norm": 5.375,
      "learning_rate": 8.998560544297176e-06,
      "loss": 1.55652294,
      "memory(GiB)": 97.17,
      "step": 19295,
      "train_speed(iter/s)": 1.632469
    },
    {
      "acc": 0.6557714,
      "epoch": 0.4895991882293252,
      "grad_norm": 5.125,
      "learning_rate": 8.997930879838646e-06,
      "loss": 1.60311966,
      "memory(GiB)": 97.17,
      "step": 19300,
      "train_speed(iter/s)": 1.632513
    },
    {
      "acc": 0.65521827,
      "epoch": 0.4897260273972603,
      "grad_norm": 5.15625,
      "learning_rate": 8.997301039533264e-06,
      "loss": 1.62269154,
      "memory(GiB)": 97.17,
      "step": 19305,
      "train_speed(iter/s)": 1.632558
    },
    {
      "acc": 0.64769917,
      "epoch": 0.48985286656519533,
      "grad_norm": 5.6875,
      "learning_rate": 8.996671023408737e-06,
      "loss": 1.63032913,
      "memory(GiB)": 97.17,
      "step": 19310,
      "train_speed(iter/s)": 1.632597
    },
    {
      "acc": 0.64696217,
      "epoch": 0.4899797057331304,
      "grad_norm": 5.15625,
      "learning_rate": 8.996040831492772e-06,
      "loss": 1.63778114,
      "memory(GiB)": 97.17,
      "step": 19315,
      "train_speed(iter/s)": 1.632639
    },
    {
      "acc": 0.63116865,
      "epoch": 0.4901065449010654,
      "grad_norm": 4.8125,
      "learning_rate": 8.995410463813093e-06,
      "loss": 1.60176773,
      "memory(GiB)": 97.17,
      "step": 19320,
      "train_speed(iter/s)": 1.632686
    },
    {
      "acc": 0.65650911,
      "epoch": 0.4902333840690005,
      "grad_norm": 4.6875,
      "learning_rate": 8.99477992039742e-06,
      "loss": 1.57855148,
      "memory(GiB)": 97.17,
      "step": 19325,
      "train_speed(iter/s)": 1.632728
    },
    {
      "acc": 0.65348015,
      "epoch": 0.49036022323693557,
      "grad_norm": 6.78125,
      "learning_rate": 8.994149201273495e-06,
      "loss": 1.6068161,
      "memory(GiB)": 97.17,
      "step": 19330,
      "train_speed(iter/s)": 1.632774
    },
    {
      "acc": 0.6420002,
      "epoch": 0.4904870624048706,
      "grad_norm": 6.21875,
      "learning_rate": 8.993518306469052e-06,
      "loss": 1.58773594,
      "memory(GiB)": 97.17,
      "step": 19335,
      "train_speed(iter/s)": 1.632818
    },
    {
      "acc": 0.63002582,
      "epoch": 0.49061390157280566,
      "grad_norm": 7.0625,
      "learning_rate": 8.992887236011847e-06,
      "loss": 1.7034874,
      "memory(GiB)": 97.17,
      "step": 19340,
      "train_speed(iter/s)": 1.632861
    },
    {
      "acc": 0.64082441,
      "epoch": 0.49074074074074076,
      "grad_norm": 5.375,
      "learning_rate": 8.992255989929632e-06,
      "loss": 1.64837914,
      "memory(GiB)": 97.17,
      "step": 19345,
      "train_speed(iter/s)": 1.632907
    },
    {
      "acc": 0.67334785,
      "epoch": 0.4908675799086758,
      "grad_norm": 5.0625,
      "learning_rate": 8.991624568250175e-06,
      "loss": 1.55427952,
      "memory(GiB)": 97.17,
      "step": 19350,
      "train_speed(iter/s)": 1.63295
    },
    {
      "acc": 0.65000949,
      "epoch": 0.49099441907661084,
      "grad_norm": 6.21875,
      "learning_rate": 8.99099297100125e-06,
      "loss": 1.68209858,
      "memory(GiB)": 97.17,
      "step": 19355,
      "train_speed(iter/s)": 1.632991
    },
    {
      "acc": 0.6570837,
      "epoch": 0.4911212582445459,
      "grad_norm": 7.78125,
      "learning_rate": 8.990361198210634e-06,
      "loss": 1.61916962,
      "memory(GiB)": 97.17,
      "step": 19360,
      "train_speed(iter/s)": 1.633036
    },
    {
      "acc": 0.64149675,
      "epoch": 0.491248097412481,
      "grad_norm": 4.65625,
      "learning_rate": 8.989729249906116e-06,
      "loss": 1.70096588,
      "memory(GiB)": 97.17,
      "step": 19365,
      "train_speed(iter/s)": 1.633075
    },
    {
      "acc": 0.6558856,
      "epoch": 0.49137493658041603,
      "grad_norm": 5.09375,
      "learning_rate": 8.989097126115493e-06,
      "loss": 1.58833504,
      "memory(GiB)": 97.17,
      "step": 19370,
      "train_speed(iter/s)": 1.633118
    },
    {
      "acc": 0.64996033,
      "epoch": 0.4915017757483511,
      "grad_norm": 5.5,
      "learning_rate": 8.98846482686657e-06,
      "loss": 1.66905727,
      "memory(GiB)": 97.17,
      "step": 19375,
      "train_speed(iter/s)": 1.633162
    },
    {
      "acc": 0.66091418,
      "epoch": 0.4916286149162861,
      "grad_norm": 6.0625,
      "learning_rate": 8.987832352187156e-06,
      "loss": 1.64019527,
      "memory(GiB)": 97.17,
      "step": 19380,
      "train_speed(iter/s)": 1.633208
    },
    {
      "acc": 0.65745149,
      "epoch": 0.4917554540842212,
      "grad_norm": 5.84375,
      "learning_rate": 8.987199702105071e-06,
      "loss": 1.66534767,
      "memory(GiB)": 97.17,
      "step": 19385,
      "train_speed(iter/s)": 1.633252
    },
    {
      "acc": 0.64414577,
      "epoch": 0.49188229325215627,
      "grad_norm": 5.65625,
      "learning_rate": 8.986566876648141e-06,
      "loss": 1.66644592,
      "memory(GiB)": 97.17,
      "step": 19390,
      "train_speed(iter/s)": 1.633297
    },
    {
      "acc": 0.64303875,
      "epoch": 0.4920091324200913,
      "grad_norm": 6.46875,
      "learning_rate": 8.985933875844202e-06,
      "loss": 1.62279053,
      "memory(GiB)": 97.17,
      "step": 19395,
      "train_speed(iter/s)": 1.633341
    },
    {
      "acc": 0.67910142,
      "epoch": 0.49213597158802636,
      "grad_norm": 5.375,
      "learning_rate": 8.985300699721094e-06,
      "loss": 1.55852766,
      "memory(GiB)": 97.17,
      "step": 19400,
      "train_speed(iter/s)": 1.633384
    },
    {
      "acc": 0.65786572,
      "epoch": 0.49226281075596146,
      "grad_norm": 5.5,
      "learning_rate": 8.984667348306669e-06,
      "loss": 1.64398079,
      "memory(GiB)": 97.17,
      "step": 19405,
      "train_speed(iter/s)": 1.633429
    },
    {
      "acc": 0.66132321,
      "epoch": 0.4923896499238965,
      "grad_norm": 5.65625,
      "learning_rate": 8.984033821628782e-06,
      "loss": 1.57578688,
      "memory(GiB)": 97.17,
      "step": 19410,
      "train_speed(iter/s)": 1.633473
    },
    {
      "acc": 0.66964483,
      "epoch": 0.49251648909183154,
      "grad_norm": 5.34375,
      "learning_rate": 8.983400119715303e-06,
      "loss": 1.54320574,
      "memory(GiB)": 97.17,
      "step": 19415,
      "train_speed(iter/s)": 1.633518
    },
    {
      "acc": 0.65455723,
      "epoch": 0.4926433282597666,
      "grad_norm": 5.96875,
      "learning_rate": 8.982766242594099e-06,
      "loss": 1.61968479,
      "memory(GiB)": 97.17,
      "step": 19420,
      "train_speed(iter/s)": 1.633563
    },
    {
      "acc": 0.67131414,
      "epoch": 0.4927701674277017,
      "grad_norm": 7.21875,
      "learning_rate": 8.982132190293056e-06,
      "loss": 1.61747837,
      "memory(GiB)": 97.17,
      "step": 19425,
      "train_speed(iter/s)": 1.63361
    },
    {
      "acc": 0.65977802,
      "epoch": 0.49289700659563673,
      "grad_norm": 5.78125,
      "learning_rate": 8.98149796284006e-06,
      "loss": 1.58889999,
      "memory(GiB)": 97.17,
      "step": 19430,
      "train_speed(iter/s)": 1.633653
    },
    {
      "acc": 0.653619,
      "epoch": 0.4930238457635718,
      "grad_norm": 6.84375,
      "learning_rate": 8.980863560263007e-06,
      "loss": 1.60346317,
      "memory(GiB)": 97.17,
      "step": 19435,
      "train_speed(iter/s)": 1.6337
    },
    {
      "acc": 0.65869341,
      "epoch": 0.4931506849315068,
      "grad_norm": 5.6875,
      "learning_rate": 8.980228982589802e-06,
      "loss": 1.59405651,
      "memory(GiB)": 97.17,
      "step": 19440,
      "train_speed(iter/s)": 1.633742
    },
    {
      "acc": 0.66383777,
      "epoch": 0.4932775240994419,
      "grad_norm": 5.40625,
      "learning_rate": 8.979594229848355e-06,
      "loss": 1.61059952,
      "memory(GiB)": 97.17,
      "step": 19445,
      "train_speed(iter/s)": 1.633786
    },
    {
      "acc": 0.65295792,
      "epoch": 0.49340436326737697,
      "grad_norm": 5.78125,
      "learning_rate": 8.978959302066587e-06,
      "loss": 1.66142769,
      "memory(GiB)": 97.17,
      "step": 19450,
      "train_speed(iter/s)": 1.633834
    },
    {
      "acc": 0.64952669,
      "epoch": 0.493531202435312,
      "grad_norm": 5.0,
      "learning_rate": 8.978324199272423e-06,
      "loss": 1.60217705,
      "memory(GiB)": 97.17,
      "step": 19455,
      "train_speed(iter/s)": 1.633881
    },
    {
      "acc": 0.66519308,
      "epoch": 0.49365804160324706,
      "grad_norm": 6.1875,
      "learning_rate": 8.977688921493799e-06,
      "loss": 1.56891441,
      "memory(GiB)": 97.17,
      "step": 19460,
      "train_speed(iter/s)": 1.633929
    },
    {
      "acc": 0.66260753,
      "epoch": 0.49378488077118216,
      "grad_norm": 5.09375,
      "learning_rate": 8.977053468758659e-06,
      "loss": 1.56150455,
      "memory(GiB)": 97.17,
      "step": 19465,
      "train_speed(iter/s)": 1.633974
    },
    {
      "acc": 0.66025896,
      "epoch": 0.4939117199391172,
      "grad_norm": 5.65625,
      "learning_rate": 8.976417841094949e-06,
      "loss": 1.61268768,
      "memory(GiB)": 97.17,
      "step": 19470,
      "train_speed(iter/s)": 1.63402
    },
    {
      "acc": 0.65946765,
      "epoch": 0.49403855910705224,
      "grad_norm": 5.875,
      "learning_rate": 8.97578203853063e-06,
      "loss": 1.64046898,
      "memory(GiB)": 97.17,
      "step": 19475,
      "train_speed(iter/s)": 1.634066
    },
    {
      "acc": 0.65515213,
      "epoch": 0.4941653982749873,
      "grad_norm": 4.9375,
      "learning_rate": 8.975146061093667e-06,
      "loss": 1.60014114,
      "memory(GiB)": 97.17,
      "step": 19480,
      "train_speed(iter/s)": 1.634106
    },
    {
      "acc": 0.64681988,
      "epoch": 0.4942922374429224,
      "grad_norm": 5.3125,
      "learning_rate": 8.97450990881203e-06,
      "loss": 1.69216728,
      "memory(GiB)": 97.17,
      "step": 19485,
      "train_speed(iter/s)": 1.634151
    },
    {
      "acc": 0.66483059,
      "epoch": 0.49441907661085743,
      "grad_norm": 9.1875,
      "learning_rate": 8.973873581713705e-06,
      "loss": 1.59817677,
      "memory(GiB)": 97.17,
      "step": 19490,
      "train_speed(iter/s)": 1.634198
    },
    {
      "acc": 0.65917969,
      "epoch": 0.4945459157787925,
      "grad_norm": 5.59375,
      "learning_rate": 8.973237079826676e-06,
      "loss": 1.62293243,
      "memory(GiB)": 97.17,
      "step": 19495,
      "train_speed(iter/s)": 1.634239
    },
    {
      "acc": 0.6548646,
      "epoch": 0.4946727549467275,
      "grad_norm": 6.84375,
      "learning_rate": 8.972600403178941e-06,
      "loss": 1.63616447,
      "memory(GiB)": 97.17,
      "step": 19500,
      "train_speed(iter/s)": 1.634281
    },
    {
      "acc": 0.65196009,
      "epoch": 0.4947995941146626,
      "grad_norm": 6.25,
      "learning_rate": 8.971963551798506e-06,
      "loss": 1.56809616,
      "memory(GiB)": 97.17,
      "step": 19505,
      "train_speed(iter/s)": 1.634322
    },
    {
      "acc": 0.65663686,
      "epoch": 0.49492643328259767,
      "grad_norm": 6.3125,
      "learning_rate": 8.971326525713378e-06,
      "loss": 1.54991417,
      "memory(GiB)": 97.17,
      "step": 19510,
      "train_speed(iter/s)": 1.634365
    },
    {
      "acc": 0.64974566,
      "epoch": 0.4950532724505327,
      "grad_norm": 6.46875,
      "learning_rate": 8.97068932495158e-06,
      "loss": 1.65213737,
      "memory(GiB)": 97.17,
      "step": 19515,
      "train_speed(iter/s)": 1.634409
    },
    {
      "acc": 0.65483007,
      "epoch": 0.49518011161846776,
      "grad_norm": 9.25,
      "learning_rate": 8.970051949541137e-06,
      "loss": 1.6299305,
      "memory(GiB)": 97.17,
      "step": 19520,
      "train_speed(iter/s)": 1.634456
    },
    {
      "acc": 0.65145993,
      "epoch": 0.49530695078640286,
      "grad_norm": 5.875,
      "learning_rate": 8.969414399510085e-06,
      "loss": 1.60536118,
      "memory(GiB)": 97.17,
      "step": 19525,
      "train_speed(iter/s)": 1.634498
    },
    {
      "acc": 0.66421452,
      "epoch": 0.4954337899543379,
      "grad_norm": 5.375,
      "learning_rate": 8.968776674886466e-06,
      "loss": 1.61416245,
      "memory(GiB)": 97.17,
      "step": 19530,
      "train_speed(iter/s)": 1.634543
    },
    {
      "acc": 0.63569074,
      "epoch": 0.49556062912227294,
      "grad_norm": 8.75,
      "learning_rate": 8.968138775698328e-06,
      "loss": 1.70665836,
      "memory(GiB)": 97.17,
      "step": 19535,
      "train_speed(iter/s)": 1.634593
    },
    {
      "acc": 0.66848354,
      "epoch": 0.495687468290208,
      "grad_norm": 6.1875,
      "learning_rate": 8.96750070197373e-06,
      "loss": 1.57719469,
      "memory(GiB)": 97.17,
      "step": 19540,
      "train_speed(iter/s)": 1.63464
    },
    {
      "acc": 0.64501586,
      "epoch": 0.4958143074581431,
      "grad_norm": 5.90625,
      "learning_rate": 8.966862453740738e-06,
      "loss": 1.60796223,
      "memory(GiB)": 97.17,
      "step": 19545,
      "train_speed(iter/s)": 1.634687
    },
    {
      "acc": 0.65066652,
      "epoch": 0.49594114662607813,
      "grad_norm": 5.15625,
      "learning_rate": 8.966224031027426e-06,
      "loss": 1.66179066,
      "memory(GiB)": 97.17,
      "step": 19550,
      "train_speed(iter/s)": 1.63473
    },
    {
      "acc": 0.64585896,
      "epoch": 0.4960679857940132,
      "grad_norm": 5.5625,
      "learning_rate": 8.965585433861871e-06,
      "loss": 1.6374012,
      "memory(GiB)": 97.17,
      "step": 19555,
      "train_speed(iter/s)": 1.63478
    },
    {
      "acc": 0.65385785,
      "epoch": 0.4961948249619482,
      "grad_norm": 5.59375,
      "learning_rate": 8.964946662272167e-06,
      "loss": 1.5826478,
      "memory(GiB)": 97.17,
      "step": 19560,
      "train_speed(iter/s)": 1.634827
    },
    {
      "acc": 0.65989742,
      "epoch": 0.4963216641298833,
      "grad_norm": 5.59375,
      "learning_rate": 8.964307716286404e-06,
      "loss": 1.58067436,
      "memory(GiB)": 97.17,
      "step": 19565,
      "train_speed(iter/s)": 1.63487
    },
    {
      "acc": 0.66022587,
      "epoch": 0.49644850329781837,
      "grad_norm": 6.21875,
      "learning_rate": 8.963668595932689e-06,
      "loss": 1.646978,
      "memory(GiB)": 97.17,
      "step": 19570,
      "train_speed(iter/s)": 1.634917
    },
    {
      "acc": 0.65500035,
      "epoch": 0.4965753424657534,
      "grad_norm": 5.6875,
      "learning_rate": 8.96302930123913e-06,
      "loss": 1.6167778,
      "memory(GiB)": 97.17,
      "step": 19575,
      "train_speed(iter/s)": 1.634962
    },
    {
      "acc": 0.66047921,
      "epoch": 0.49670218163368846,
      "grad_norm": 5.15625,
      "learning_rate": 8.962389832233853e-06,
      "loss": 1.57737408,
      "memory(GiB)": 97.17,
      "step": 19580,
      "train_speed(iter/s)": 1.635008
    },
    {
      "acc": 0.64837852,
      "epoch": 0.49682902080162356,
      "grad_norm": 7.90625,
      "learning_rate": 8.961750188944978e-06,
      "loss": 1.69679546,
      "memory(GiB)": 97.17,
      "step": 19585,
      "train_speed(iter/s)": 1.635057
    },
    {
      "acc": 0.6454319,
      "epoch": 0.4969558599695586,
      "grad_norm": 5.625,
      "learning_rate": 8.96111037140064e-06,
      "loss": 1.61879864,
      "memory(GiB)": 97.17,
      "step": 19590,
      "train_speed(iter/s)": 1.6351
    },
    {
      "acc": 0.65377541,
      "epoch": 0.49708269913749364,
      "grad_norm": 4.78125,
      "learning_rate": 8.960470379628986e-06,
      "loss": 1.5757411,
      "memory(GiB)": 97.17,
      "step": 19595,
      "train_speed(iter/s)": 1.635146
    },
    {
      "acc": 0.64588599,
      "epoch": 0.4972095383054287,
      "grad_norm": 6.0,
      "learning_rate": 8.959830213658161e-06,
      "loss": 1.6123106,
      "memory(GiB)": 97.17,
      "step": 19600,
      "train_speed(iter/s)": 1.635192
    },
    {
      "acc": 0.64932814,
      "epoch": 0.4973363774733638,
      "grad_norm": 5.46875,
      "learning_rate": 8.959189873516324e-06,
      "loss": 1.67450199,
      "memory(GiB)": 97.17,
      "step": 19605,
      "train_speed(iter/s)": 1.635236
    },
    {
      "acc": 0.64351878,
      "epoch": 0.49746321664129883,
      "grad_norm": 5.15625,
      "learning_rate": 8.95854935923164e-06,
      "loss": 1.6929636,
      "memory(GiB)": 97.17,
      "step": 19610,
      "train_speed(iter/s)": 1.635281
    },
    {
      "acc": 0.65900097,
      "epoch": 0.4975900558092339,
      "grad_norm": 7.09375,
      "learning_rate": 8.95790867083228e-06,
      "loss": 1.67759018,
      "memory(GiB)": 97.17,
      "step": 19615,
      "train_speed(iter/s)": 1.635327
    },
    {
      "acc": 0.63874559,
      "epoch": 0.4977168949771689,
      "grad_norm": 6.6875,
      "learning_rate": 8.957267808346428e-06,
      "loss": 1.67475052,
      "memory(GiB)": 97.17,
      "step": 19620,
      "train_speed(iter/s)": 1.635372
    },
    {
      "acc": 0.66545205,
      "epoch": 0.497843734145104,
      "grad_norm": 5.59375,
      "learning_rate": 8.95662677180227e-06,
      "loss": 1.55602398,
      "memory(GiB)": 97.17,
      "step": 19625,
      "train_speed(iter/s)": 1.635419
    },
    {
      "acc": 0.66076527,
      "epoch": 0.49797057331303907,
      "grad_norm": 5.53125,
      "learning_rate": 8.955985561228e-06,
      "loss": 1.59383564,
      "memory(GiB)": 97.17,
      "step": 19630,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.66685324,
      "epoch": 0.4980974124809741,
      "grad_norm": 6.25,
      "learning_rate": 8.955344176651824e-06,
      "loss": 1.60934792,
      "memory(GiB)": 97.17,
      "step": 19635,
      "train_speed(iter/s)": 1.635509
    },
    {
      "acc": 0.66330113,
      "epoch": 0.49822425164890916,
      "grad_norm": 6.625,
      "learning_rate": 8.954702618101952e-06,
      "loss": 1.57660933,
      "memory(GiB)": 97.17,
      "step": 19640,
      "train_speed(iter/s)": 1.635555
    },
    {
      "acc": 0.64868908,
      "epoch": 0.49835109081684426,
      "grad_norm": 7.3125,
      "learning_rate": 8.9540608856066e-06,
      "loss": 1.63900948,
      "memory(GiB)": 97.17,
      "step": 19645,
      "train_speed(iter/s)": 1.635599
    },
    {
      "acc": 0.64220729,
      "epoch": 0.4984779299847793,
      "grad_norm": 7.625,
      "learning_rate": 8.953418979194e-06,
      "loss": 1.71169529,
      "memory(GiB)": 97.17,
      "step": 19650,
      "train_speed(iter/s)": 1.635641
    },
    {
      "acc": 0.6536541,
      "epoch": 0.49860476915271434,
      "grad_norm": 5.90625,
      "learning_rate": 8.95277689889238e-06,
      "loss": 1.58922529,
      "memory(GiB)": 97.17,
      "step": 19655,
      "train_speed(iter/s)": 1.635686
    },
    {
      "acc": 0.64948592,
      "epoch": 0.4987316083206494,
      "grad_norm": 6.09375,
      "learning_rate": 8.952134644729985e-06,
      "loss": 1.63966732,
      "memory(GiB)": 97.17,
      "step": 19660,
      "train_speed(iter/s)": 1.635731
    },
    {
      "acc": 0.66129317,
      "epoch": 0.4988584474885845,
      "grad_norm": 5.9375,
      "learning_rate": 8.951492216735062e-06,
      "loss": 1.58415575,
      "memory(GiB)": 97.17,
      "step": 19665,
      "train_speed(iter/s)": 1.635777
    },
    {
      "acc": 0.63886471,
      "epoch": 0.49898528665651953,
      "grad_norm": 5.625,
      "learning_rate": 8.950849614935872e-06,
      "loss": 1.65664673,
      "memory(GiB)": 97.17,
      "step": 19670,
      "train_speed(iter/s)": 1.635822
    },
    {
      "acc": 0.64195886,
      "epoch": 0.4991121258244546,
      "grad_norm": 6.3125,
      "learning_rate": 8.950206839360674e-06,
      "loss": 1.69861031,
      "memory(GiB)": 97.17,
      "step": 19675,
      "train_speed(iter/s)": 1.635865
    },
    {
      "acc": 0.64352846,
      "epoch": 0.4992389649923896,
      "grad_norm": 4.71875,
      "learning_rate": 8.949563890037745e-06,
      "loss": 1.67011986,
      "memory(GiB)": 97.17,
      "step": 19680,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.64192257,
      "epoch": 0.4993658041603247,
      "grad_norm": 5.59375,
      "learning_rate": 8.948920766995362e-06,
      "loss": 1.6732151,
      "memory(GiB)": 97.17,
      "step": 19685,
      "train_speed(iter/s)": 1.635954
    },
    {
      "acc": 0.6566762,
      "epoch": 0.49949264332825977,
      "grad_norm": 5.875,
      "learning_rate": 8.948277470261812e-06,
      "loss": 1.59000168,
      "memory(GiB)": 97.17,
      "step": 19690,
      "train_speed(iter/s)": 1.635996
    },
    {
      "acc": 0.63902798,
      "epoch": 0.4996194824961948,
      "grad_norm": 5.96875,
      "learning_rate": 8.94763399986539e-06,
      "loss": 1.63371792,
      "memory(GiB)": 97.17,
      "step": 19695,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.66027279,
      "epoch": 0.49974632166412986,
      "grad_norm": 4.625,
      "learning_rate": 8.946990355834401e-06,
      "loss": 1.58149529,
      "memory(GiB)": 97.17,
      "step": 19700,
      "train_speed(iter/s)": 1.636085
    },
    {
      "acc": 0.65635209,
      "epoch": 0.49987316083206496,
      "grad_norm": 5.75,
      "learning_rate": 8.946346538197156e-06,
      "loss": 1.55523911,
      "memory(GiB)": 97.17,
      "step": 19705,
      "train_speed(iter/s)": 1.63613
    },
    {
      "acc": 0.64486589,
      "epoch": 0.5,
      "grad_norm": 5.15625,
      "learning_rate": 8.94570254698197e-06,
      "loss": 1.64925861,
      "memory(GiB)": 97.17,
      "step": 19710,
      "train_speed(iter/s)": 1.636173
    },
    {
      "acc": 0.6442924,
      "epoch": 0.5001268391679351,
      "grad_norm": 5.59375,
      "learning_rate": 8.945058382217168e-06,
      "loss": 1.63886719,
      "memory(GiB)": 97.17,
      "step": 19715,
      "train_speed(iter/s)": 1.636219
    },
    {
      "acc": 0.64146109,
      "epoch": 0.5002536783358701,
      "grad_norm": 5.34375,
      "learning_rate": 8.944414043931086e-06,
      "loss": 1.68201981,
      "memory(GiB)": 97.17,
      "step": 19720,
      "train_speed(iter/s)": 1.63626
    },
    {
      "acc": 0.65783596,
      "epoch": 0.5003805175038052,
      "grad_norm": 4.4375,
      "learning_rate": 8.943769532152065e-06,
      "loss": 1.56636171,
      "memory(GiB)": 97.17,
      "step": 19725,
      "train_speed(iter/s)": 1.636302
    },
    {
      "acc": 0.64758997,
      "epoch": 0.5005073566717403,
      "grad_norm": 5.25,
      "learning_rate": 8.94312484690845e-06,
      "loss": 1.60370312,
      "memory(GiB)": 97.17,
      "step": 19730,
      "train_speed(iter/s)": 1.63635
    },
    {
      "acc": 0.65292311,
      "epoch": 0.5006341958396753,
      "grad_norm": 5.5,
      "learning_rate": 8.9424799882286e-06,
      "loss": 1.62388172,
      "memory(GiB)": 97.17,
      "step": 19735,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.66491714,
      "epoch": 0.5007610350076104,
      "grad_norm": 4.6875,
      "learning_rate": 8.94183495614088e-06,
      "loss": 1.58442364,
      "memory(GiB)": 97.17,
      "step": 19740,
      "train_speed(iter/s)": 1.636436
    },
    {
      "acc": 0.67173519,
      "epoch": 0.5008878741755454,
      "grad_norm": 5.28125,
      "learning_rate": 8.941189750673658e-06,
      "loss": 1.56227512,
      "memory(GiB)": 97.17,
      "step": 19745,
      "train_speed(iter/s)": 1.636478
    },
    {
      "acc": 0.64533234,
      "epoch": 0.5010147133434805,
      "grad_norm": 5.3125,
      "learning_rate": 8.940544371855315e-06,
      "loss": 1.60845604,
      "memory(GiB)": 97.17,
      "step": 19750,
      "train_speed(iter/s)": 1.636525
    },
    {
      "acc": 0.66256008,
      "epoch": 0.5011415525114156,
      "grad_norm": 6.65625,
      "learning_rate": 8.939898819714237e-06,
      "loss": 1.53998079,
      "memory(GiB)": 97.17,
      "step": 19755,
      "train_speed(iter/s)": 1.63657
    },
    {
      "acc": 0.65779552,
      "epoch": 0.5012683916793506,
      "grad_norm": 5.25,
      "learning_rate": 8.93925309427882e-06,
      "loss": 1.58629084,
      "memory(GiB)": 97.17,
      "step": 19760,
      "train_speed(iter/s)": 1.636618
    },
    {
      "acc": 0.652987,
      "epoch": 0.5013952308472857,
      "grad_norm": 5.65625,
      "learning_rate": 8.938607195577462e-06,
      "loss": 1.66644135,
      "memory(GiB)": 97.17,
      "step": 19765,
      "train_speed(iter/s)": 1.63666
    },
    {
      "acc": 0.64125495,
      "epoch": 0.5015220700152208,
      "grad_norm": 4.90625,
      "learning_rate": 8.937961123638577e-06,
      "loss": 1.67882233,
      "memory(GiB)": 97.17,
      "step": 19770,
      "train_speed(iter/s)": 1.636706
    },
    {
      "acc": 0.64762201,
      "epoch": 0.5016489091831557,
      "grad_norm": 5.78125,
      "learning_rate": 8.93731487849058e-06,
      "loss": 1.65156441,
      "memory(GiB)": 97.17,
      "step": 19775,
      "train_speed(iter/s)": 1.636749
    },
    {
      "acc": 0.63744946,
      "epoch": 0.5017757483510908,
      "grad_norm": 6.9375,
      "learning_rate": 8.936668460161895e-06,
      "loss": 1.66271744,
      "memory(GiB)": 97.17,
      "step": 19780,
      "train_speed(iter/s)": 1.636796
    },
    {
      "acc": 0.65909729,
      "epoch": 0.5019025875190258,
      "grad_norm": 6.34375,
      "learning_rate": 8.936021868680956e-06,
      "loss": 1.68622627,
      "memory(GiB)": 97.17,
      "step": 19785,
      "train_speed(iter/s)": 1.636837
    },
    {
      "acc": 0.64431381,
      "epoch": 0.5020294266869609,
      "grad_norm": 6.09375,
      "learning_rate": 8.935375104076201e-06,
      "loss": 1.68079185,
      "memory(GiB)": 97.17,
      "step": 19790,
      "train_speed(iter/s)": 1.636882
    },
    {
      "acc": 0.65269499,
      "epoch": 0.502156265854896,
      "grad_norm": 5.59375,
      "learning_rate": 8.93472816637608e-06,
      "loss": 1.61054878,
      "memory(GiB)": 97.17,
      "step": 19795,
      "train_speed(iter/s)": 1.636925
    },
    {
      "acc": 0.65919523,
      "epoch": 0.502283105022831,
      "grad_norm": 5.5625,
      "learning_rate": 8.934081055609046e-06,
      "loss": 1.63678875,
      "memory(GiB)": 97.17,
      "step": 19800,
      "train_speed(iter/s)": 1.63697
    },
    {
      "acc": 0.66355925,
      "epoch": 0.5024099441907661,
      "grad_norm": 5.75,
      "learning_rate": 8.933433771803562e-06,
      "loss": 1.58405828,
      "memory(GiB)": 97.17,
      "step": 19805,
      "train_speed(iter/s)": 1.637015
    },
    {
      "acc": 0.65147705,
      "epoch": 0.5025367833587012,
      "grad_norm": 5.25,
      "learning_rate": 8.932786314988099e-06,
      "loss": 1.61141987,
      "memory(GiB)": 97.17,
      "step": 19810,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.65990744,
      "epoch": 0.5026636225266362,
      "grad_norm": 6.84375,
      "learning_rate": 8.932138685191136e-06,
      "loss": 1.60928097,
      "memory(GiB)": 97.17,
      "step": 19815,
      "train_speed(iter/s)": 1.637099
    },
    {
      "acc": 0.66688509,
      "epoch": 0.5027904616945713,
      "grad_norm": 6.90625,
      "learning_rate": 8.931490882441159e-06,
      "loss": 1.55720978,
      "memory(GiB)": 97.17,
      "step": 19820,
      "train_speed(iter/s)": 1.637143
    },
    {
      "acc": 0.64496918,
      "epoch": 0.5029173008625063,
      "grad_norm": 7.09375,
      "learning_rate": 8.930842906766659e-06,
      "loss": 1.65523148,
      "memory(GiB)": 97.17,
      "step": 19825,
      "train_speed(iter/s)": 1.637188
    },
    {
      "acc": 0.66474676,
      "epoch": 0.5030441400304414,
      "grad_norm": 6.46875,
      "learning_rate": 8.930194758196138e-06,
      "loss": 1.56693897,
      "memory(GiB)": 97.17,
      "step": 19830,
      "train_speed(iter/s)": 1.637232
    },
    {
      "acc": 0.64261909,
      "epoch": 0.5031709791983765,
      "grad_norm": 4.9375,
      "learning_rate": 8.929546436758105e-06,
      "loss": 1.6713604,
      "memory(GiB)": 97.17,
      "step": 19835,
      "train_speed(iter/s)": 1.637276
    },
    {
      "acc": 0.65234485,
      "epoch": 0.5032978183663115,
      "grad_norm": 5.09375,
      "learning_rate": 8.928897942481075e-06,
      "loss": 1.58415432,
      "memory(GiB)": 97.17,
      "step": 19840,
      "train_speed(iter/s)": 1.637318
    },
    {
      "acc": 0.64369764,
      "epoch": 0.5034246575342466,
      "grad_norm": 7.46875,
      "learning_rate": 8.928249275393572e-06,
      "loss": 1.61291428,
      "memory(GiB)": 97.17,
      "step": 19845,
      "train_speed(iter/s)": 1.637362
    },
    {
      "acc": 0.65665665,
      "epoch": 0.5035514967021817,
      "grad_norm": 5.28125,
      "learning_rate": 8.927600435524129e-06,
      "loss": 1.57517281,
      "memory(GiB)": 97.17,
      "step": 19850,
      "train_speed(iter/s)": 1.637404
    },
    {
      "acc": 0.64826097,
      "epoch": 0.5036783358701167,
      "grad_norm": 5.625,
      "learning_rate": 8.926951422901282e-06,
      "loss": 1.67055874,
      "memory(GiB)": 97.17,
      "step": 19855,
      "train_speed(iter/s)": 1.637446
    },
    {
      "acc": 0.66410532,
      "epoch": 0.5038051750380518,
      "grad_norm": 8.25,
      "learning_rate": 8.926302237553578e-06,
      "loss": 1.59546185,
      "memory(GiB)": 97.17,
      "step": 19860,
      "train_speed(iter/s)": 1.63749
    },
    {
      "acc": 0.64603043,
      "epoch": 0.5039320142059868,
      "grad_norm": 6.375,
      "learning_rate": 8.925652879509575e-06,
      "loss": 1.61109486,
      "memory(GiB)": 97.17,
      "step": 19865,
      "train_speed(iter/s)": 1.637533
    },
    {
      "acc": 0.65816126,
      "epoch": 0.5040588533739219,
      "grad_norm": 6.875,
      "learning_rate": 8.925003348797829e-06,
      "loss": 1.68585777,
      "memory(GiB)": 97.17,
      "step": 19870,
      "train_speed(iter/s)": 1.637578
    },
    {
      "acc": 0.65360856,
      "epoch": 0.504185692541857,
      "grad_norm": 5.25,
      "learning_rate": 8.924353645446912e-06,
      "loss": 1.55985394,
      "memory(GiB)": 97.17,
      "step": 19875,
      "train_speed(iter/s)": 1.637622
    },
    {
      "acc": 0.65019102,
      "epoch": 0.504312531709792,
      "grad_norm": 5.6875,
      "learning_rate": 8.923703769485403e-06,
      "loss": 1.62335587,
      "memory(GiB)": 97.17,
      "step": 19880,
      "train_speed(iter/s)": 1.637665
    },
    {
      "acc": 0.65725064,
      "epoch": 0.5044393708777271,
      "grad_norm": 6.28125,
      "learning_rate": 8.92305372094188e-06,
      "loss": 1.62615643,
      "memory(GiB)": 97.17,
      "step": 19885,
      "train_speed(iter/s)": 1.637708
    },
    {
      "acc": 0.66691246,
      "epoch": 0.5045662100456622,
      "grad_norm": 6.5625,
      "learning_rate": 8.922403499844943e-06,
      "loss": 1.56492739,
      "memory(GiB)": 97.17,
      "step": 19890,
      "train_speed(iter/s)": 1.63775
    },
    {
      "acc": 0.65334301,
      "epoch": 0.5046930492135971,
      "grad_norm": 5.78125,
      "learning_rate": 8.921753106223186e-06,
      "loss": 1.58829002,
      "memory(GiB)": 97.17,
      "step": 19895,
      "train_speed(iter/s)": 1.637793
    },
    {
      "acc": 0.62701683,
      "epoch": 0.5048198883815322,
      "grad_norm": 5.78125,
      "learning_rate": 8.92110254010522e-06,
      "loss": 1.7714592,
      "memory(GiB)": 97.17,
      "step": 19900,
      "train_speed(iter/s)": 1.637837
    },
    {
      "acc": 0.63515863,
      "epoch": 0.5049467275494672,
      "grad_norm": 4.5,
      "learning_rate": 8.920451801519656e-06,
      "loss": 1.70014973,
      "memory(GiB)": 97.17,
      "step": 19905,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.65417986,
      "epoch": 0.5050735667174023,
      "grad_norm": 6.78125,
      "learning_rate": 8.919800890495118e-06,
      "loss": 1.62693977,
      "memory(GiB)": 97.17,
      "step": 19910,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.6486495,
      "epoch": 0.5052004058853374,
      "grad_norm": 5.25,
      "learning_rate": 8.919149807060237e-06,
      "loss": 1.65757732,
      "memory(GiB)": 97.17,
      "step": 19915,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.64851494,
      "epoch": 0.5053272450532724,
      "grad_norm": 5.75,
      "learning_rate": 8.918498551243649e-06,
      "loss": 1.67344894,
      "memory(GiB)": 97.17,
      "step": 19920,
      "train_speed(iter/s)": 1.638007
    },
    {
      "acc": 0.64613094,
      "epoch": 0.5054540842212075,
      "grad_norm": 5.03125,
      "learning_rate": 8.917847123073999e-06,
      "loss": 1.63346424,
      "memory(GiB)": 97.17,
      "step": 19925,
      "train_speed(iter/s)": 1.638051
    },
    {
      "acc": 0.64864411,
      "epoch": 0.5055809233891426,
      "grad_norm": 8.1875,
      "learning_rate": 8.917195522579943e-06,
      "loss": 1.5969182,
      "memory(GiB)": 97.17,
      "step": 19930,
      "train_speed(iter/s)": 1.638098
    },
    {
      "acc": 0.65429907,
      "epoch": 0.5057077625570776,
      "grad_norm": 5.96875,
      "learning_rate": 8.916543749790139e-06,
      "loss": 1.63215523,
      "memory(GiB)": 97.17,
      "step": 19935,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.64883585,
      "epoch": 0.5058346017250127,
      "grad_norm": 7.25,
      "learning_rate": 8.915891804733253e-06,
      "loss": 1.64344845,
      "memory(GiB)": 97.17,
      "step": 19940,
      "train_speed(iter/s)": 1.63819
    },
    {
      "acc": 0.66215897,
      "epoch": 0.5059614408929477,
      "grad_norm": 5.375,
      "learning_rate": 8.915239687437963e-06,
      "loss": 1.5686039,
      "memory(GiB)": 97.17,
      "step": 19945,
      "train_speed(iter/s)": 1.638233
    },
    {
      "acc": 0.65325317,
      "epoch": 0.5060882800608828,
      "grad_norm": 5.125,
      "learning_rate": 8.91458739793295e-06,
      "loss": 1.61301193,
      "memory(GiB)": 97.17,
      "step": 19950,
      "train_speed(iter/s)": 1.638275
    },
    {
      "acc": 0.65407767,
      "epoch": 0.5062151192288179,
      "grad_norm": 5.65625,
      "learning_rate": 8.913934936246907e-06,
      "loss": 1.65999413,
      "memory(GiB)": 97.17,
      "step": 19955,
      "train_speed(iter/s)": 1.638319
    },
    {
      "acc": 0.63604031,
      "epoch": 0.5063419583967529,
      "grad_norm": 5.71875,
      "learning_rate": 8.91328230240853e-06,
      "loss": 1.71096077,
      "memory(GiB)": 97.17,
      "step": 19960,
      "train_speed(iter/s)": 1.638362
    },
    {
      "acc": 0.65486422,
      "epoch": 0.506468797564688,
      "grad_norm": 5.78125,
      "learning_rate": 8.912629496446528e-06,
      "loss": 1.56815701,
      "memory(GiB)": 97.17,
      "step": 19965,
      "train_speed(iter/s)": 1.638406
    },
    {
      "acc": 0.66075773,
      "epoch": 0.5065956367326231,
      "grad_norm": 7.5,
      "learning_rate": 8.911976518389612e-06,
      "loss": 1.51954403,
      "memory(GiB)": 97.17,
      "step": 19970,
      "train_speed(iter/s)": 1.638449
    },
    {
      "acc": 0.66865425,
      "epoch": 0.5067224759005581,
      "grad_norm": 5.875,
      "learning_rate": 8.9113233682665e-06,
      "loss": 1.57399855,
      "memory(GiB)": 97.17,
      "step": 19975,
      "train_speed(iter/s)": 1.638493
    },
    {
      "acc": 0.64773932,
      "epoch": 0.5068493150684932,
      "grad_norm": 5.25,
      "learning_rate": 8.910670046105927e-06,
      "loss": 1.60767498,
      "memory(GiB)": 97.17,
      "step": 19980,
      "train_speed(iter/s)": 1.63854
    },
    {
      "acc": 0.65828819,
      "epoch": 0.5069761542364282,
      "grad_norm": 5.9375,
      "learning_rate": 8.910016551936623e-06,
      "loss": 1.63433437,
      "memory(GiB)": 97.17,
      "step": 19985,
      "train_speed(iter/s)": 1.638583
    },
    {
      "acc": 0.64350533,
      "epoch": 0.5071029934043633,
      "grad_norm": 6.0,
      "learning_rate": 8.909362885787333e-06,
      "loss": 1.66422157,
      "memory(GiB)": 97.17,
      "step": 19990,
      "train_speed(iter/s)": 1.638626
    },
    {
      "acc": 0.64171085,
      "epoch": 0.5072298325722984,
      "grad_norm": 5.25,
      "learning_rate": 8.908709047686813e-06,
      "loss": 1.67835464,
      "memory(GiB)": 97.17,
      "step": 19995,
      "train_speed(iter/s)": 1.63867
    },
    {
      "acc": 0.65791407,
      "epoch": 0.5073566717402334,
      "grad_norm": 6.625,
      "learning_rate": 8.908055037663815e-06,
      "loss": 1.63178844,
      "memory(GiB)": 97.17,
      "step": 20000,
      "train_speed(iter/s)": 1.638715
    },
    {
      "epoch": 0.5073566717402334,
      "eval_acc": 0.6437379008764621,
      "eval_loss": 1.5866650342941284,
      "eval_runtime": 58.3523,
      "eval_samples_per_second": 109.164,
      "eval_steps_per_second": 27.3,
      "step": 20000
    },
    {
      "acc": 0.66467433,
      "epoch": 0.5074835109081685,
      "grad_norm": 6.25,
      "learning_rate": 8.907400855747111e-06,
      "loss": 1.60907555,
      "memory(GiB)": 97.17,
      "step": 20005,
      "train_speed(iter/s)": 1.6304
    },
    {
      "acc": 0.62223005,
      "epoch": 0.5076103500761036,
      "grad_norm": 5.4375,
      "learning_rate": 8.906746501965468e-06,
      "loss": 1.74345722,
      "memory(GiB)": 97.17,
      "step": 20010,
      "train_speed(iter/s)": 1.630442
    },
    {
      "acc": 0.67485809,
      "epoch": 0.5077371892440385,
      "grad_norm": 6.53125,
      "learning_rate": 8.906091976347675e-06,
      "loss": 1.53418055,
      "memory(GiB)": 97.17,
      "step": 20015,
      "train_speed(iter/s)": 1.630483
    },
    {
      "acc": 0.64181247,
      "epoch": 0.5078640284119736,
      "grad_norm": 5.0625,
      "learning_rate": 8.905437278922516e-06,
      "loss": 1.62519455,
      "memory(GiB)": 97.17,
      "step": 20020,
      "train_speed(iter/s)": 1.630524
    },
    {
      "acc": 0.657902,
      "epoch": 0.5079908675799086,
      "grad_norm": 6.3125,
      "learning_rate": 8.90478240971879e-06,
      "loss": 1.61010838,
      "memory(GiB)": 97.17,
      "step": 20025,
      "train_speed(iter/s)": 1.630571
    },
    {
      "acc": 0.65885143,
      "epoch": 0.5081177067478437,
      "grad_norm": 5.34375,
      "learning_rate": 8.904127368765298e-06,
      "loss": 1.62665596,
      "memory(GiB)": 97.17,
      "step": 20030,
      "train_speed(iter/s)": 1.630613
    },
    {
      "acc": 0.63172417,
      "epoch": 0.5082445459157788,
      "grad_norm": 6.625,
      "learning_rate": 8.903472156090856e-06,
      "loss": 1.62974129,
      "memory(GiB)": 97.17,
      "step": 20035,
      "train_speed(iter/s)": 1.630658
    },
    {
      "acc": 0.66237855,
      "epoch": 0.5083713850837138,
      "grad_norm": 5.53125,
      "learning_rate": 8.902816771724279e-06,
      "loss": 1.56528263,
      "memory(GiB)": 97.17,
      "step": 20040,
      "train_speed(iter/s)": 1.6307
    },
    {
      "acc": 0.64765735,
      "epoch": 0.5084982242516489,
      "grad_norm": 5.28125,
      "learning_rate": 8.902161215694396e-06,
      "loss": 1.59422989,
      "memory(GiB)": 97.17,
      "step": 20045,
      "train_speed(iter/s)": 1.630745
    },
    {
      "acc": 0.65953779,
      "epoch": 0.508625063419584,
      "grad_norm": 5.4375,
      "learning_rate": 8.901505488030042e-06,
      "loss": 1.61008282,
      "memory(GiB)": 97.17,
      "step": 20050,
      "train_speed(iter/s)": 1.630789
    },
    {
      "acc": 0.63473411,
      "epoch": 0.508751902587519,
      "grad_norm": 6.09375,
      "learning_rate": 8.900849588760057e-06,
      "loss": 1.6123127,
      "memory(GiB)": 97.17,
      "step": 20055,
      "train_speed(iter/s)": 1.630833
    },
    {
      "acc": 0.6434783,
      "epoch": 0.5088787417554541,
      "grad_norm": 5.46875,
      "learning_rate": 8.90019351791329e-06,
      "loss": 1.65244026,
      "memory(GiB)": 97.17,
      "step": 20060,
      "train_speed(iter/s)": 1.630869
    },
    {
      "acc": 0.65200148,
      "epoch": 0.5090055809233891,
      "grad_norm": 7.15625,
      "learning_rate": 8.8995372755186e-06,
      "loss": 1.59405804,
      "memory(GiB)": 97.17,
      "step": 20065,
      "train_speed(iter/s)": 1.630912
    },
    {
      "acc": 0.64505138,
      "epoch": 0.5091324200913242,
      "grad_norm": 8.4375,
      "learning_rate": 8.89888086160485e-06,
      "loss": 1.60657616,
      "memory(GiB)": 97.17,
      "step": 20070,
      "train_speed(iter/s)": 1.630955
    },
    {
      "acc": 0.64989672,
      "epoch": 0.5092592592592593,
      "grad_norm": 5.46875,
      "learning_rate": 8.898224276200913e-06,
      "loss": 1.63871651,
      "memory(GiB)": 97.17,
      "step": 20075,
      "train_speed(iter/s)": 1.630998
    },
    {
      "acc": 0.6588295,
      "epoch": 0.5093860984271943,
      "grad_norm": 5.34375,
      "learning_rate": 8.897567519335669e-06,
      "loss": 1.59247589,
      "memory(GiB)": 97.17,
      "step": 20080,
      "train_speed(iter/s)": 1.631041
    },
    {
      "acc": 0.64510145,
      "epoch": 0.5095129375951294,
      "grad_norm": 5.84375,
      "learning_rate": 8.896910591038002e-06,
      "loss": 1.61216774,
      "memory(GiB)": 97.17,
      "step": 20085,
      "train_speed(iter/s)": 1.631087
    },
    {
      "acc": 0.64206228,
      "epoch": 0.5096397767630645,
      "grad_norm": 8.0625,
      "learning_rate": 8.896253491336809e-06,
      "loss": 1.69341984,
      "memory(GiB)": 97.17,
      "step": 20090,
      "train_speed(iter/s)": 1.631132
    },
    {
      "acc": 0.63603344,
      "epoch": 0.5097666159309995,
      "grad_norm": 5.875,
      "learning_rate": 8.895596220260993e-06,
      "loss": 1.64787731,
      "memory(GiB)": 97.17,
      "step": 20095,
      "train_speed(iter/s)": 1.631175
    },
    {
      "acc": 0.65597162,
      "epoch": 0.5098934550989346,
      "grad_norm": 7.0625,
      "learning_rate": 8.894938777839462e-06,
      "loss": 1.63399353,
      "memory(GiB)": 97.17,
      "step": 20100,
      "train_speed(iter/s)": 1.631216
    },
    {
      "acc": 0.65730529,
      "epoch": 0.5100202942668696,
      "grad_norm": 5.8125,
      "learning_rate": 8.894281164101134e-06,
      "loss": 1.60963593,
      "memory(GiB)": 97.17,
      "step": 20105,
      "train_speed(iter/s)": 1.631256
    },
    {
      "acc": 0.63457851,
      "epoch": 0.5101471334348047,
      "grad_norm": 5.1875,
      "learning_rate": 8.893623379074934e-06,
      "loss": 1.64609795,
      "memory(GiB)": 97.17,
      "step": 20110,
      "train_speed(iter/s)": 1.631297
    },
    {
      "acc": 0.65438733,
      "epoch": 0.5102739726027398,
      "grad_norm": 5.75,
      "learning_rate": 8.892965422789793e-06,
      "loss": 1.61619415,
      "memory(GiB)": 97.17,
      "step": 20115,
      "train_speed(iter/s)": 1.631344
    },
    {
      "acc": 0.65672588,
      "epoch": 0.5104008117706748,
      "grad_norm": 6.15625,
      "learning_rate": 8.892307295274654e-06,
      "loss": 1.65938377,
      "memory(GiB)": 97.17,
      "step": 20120,
      "train_speed(iter/s)": 1.63139
    },
    {
      "acc": 0.65587487,
      "epoch": 0.5105276509386099,
      "grad_norm": 6.78125,
      "learning_rate": 8.89164899655846e-06,
      "loss": 1.62233658,
      "memory(GiB)": 97.17,
      "step": 20125,
      "train_speed(iter/s)": 1.631434
    },
    {
      "acc": 0.67769499,
      "epoch": 0.510654490106545,
      "grad_norm": 5.46875,
      "learning_rate": 8.89099052667017e-06,
      "loss": 1.50240917,
      "memory(GiB)": 97.17,
      "step": 20130,
      "train_speed(iter/s)": 1.631474
    },
    {
      "acc": 0.64973674,
      "epoch": 0.51078132927448,
      "grad_norm": 5.34375,
      "learning_rate": 8.890331885638744e-06,
      "loss": 1.65610256,
      "memory(GiB)": 97.17,
      "step": 20135,
      "train_speed(iter/s)": 1.63152
    },
    {
      "acc": 0.65786638,
      "epoch": 0.510908168442415,
      "grad_norm": 4.96875,
      "learning_rate": 8.889673073493151e-06,
      "loss": 1.59287968,
      "memory(GiB)": 97.17,
      "step": 20140,
      "train_speed(iter/s)": 1.631562
    },
    {
      "acc": 0.65741429,
      "epoch": 0.51103500761035,
      "grad_norm": 6.125,
      "learning_rate": 8.889014090262371e-06,
      "loss": 1.66737022,
      "memory(GiB)": 97.17,
      "step": 20145,
      "train_speed(iter/s)": 1.631592
    },
    {
      "acc": 0.64493418,
      "epoch": 0.5111618467782851,
      "grad_norm": 5.5625,
      "learning_rate": 8.888354935975388e-06,
      "loss": 1.56829071,
      "memory(GiB)": 97.17,
      "step": 20150,
      "train_speed(iter/s)": 1.631633
    },
    {
      "acc": 0.65764189,
      "epoch": 0.5112886859462202,
      "grad_norm": 4.4375,
      "learning_rate": 8.887695610661196e-06,
      "loss": 1.58194008,
      "memory(GiB)": 97.17,
      "step": 20155,
      "train_speed(iter/s)": 1.631675
    },
    {
      "acc": 0.65194035,
      "epoch": 0.5114155251141552,
      "grad_norm": 5.40625,
      "learning_rate": 8.887036114348792e-06,
      "loss": 1.67314415,
      "memory(GiB)": 97.17,
      "step": 20160,
      "train_speed(iter/s)": 1.631718
    },
    {
      "acc": 0.6675209,
      "epoch": 0.5115423642820903,
      "grad_norm": 5.40625,
      "learning_rate": 8.886376447067186e-06,
      "loss": 1.63950768,
      "memory(GiB)": 97.17,
      "step": 20165,
      "train_speed(iter/s)": 1.631762
    },
    {
      "acc": 0.64694824,
      "epoch": 0.5116692034500254,
      "grad_norm": 6.59375,
      "learning_rate": 8.885716608845394e-06,
      "loss": 1.64419384,
      "memory(GiB)": 97.17,
      "step": 20170,
      "train_speed(iter/s)": 1.631803
    },
    {
      "acc": 0.65578842,
      "epoch": 0.5117960426179604,
      "grad_norm": 4.8125,
      "learning_rate": 8.885056599712436e-06,
      "loss": 1.58569765,
      "memory(GiB)": 97.17,
      "step": 20175,
      "train_speed(iter/s)": 1.631846
    },
    {
      "acc": 0.64497242,
      "epoch": 0.5119228817858955,
      "grad_norm": 5.8125,
      "learning_rate": 8.884396419697343e-06,
      "loss": 1.65522842,
      "memory(GiB)": 97.17,
      "step": 20180,
      "train_speed(iter/s)": 1.631887
    },
    {
      "acc": 0.65047626,
      "epoch": 0.5120497209538305,
      "grad_norm": 5.875,
      "learning_rate": 8.883736068829151e-06,
      "loss": 1.58411598,
      "memory(GiB)": 97.17,
      "step": 20185,
      "train_speed(iter/s)": 1.631933
    },
    {
      "acc": 0.65407124,
      "epoch": 0.5121765601217656,
      "grad_norm": 5.34375,
      "learning_rate": 8.88307554713691e-06,
      "loss": 1.59185514,
      "memory(GiB)": 97.17,
      "step": 20190,
      "train_speed(iter/s)": 1.631977
    },
    {
      "acc": 0.66403227,
      "epoch": 0.5123033992897007,
      "grad_norm": 5.625,
      "learning_rate": 8.882414854649667e-06,
      "loss": 1.58980694,
      "memory(GiB)": 97.17,
      "step": 20195,
      "train_speed(iter/s)": 1.632023
    },
    {
      "acc": 0.65741043,
      "epoch": 0.5124302384576357,
      "grad_norm": 5.375,
      "learning_rate": 8.881753991396488e-06,
      "loss": 1.64481468,
      "memory(GiB)": 97.17,
      "step": 20200,
      "train_speed(iter/s)": 1.632064
    },
    {
      "acc": 0.65740623,
      "epoch": 0.5125570776255708,
      "grad_norm": 5.8125,
      "learning_rate": 8.881092957406436e-06,
      "loss": 1.5566761,
      "memory(GiB)": 97.17,
      "step": 20205,
      "train_speed(iter/s)": 1.632106
    },
    {
      "acc": 0.65452919,
      "epoch": 0.5126839167935059,
      "grad_norm": 6.0625,
      "learning_rate": 8.880431752708588e-06,
      "loss": 1.5680829,
      "memory(GiB)": 97.17,
      "step": 20210,
      "train_speed(iter/s)": 1.632146
    },
    {
      "acc": 0.67387424,
      "epoch": 0.5128107559614409,
      "grad_norm": 5.75,
      "learning_rate": 8.879770377332026e-06,
      "loss": 1.55310574,
      "memory(GiB)": 97.17,
      "step": 20215,
      "train_speed(iter/s)": 1.632191
    },
    {
      "acc": 0.656464,
      "epoch": 0.512937595129376,
      "grad_norm": 6.40625,
      "learning_rate": 8.879108831305842e-06,
      "loss": 1.60006008,
      "memory(GiB)": 97.17,
      "step": 20220,
      "train_speed(iter/s)": 1.632233
    },
    {
      "acc": 0.6463912,
      "epoch": 0.513064434297311,
      "grad_norm": 5.9375,
      "learning_rate": 8.878447114659131e-06,
      "loss": 1.61846733,
      "memory(GiB)": 97.17,
      "step": 20225,
      "train_speed(iter/s)": 1.632278
    },
    {
      "acc": 0.67182813,
      "epoch": 0.5131912734652461,
      "grad_norm": 7.5,
      "learning_rate": 8.877785227421003e-06,
      "loss": 1.53214722,
      "memory(GiB)": 97.17,
      "step": 20230,
      "train_speed(iter/s)": 1.632324
    },
    {
      "acc": 0.66676221,
      "epoch": 0.5133181126331812,
      "grad_norm": 6.375,
      "learning_rate": 8.877123169620565e-06,
      "loss": 1.59670267,
      "memory(GiB)": 97.17,
      "step": 20235,
      "train_speed(iter/s)": 1.632369
    },
    {
      "acc": 0.64495134,
      "epoch": 0.5134449518011162,
      "grad_norm": 5.0,
      "learning_rate": 8.876460941286941e-06,
      "loss": 1.62505302,
      "memory(GiB)": 97.17,
      "step": 20240,
      "train_speed(iter/s)": 1.632411
    },
    {
      "acc": 0.63942709,
      "epoch": 0.5135717909690513,
      "grad_norm": 5.96875,
      "learning_rate": 8.875798542449256e-06,
      "loss": 1.66225662,
      "memory(GiB)": 97.17,
      "step": 20245,
      "train_speed(iter/s)": 1.632457
    },
    {
      "acc": 0.63722963,
      "epoch": 0.5136986301369864,
      "grad_norm": 5.5,
      "learning_rate": 8.87513597313665e-06,
      "loss": 1.67115879,
      "memory(GiB)": 97.17,
      "step": 20250,
      "train_speed(iter/s)": 1.632501
    },
    {
      "acc": 0.64086881,
      "epoch": 0.5138254693049213,
      "grad_norm": 6.75,
      "learning_rate": 8.87447323337826e-06,
      "loss": 1.6811409,
      "memory(GiB)": 97.17,
      "step": 20255,
      "train_speed(iter/s)": 1.632544
    },
    {
      "acc": 0.67046118,
      "epoch": 0.5139523084728564,
      "grad_norm": 5.71875,
      "learning_rate": 8.87381032320324e-06,
      "loss": 1.5784833,
      "memory(GiB)": 97.17,
      "step": 20260,
      "train_speed(iter/s)": 1.632584
    },
    {
      "acc": 0.64712725,
      "epoch": 0.5140791476407914,
      "grad_norm": 5.46875,
      "learning_rate": 8.873147242640746e-06,
      "loss": 1.63043079,
      "memory(GiB)": 97.17,
      "step": 20265,
      "train_speed(iter/s)": 1.632628
    },
    {
      "acc": 0.64511232,
      "epoch": 0.5142059868087265,
      "grad_norm": 4.59375,
      "learning_rate": 8.872483991719944e-06,
      "loss": 1.65729504,
      "memory(GiB)": 97.17,
      "step": 20270,
      "train_speed(iter/s)": 1.632671
    },
    {
      "acc": 0.63720894,
      "epoch": 0.5143328259766616,
      "grad_norm": 5.125,
      "learning_rate": 8.871820570470009e-06,
      "loss": 1.67285767,
      "memory(GiB)": 97.17,
      "step": 20275,
      "train_speed(iter/s)": 1.63271
    },
    {
      "acc": 0.65420094,
      "epoch": 0.5144596651445966,
      "grad_norm": 8.6875,
      "learning_rate": 8.871156978920116e-06,
      "loss": 1.65575237,
      "memory(GiB)": 97.17,
      "step": 20280,
      "train_speed(iter/s)": 1.632756
    },
    {
      "acc": 0.66418495,
      "epoch": 0.5145865043125317,
      "grad_norm": 5.6875,
      "learning_rate": 8.870493217099456e-06,
      "loss": 1.50092735,
      "memory(GiB)": 97.17,
      "step": 20285,
      "train_speed(iter/s)": 1.632797
    },
    {
      "acc": 0.65824113,
      "epoch": 0.5147133434804668,
      "grad_norm": 5.34375,
      "learning_rate": 8.869829285037224e-06,
      "loss": 1.53967924,
      "memory(GiB)": 97.17,
      "step": 20290,
      "train_speed(iter/s)": 1.632841
    },
    {
      "acc": 0.64342742,
      "epoch": 0.5148401826484018,
      "grad_norm": 6.59375,
      "learning_rate": 8.869165182762623e-06,
      "loss": 1.66172791,
      "memory(GiB)": 97.17,
      "step": 20295,
      "train_speed(iter/s)": 1.632884
    },
    {
      "acc": 0.65398817,
      "epoch": 0.5149670218163369,
      "grad_norm": 5.53125,
      "learning_rate": 8.868500910304863e-06,
      "loss": 1.65386562,
      "memory(GiB)": 97.17,
      "step": 20300,
      "train_speed(iter/s)": 1.632927
    },
    {
      "acc": 0.65053244,
      "epoch": 0.5150938609842719,
      "grad_norm": 4.9375,
      "learning_rate": 8.86783646769316e-06,
      "loss": 1.59877939,
      "memory(GiB)": 97.17,
      "step": 20305,
      "train_speed(iter/s)": 1.63297
    },
    {
      "acc": 0.66531544,
      "epoch": 0.515220700152207,
      "grad_norm": 5.21875,
      "learning_rate": 8.867171854956742e-06,
      "loss": 1.6009716,
      "memory(GiB)": 97.17,
      "step": 20310,
      "train_speed(iter/s)": 1.633011
    },
    {
      "acc": 0.64752612,
      "epoch": 0.5153475393201421,
      "grad_norm": 8.3125,
      "learning_rate": 8.86650707212484e-06,
      "loss": 1.62827721,
      "memory(GiB)": 97.17,
      "step": 20315,
      "train_speed(iter/s)": 1.633054
    },
    {
      "acc": 0.65645323,
      "epoch": 0.5154743784880771,
      "grad_norm": 5.65625,
      "learning_rate": 8.865842119226693e-06,
      "loss": 1.63121605,
      "memory(GiB)": 97.17,
      "step": 20320,
      "train_speed(iter/s)": 1.6331
    },
    {
      "acc": 0.64139228,
      "epoch": 0.5156012176560122,
      "grad_norm": 6.15625,
      "learning_rate": 8.86517699629155e-06,
      "loss": 1.68119621,
      "memory(GiB)": 97.17,
      "step": 20325,
      "train_speed(iter/s)": 1.633144
    },
    {
      "acc": 0.64034743,
      "epoch": 0.5157280568239473,
      "grad_norm": 5.09375,
      "learning_rate": 8.864511703348666e-06,
      "loss": 1.65172272,
      "memory(GiB)": 97.17,
      "step": 20330,
      "train_speed(iter/s)": 1.633186
    },
    {
      "acc": 0.64373608,
      "epoch": 0.5158548959918823,
      "grad_norm": 6.5,
      "learning_rate": 8.863846240427306e-06,
      "loss": 1.68302612,
      "memory(GiB)": 97.17,
      "step": 20335,
      "train_speed(iter/s)": 1.633227
    },
    {
      "acc": 0.66251297,
      "epoch": 0.5159817351598174,
      "grad_norm": 6.28125,
      "learning_rate": 8.863180607556733e-06,
      "loss": 1.51638041,
      "memory(GiB)": 97.17,
      "step": 20340,
      "train_speed(iter/s)": 1.633269
    },
    {
      "acc": 0.64889421,
      "epoch": 0.5161085743277524,
      "grad_norm": 6.5,
      "learning_rate": 8.862514804766234e-06,
      "loss": 1.57986498,
      "memory(GiB)": 97.17,
      "step": 20345,
      "train_speed(iter/s)": 1.633314
    },
    {
      "acc": 0.64768066,
      "epoch": 0.5162354134956875,
      "grad_norm": 6.03125,
      "learning_rate": 8.861848832085084e-06,
      "loss": 1.62796288,
      "memory(GiB)": 97.17,
      "step": 20350,
      "train_speed(iter/s)": 1.63336
    },
    {
      "acc": 0.65385237,
      "epoch": 0.5163622526636226,
      "grad_norm": 5.78125,
      "learning_rate": 8.861182689542585e-06,
      "loss": 1.60219955,
      "memory(GiB)": 97.17,
      "step": 20355,
      "train_speed(iter/s)": 1.633401
    },
    {
      "acc": 0.65883746,
      "epoch": 0.5164890918315576,
      "grad_norm": 5.90625,
      "learning_rate": 8.86051637716803e-06,
      "loss": 1.55787125,
      "memory(GiB)": 97.17,
      "step": 20360,
      "train_speed(iter/s)": 1.633445
    },
    {
      "acc": 0.64599462,
      "epoch": 0.5166159309994927,
      "grad_norm": 5.4375,
      "learning_rate": 8.859849894990728e-06,
      "loss": 1.66964417,
      "memory(GiB)": 97.17,
      "step": 20365,
      "train_speed(iter/s)": 1.633488
    },
    {
      "acc": 0.64820824,
      "epoch": 0.5167427701674278,
      "grad_norm": 5.53125,
      "learning_rate": 8.859183243039995e-06,
      "loss": 1.65293179,
      "memory(GiB)": 97.17,
      "step": 20370,
      "train_speed(iter/s)": 1.63353
    },
    {
      "acc": 0.63783326,
      "epoch": 0.5168696093353627,
      "grad_norm": 6.03125,
      "learning_rate": 8.858516421345154e-06,
      "loss": 1.63754845,
      "memory(GiB)": 97.17,
      "step": 20375,
      "train_speed(iter/s)": 1.633574
    },
    {
      "acc": 0.65004282,
      "epoch": 0.5169964485032978,
      "grad_norm": 5.65625,
      "learning_rate": 8.857849429935534e-06,
      "loss": 1.66275406,
      "memory(GiB)": 97.17,
      "step": 20380,
      "train_speed(iter/s)": 1.633617
    },
    {
      "acc": 0.65984912,
      "epoch": 0.5171232876712328,
      "grad_norm": 6.3125,
      "learning_rate": 8.857182268840471e-06,
      "loss": 1.6439785,
      "memory(GiB)": 97.17,
      "step": 20385,
      "train_speed(iter/s)": 1.633661
    },
    {
      "acc": 0.65073705,
      "epoch": 0.5172501268391679,
      "grad_norm": 5.1875,
      "learning_rate": 8.856514938089312e-06,
      "loss": 1.60009708,
      "memory(GiB)": 97.17,
      "step": 20390,
      "train_speed(iter/s)": 1.633701
    },
    {
      "acc": 0.6440773,
      "epoch": 0.517376966007103,
      "grad_norm": 5.59375,
      "learning_rate": 8.855847437711407e-06,
      "loss": 1.70226879,
      "memory(GiB)": 97.17,
      "step": 20395,
      "train_speed(iter/s)": 1.633744
    },
    {
      "acc": 0.64767084,
      "epoch": 0.517503805175038,
      "grad_norm": 6.25,
      "learning_rate": 8.855179767736117e-06,
      "loss": 1.65001278,
      "memory(GiB)": 97.17,
      "step": 20400,
      "train_speed(iter/s)": 1.633788
    },
    {
      "acc": 0.64709287,
      "epoch": 0.5176306443429731,
      "grad_norm": 5.34375,
      "learning_rate": 8.854511928192808e-06,
      "loss": 1.60745468,
      "memory(GiB)": 97.17,
      "step": 20405,
      "train_speed(iter/s)": 1.63383
    },
    {
      "acc": 0.64274278,
      "epoch": 0.5177574835109082,
      "grad_norm": 5.65625,
      "learning_rate": 8.853843919110856e-06,
      "loss": 1.57408152,
      "memory(GiB)": 97.17,
      "step": 20410,
      "train_speed(iter/s)": 1.633871
    },
    {
      "acc": 0.64803972,
      "epoch": 0.5178843226788432,
      "grad_norm": 6.25,
      "learning_rate": 8.853175740519642e-06,
      "loss": 1.6757473,
      "memory(GiB)": 97.17,
      "step": 20415,
      "train_speed(iter/s)": 1.633918
    },
    {
      "acc": 0.6625638,
      "epoch": 0.5180111618467783,
      "grad_norm": 6.09375,
      "learning_rate": 8.852507392448555e-06,
      "loss": 1.58051805,
      "memory(GiB)": 97.17,
      "step": 20420,
      "train_speed(iter/s)": 1.633961
    },
    {
      "acc": 0.65181627,
      "epoch": 0.5181380010147133,
      "grad_norm": 6.9375,
      "learning_rate": 8.851838874926994e-06,
      "loss": 1.63936996,
      "memory(GiB)": 97.17,
      "step": 20425,
      "train_speed(iter/s)": 1.634003
    },
    {
      "acc": 0.64780436,
      "epoch": 0.5182648401826484,
      "grad_norm": 6.65625,
      "learning_rate": 8.851170187984362e-06,
      "loss": 1.61182823,
      "memory(GiB)": 97.17,
      "step": 20430,
      "train_speed(iter/s)": 1.634043
    },
    {
      "acc": 0.65935707,
      "epoch": 0.5183916793505835,
      "grad_norm": 6.34375,
      "learning_rate": 8.850501331650069e-06,
      "loss": 1.57177334,
      "memory(GiB)": 97.17,
      "step": 20435,
      "train_speed(iter/s)": 1.634087
    },
    {
      "acc": 0.65161285,
      "epoch": 0.5185185185185185,
      "grad_norm": 5.71875,
      "learning_rate": 8.849832305953536e-06,
      "loss": 1.66026421,
      "memory(GiB)": 97.17,
      "step": 20440,
      "train_speed(iter/s)": 1.634129
    },
    {
      "acc": 0.65268941,
      "epoch": 0.5186453576864536,
      "grad_norm": 4.59375,
      "learning_rate": 8.849163110924193e-06,
      "loss": 1.61938477,
      "memory(GiB)": 97.17,
      "step": 20445,
      "train_speed(iter/s)": 1.634172
    },
    {
      "acc": 0.64094872,
      "epoch": 0.5187721968543887,
      "grad_norm": 5.0625,
      "learning_rate": 8.84849374659147e-06,
      "loss": 1.67266521,
      "memory(GiB)": 97.17,
      "step": 20450,
      "train_speed(iter/s)": 1.634212
    },
    {
      "acc": 0.65699377,
      "epoch": 0.5188990360223237,
      "grad_norm": 4.59375,
      "learning_rate": 8.847824212984807e-06,
      "loss": 1.55342484,
      "memory(GiB)": 97.17,
      "step": 20455,
      "train_speed(iter/s)": 1.634258
    },
    {
      "acc": 0.66278982,
      "epoch": 0.5190258751902588,
      "grad_norm": 5.3125,
      "learning_rate": 8.84715451013366e-06,
      "loss": 1.50579128,
      "memory(GiB)": 97.17,
      "step": 20460,
      "train_speed(iter/s)": 1.6343
    },
    {
      "acc": 0.65456476,
      "epoch": 0.5191527143581938,
      "grad_norm": 6.59375,
      "learning_rate": 8.846484638067478e-06,
      "loss": 1.5660162,
      "memory(GiB)": 97.17,
      "step": 20465,
      "train_speed(iter/s)": 1.634344
    },
    {
      "acc": 0.64687386,
      "epoch": 0.5192795535261289,
      "grad_norm": 5.21875,
      "learning_rate": 8.84581459681573e-06,
      "loss": 1.62442856,
      "memory(GiB)": 97.17,
      "step": 20470,
      "train_speed(iter/s)": 1.634388
    },
    {
      "acc": 0.66794186,
      "epoch": 0.519406392694064,
      "grad_norm": 6.15625,
      "learning_rate": 8.845144386407884e-06,
      "loss": 1.59475269,
      "memory(GiB)": 97.17,
      "step": 20475,
      "train_speed(iter/s)": 1.634424
    },
    {
      "acc": 0.64997854,
      "epoch": 0.519533231861999,
      "grad_norm": 5.0625,
      "learning_rate": 8.844474006873422e-06,
      "loss": 1.63259506,
      "memory(GiB)": 97.17,
      "step": 20480,
      "train_speed(iter/s)": 1.634466
    },
    {
      "acc": 0.6645627,
      "epoch": 0.5196600710299341,
      "grad_norm": 6.03125,
      "learning_rate": 8.84380345824183e-06,
      "loss": 1.55536518,
      "memory(GiB)": 97.17,
      "step": 20485,
      "train_speed(iter/s)": 1.634508
    },
    {
      "acc": 0.6481904,
      "epoch": 0.5197869101978692,
      "grad_norm": 6.34375,
      "learning_rate": 8.843132740542599e-06,
      "loss": 1.63135262,
      "memory(GiB)": 97.17,
      "step": 20490,
      "train_speed(iter/s)": 1.634552
    },
    {
      "acc": 0.66286187,
      "epoch": 0.5199137493658041,
      "grad_norm": 12.875,
      "learning_rate": 8.842461853805232e-06,
      "loss": 1.52390251,
      "memory(GiB)": 97.17,
      "step": 20495,
      "train_speed(iter/s)": 1.634593
    },
    {
      "acc": 0.65427513,
      "epoch": 0.5200405885337392,
      "grad_norm": 5.65625,
      "learning_rate": 8.841790798059237e-06,
      "loss": 1.60269718,
      "memory(GiB)": 97.17,
      "step": 20500,
      "train_speed(iter/s)": 1.634634
    },
    {
      "acc": 0.67601938,
      "epoch": 0.5201674277016742,
      "grad_norm": 6.6875,
      "learning_rate": 8.84111957333413e-06,
      "loss": 1.52135229,
      "memory(GiB)": 97.17,
      "step": 20505,
      "train_speed(iter/s)": 1.634678
    },
    {
      "acc": 0.63770156,
      "epoch": 0.5202942668696093,
      "grad_norm": 6.0,
      "learning_rate": 8.840448179659436e-06,
      "loss": 1.74112816,
      "memory(GiB)": 97.17,
      "step": 20510,
      "train_speed(iter/s)": 1.634722
    },
    {
      "acc": 0.64741273,
      "epoch": 0.5204211060375444,
      "grad_norm": 5.03125,
      "learning_rate": 8.839776617064683e-06,
      "loss": 1.63785095,
      "memory(GiB)": 97.17,
      "step": 20515,
      "train_speed(iter/s)": 1.634763
    },
    {
      "acc": 0.66083527,
      "epoch": 0.5205479452054794,
      "grad_norm": 4.96875,
      "learning_rate": 8.839104885579413e-06,
      "loss": 1.61422634,
      "memory(GiB)": 97.17,
      "step": 20520,
      "train_speed(iter/s)": 1.634806
    },
    {
      "acc": 0.64398055,
      "epoch": 0.5206747843734145,
      "grad_norm": 6.4375,
      "learning_rate": 8.83843298523317e-06,
      "loss": 1.65961456,
      "memory(GiB)": 97.17,
      "step": 20525,
      "train_speed(iter/s)": 1.63485
    },
    {
      "acc": 0.64862199,
      "epoch": 0.5208016235413496,
      "grad_norm": 6.0625,
      "learning_rate": 8.837760916055505e-06,
      "loss": 1.64898834,
      "memory(GiB)": 97.17,
      "step": 20530,
      "train_speed(iter/s)": 1.634891
    },
    {
      "acc": 0.64246101,
      "epoch": 0.5209284627092846,
      "grad_norm": 5.5625,
      "learning_rate": 8.837088678075983e-06,
      "loss": 1.69156151,
      "memory(GiB)": 97.17,
      "step": 20535,
      "train_speed(iter/s)": 1.634933
    },
    {
      "acc": 0.6487082,
      "epoch": 0.5210553018772197,
      "grad_norm": 4.78125,
      "learning_rate": 8.836416271324166e-06,
      "loss": 1.62909603,
      "memory(GiB)": 97.17,
      "step": 20540,
      "train_speed(iter/s)": 1.634976
    },
    {
      "acc": 0.67679324,
      "epoch": 0.5211821410451547,
      "grad_norm": 4.5,
      "learning_rate": 8.835743695829635e-06,
      "loss": 1.49806566,
      "memory(GiB)": 97.17,
      "step": 20545,
      "train_speed(iter/s)": 1.635021
    },
    {
      "acc": 0.64529123,
      "epoch": 0.5213089802130898,
      "grad_norm": 6.78125,
      "learning_rate": 8.835070951621971e-06,
      "loss": 1.66490364,
      "memory(GiB)": 97.17,
      "step": 20550,
      "train_speed(iter/s)": 1.635065
    },
    {
      "acc": 0.65202789,
      "epoch": 0.5214358193810249,
      "grad_norm": 6.03125,
      "learning_rate": 8.834398038730765e-06,
      "loss": 1.61711502,
      "memory(GiB)": 97.17,
      "step": 20555,
      "train_speed(iter/s)": 1.635108
    },
    {
      "acc": 0.68759584,
      "epoch": 0.5215626585489599,
      "grad_norm": 6.5,
      "learning_rate": 8.833724957185612e-06,
      "loss": 1.49708958,
      "memory(GiB)": 97.17,
      "step": 20560,
      "train_speed(iter/s)": 1.635152
    },
    {
      "acc": 0.64750705,
      "epoch": 0.521689497716895,
      "grad_norm": 6.6875,
      "learning_rate": 8.83305170701612e-06,
      "loss": 1.67592335,
      "memory(GiB)": 97.17,
      "step": 20565,
      "train_speed(iter/s)": 1.635195
    },
    {
      "acc": 0.66132841,
      "epoch": 0.5218163368848301,
      "grad_norm": 5.53125,
      "learning_rate": 8.832378288251902e-06,
      "loss": 1.61383629,
      "memory(GiB)": 97.17,
      "step": 20570,
      "train_speed(iter/s)": 1.635241
    },
    {
      "acc": 0.65200987,
      "epoch": 0.5219431760527651,
      "grad_norm": 5.8125,
      "learning_rate": 8.831704700922574e-06,
      "loss": 1.64687843,
      "memory(GiB)": 97.17,
      "step": 20575,
      "train_speed(iter/s)": 1.635282
    },
    {
      "acc": 0.65637298,
      "epoch": 0.5220700152207002,
      "grad_norm": 5.96875,
      "learning_rate": 8.831030945057767e-06,
      "loss": 1.60000801,
      "memory(GiB)": 97.17,
      "step": 20580,
      "train_speed(iter/s)": 1.635324
    },
    {
      "acc": 0.66913633,
      "epoch": 0.5221968543886352,
      "grad_norm": 5.375,
      "learning_rate": 8.830357020687115e-06,
      "loss": 1.52004366,
      "memory(GiB)": 97.17,
      "step": 20585,
      "train_speed(iter/s)": 1.635364
    },
    {
      "acc": 0.65937729,
      "epoch": 0.5223236935565703,
      "grad_norm": 5.8125,
      "learning_rate": 8.82968292784026e-06,
      "loss": 1.62688942,
      "memory(GiB)": 97.17,
      "step": 20590,
      "train_speed(iter/s)": 1.635407
    },
    {
      "acc": 0.65578389,
      "epoch": 0.5224505327245054,
      "grad_norm": 6.40625,
      "learning_rate": 8.82900866654685e-06,
      "loss": 1.60818672,
      "memory(GiB)": 97.17,
      "step": 20595,
      "train_speed(iter/s)": 1.63545
    },
    {
      "acc": 0.65954952,
      "epoch": 0.5225773718924404,
      "grad_norm": 4.875,
      "learning_rate": 8.828334236836546e-06,
      "loss": 1.59411736,
      "memory(GiB)": 97.17,
      "step": 20600,
      "train_speed(iter/s)": 1.635489
    },
    {
      "acc": 0.64722118,
      "epoch": 0.5227042110603755,
      "grad_norm": 5.0625,
      "learning_rate": 8.827659638739007e-06,
      "loss": 1.66816025,
      "memory(GiB)": 97.17,
      "step": 20605,
      "train_speed(iter/s)": 1.635531
    },
    {
      "acc": 0.65049634,
      "epoch": 0.5228310502283106,
      "grad_norm": 5.40625,
      "learning_rate": 8.82698487228391e-06,
      "loss": 1.58742905,
      "memory(GiB)": 97.17,
      "step": 20610,
      "train_speed(iter/s)": 1.635574
    },
    {
      "acc": 0.64607558,
      "epoch": 0.5229578893962455,
      "grad_norm": 5.5625,
      "learning_rate": 8.826309937500932e-06,
      "loss": 1.60876427,
      "memory(GiB)": 97.17,
      "step": 20615,
      "train_speed(iter/s)": 1.635616
    },
    {
      "acc": 0.64737964,
      "epoch": 0.5230847285641806,
      "grad_norm": 5.1875,
      "learning_rate": 8.825634834419758e-06,
      "loss": 1.66019268,
      "memory(GiB)": 97.17,
      "step": 20620,
      "train_speed(iter/s)": 1.635659
    },
    {
      "acc": 0.66241264,
      "epoch": 0.5232115677321156,
      "grad_norm": 7.03125,
      "learning_rate": 8.824959563070085e-06,
      "loss": 1.5532692,
      "memory(GiB)": 97.17,
      "step": 20625,
      "train_speed(iter/s)": 1.6357
    },
    {
      "acc": 0.6638639,
      "epoch": 0.5233384069000507,
      "grad_norm": 5.1875,
      "learning_rate": 8.824284123481614e-06,
      "loss": 1.57181873,
      "memory(GiB)": 97.17,
      "step": 20630,
      "train_speed(iter/s)": 1.635744
    },
    {
      "acc": 0.64825201,
      "epoch": 0.5234652460679858,
      "grad_norm": 5.40625,
      "learning_rate": 8.823608515684053e-06,
      "loss": 1.64821243,
      "memory(GiB)": 97.17,
      "step": 20635,
      "train_speed(iter/s)": 1.635787
    },
    {
      "acc": 0.64459519,
      "epoch": 0.5235920852359208,
      "grad_norm": 4.65625,
      "learning_rate": 8.822932739707118e-06,
      "loss": 1.67736187,
      "memory(GiB)": 97.17,
      "step": 20640,
      "train_speed(iter/s)": 1.635827
    },
    {
      "acc": 0.64496059,
      "epoch": 0.5237189244038559,
      "grad_norm": 5.75,
      "learning_rate": 8.822256795580532e-06,
      "loss": 1.65869064,
      "memory(GiB)": 97.17,
      "step": 20645,
      "train_speed(iter/s)": 1.635868
    },
    {
      "acc": 0.66012797,
      "epoch": 0.523845763571791,
      "grad_norm": 5.96875,
      "learning_rate": 8.821580683334027e-06,
      "loss": 1.60667648,
      "memory(GiB)": 97.17,
      "step": 20650,
      "train_speed(iter/s)": 1.635909
    },
    {
      "acc": 0.64167061,
      "epoch": 0.523972602739726,
      "grad_norm": 6.40625,
      "learning_rate": 8.820904402997343e-06,
      "loss": 1.64902191,
      "memory(GiB)": 97.17,
      "step": 20655,
      "train_speed(iter/s)": 1.635952
    },
    {
      "acc": 0.65029426,
      "epoch": 0.5240994419076611,
      "grad_norm": 5.375,
      "learning_rate": 8.820227954600222e-06,
      "loss": 1.55934143,
      "memory(GiB)": 97.17,
      "step": 20660,
      "train_speed(iter/s)": 1.635994
    },
    {
      "acc": 0.63539953,
      "epoch": 0.5242262810755961,
      "grad_norm": 5.1875,
      "learning_rate": 8.819551338172421e-06,
      "loss": 1.69757156,
      "memory(GiB)": 97.17,
      "step": 20665,
      "train_speed(iter/s)": 1.636034
    },
    {
      "acc": 0.66527739,
      "epoch": 0.5243531202435312,
      "grad_norm": 5.125,
      "learning_rate": 8.8188745537437e-06,
      "loss": 1.61686554,
      "memory(GiB)": 97.17,
      "step": 20670,
      "train_speed(iter/s)": 1.636075
    },
    {
      "acc": 0.6516263,
      "epoch": 0.5244799594114663,
      "grad_norm": 5.875,
      "learning_rate": 8.818197601343822e-06,
      "loss": 1.60345631,
      "memory(GiB)": 97.17,
      "step": 20675,
      "train_speed(iter/s)": 1.63612
    },
    {
      "acc": 0.65005751,
      "epoch": 0.5246067985794013,
      "grad_norm": 6.96875,
      "learning_rate": 8.81752048100257e-06,
      "loss": 1.59206944,
      "memory(GiB)": 97.17,
      "step": 20680,
      "train_speed(iter/s)": 1.636163
    },
    {
      "acc": 0.63440671,
      "epoch": 0.5247336377473364,
      "grad_norm": 6.03125,
      "learning_rate": 8.816843192749724e-06,
      "loss": 1.68802929,
      "memory(GiB)": 97.17,
      "step": 20685,
      "train_speed(iter/s)": 1.636204
    },
    {
      "acc": 0.65044222,
      "epoch": 0.5248604769152715,
      "grad_norm": 6.125,
      "learning_rate": 8.816165736615072e-06,
      "loss": 1.61524734,
      "memory(GiB)": 97.17,
      "step": 20690,
      "train_speed(iter/s)": 1.636247
    },
    {
      "acc": 0.65094843,
      "epoch": 0.5249873160832065,
      "grad_norm": 6.5625,
      "learning_rate": 8.815488112628412e-06,
      "loss": 1.62463493,
      "memory(GiB)": 97.17,
      "step": 20695,
      "train_speed(iter/s)": 1.636289
    },
    {
      "acc": 0.66658874,
      "epoch": 0.5251141552511416,
      "grad_norm": 5.71875,
      "learning_rate": 8.814810320819551e-06,
      "loss": 1.56076031,
      "memory(GiB)": 97.17,
      "step": 20700,
      "train_speed(iter/s)": 1.636328
    },
    {
      "acc": 0.65842886,
      "epoch": 0.5252409944190766,
      "grad_norm": 5.6875,
      "learning_rate": 8.8141323612183e-06,
      "loss": 1.61330528,
      "memory(GiB)": 97.17,
      "step": 20705,
      "train_speed(iter/s)": 1.636371
    },
    {
      "acc": 0.64636955,
      "epoch": 0.5253678335870117,
      "grad_norm": 7.40625,
      "learning_rate": 8.813454233854479e-06,
      "loss": 1.62866688,
      "memory(GiB)": 97.17,
      "step": 20710,
      "train_speed(iter/s)": 1.636414
    },
    {
      "acc": 0.65614576,
      "epoch": 0.5254946727549468,
      "grad_norm": 5.59375,
      "learning_rate": 8.812775938757914e-06,
      "loss": 1.61141777,
      "memory(GiB)": 97.17,
      "step": 20715,
      "train_speed(iter/s)": 1.636455
    },
    {
      "acc": 0.67145171,
      "epoch": 0.5256215119228818,
      "grad_norm": 6.25,
      "learning_rate": 8.812097475958442e-06,
      "loss": 1.56528406,
      "memory(GiB)": 97.17,
      "step": 20720,
      "train_speed(iter/s)": 1.636499
    },
    {
      "acc": 0.64792519,
      "epoch": 0.5257483510908169,
      "grad_norm": 5.625,
      "learning_rate": 8.811418845485902e-06,
      "loss": 1.65422173,
      "memory(GiB)": 97.17,
      "step": 20725,
      "train_speed(iter/s)": 1.636544
    },
    {
      "acc": 0.65836978,
      "epoch": 0.525875190258752,
      "grad_norm": 5.5625,
      "learning_rate": 8.810740047370146e-06,
      "loss": 1.64090042,
      "memory(GiB)": 97.17,
      "step": 20730,
      "train_speed(iter/s)": 1.636586
    },
    {
      "acc": 0.67015996,
      "epoch": 0.526002029426687,
      "grad_norm": 5.625,
      "learning_rate": 8.810061081641026e-06,
      "loss": 1.62822876,
      "memory(GiB)": 97.17,
      "step": 20735,
      "train_speed(iter/s)": 1.636627
    },
    {
      "acc": 0.66730595,
      "epoch": 0.526128868594622,
      "grad_norm": 4.625,
      "learning_rate": 8.809381948328412e-06,
      "loss": 1.58117638,
      "memory(GiB)": 97.17,
      "step": 20740,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.65899262,
      "epoch": 0.526255707762557,
      "grad_norm": 5.125,
      "learning_rate": 8.80870264746217e-06,
      "loss": 1.64216938,
      "memory(GiB)": 97.17,
      "step": 20745,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.64683342,
      "epoch": 0.5263825469304921,
      "grad_norm": 4.8125,
      "learning_rate": 8.808023179072183e-06,
      "loss": 1.607337,
      "memory(GiB)": 97.17,
      "step": 20750,
      "train_speed(iter/s)": 1.636749
    },
    {
      "acc": 0.6462841,
      "epoch": 0.5265093860984272,
      "grad_norm": 5.3125,
      "learning_rate": 8.807343543188333e-06,
      "loss": 1.68285522,
      "memory(GiB)": 97.17,
      "step": 20755,
      "train_speed(iter/s)": 1.636791
    },
    {
      "acc": 0.66174536,
      "epoch": 0.5266362252663622,
      "grad_norm": 6.03125,
      "learning_rate": 8.806663739840515e-06,
      "loss": 1.61272812,
      "memory(GiB)": 97.17,
      "step": 20760,
      "train_speed(iter/s)": 1.636835
    },
    {
      "acc": 0.64558692,
      "epoch": 0.5267630644342973,
      "grad_norm": 5.78125,
      "learning_rate": 8.805983769058633e-06,
      "loss": 1.65259418,
      "memory(GiB)": 97.17,
      "step": 20765,
      "train_speed(iter/s)": 1.636874
    },
    {
      "acc": 0.66302023,
      "epoch": 0.5268899036022324,
      "grad_norm": 6.6875,
      "learning_rate": 8.80530363087259e-06,
      "loss": 1.61344185,
      "memory(GiB)": 97.17,
      "step": 20770,
      "train_speed(iter/s)": 1.636916
    },
    {
      "acc": 0.65623193,
      "epoch": 0.5270167427701674,
      "grad_norm": 5.40625,
      "learning_rate": 8.804623325312305e-06,
      "loss": 1.66651974,
      "memory(GiB)": 97.17,
      "step": 20775,
      "train_speed(iter/s)": 1.636958
    },
    {
      "acc": 0.64837036,
      "epoch": 0.5271435819381025,
      "grad_norm": 5.21875,
      "learning_rate": 8.8039428524077e-06,
      "loss": 1.59288073,
      "memory(GiB)": 97.17,
      "step": 20780,
      "train_speed(iter/s)": 1.636999
    },
    {
      "acc": 0.66401401,
      "epoch": 0.5272704211060375,
      "grad_norm": 6.375,
      "learning_rate": 8.803262212188703e-06,
      "loss": 1.65262375,
      "memory(GiB)": 97.17,
      "step": 20785,
      "train_speed(iter/s)": 1.637042
    },
    {
      "acc": 0.65952148,
      "epoch": 0.5273972602739726,
      "grad_norm": 6.53125,
      "learning_rate": 8.802581404685255e-06,
      "loss": 1.62571354,
      "memory(GiB)": 97.17,
      "step": 20790,
      "train_speed(iter/s)": 1.637084
    },
    {
      "acc": 0.63022528,
      "epoch": 0.5275240994419077,
      "grad_norm": 5.71875,
      "learning_rate": 8.8019004299273e-06,
      "loss": 1.67102699,
      "memory(GiB)": 97.17,
      "step": 20795,
      "train_speed(iter/s)": 1.637126
    },
    {
      "acc": 0.64707384,
      "epoch": 0.5276509386098427,
      "grad_norm": 7.3125,
      "learning_rate": 8.801219287944788e-06,
      "loss": 1.64225349,
      "memory(GiB)": 97.17,
      "step": 20800,
      "train_speed(iter/s)": 1.637168
    },
    {
      "acc": 0.65275025,
      "epoch": 0.5277777777777778,
      "grad_norm": 6.28125,
      "learning_rate": 8.800537978767683e-06,
      "loss": 1.6052227,
      "memory(GiB)": 97.17,
      "step": 20805,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.66931887,
      "epoch": 0.5279046169457129,
      "grad_norm": 6.75,
      "learning_rate": 8.79985650242595e-06,
      "loss": 1.56394033,
      "memory(GiB)": 97.17,
      "step": 20810,
      "train_speed(iter/s)": 1.637255
    },
    {
      "acc": 0.64282403,
      "epoch": 0.5280314561136479,
      "grad_norm": 5.0,
      "learning_rate": 8.79917485894956e-06,
      "loss": 1.66500397,
      "memory(GiB)": 97.17,
      "step": 20815,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.65252514,
      "epoch": 0.528158295281583,
      "grad_norm": 5.15625,
      "learning_rate": 8.798493048368498e-06,
      "loss": 1.63258247,
      "memory(GiB)": 97.17,
      "step": 20820,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.65473585,
      "epoch": 0.528285134449518,
      "grad_norm": 4.59375,
      "learning_rate": 8.797811070712754e-06,
      "loss": 1.57983236,
      "memory(GiB)": 97.17,
      "step": 20825,
      "train_speed(iter/s)": 1.637377
    },
    {
      "acc": 0.64744306,
      "epoch": 0.5284119736174531,
      "grad_norm": 5.65625,
      "learning_rate": 8.797128926012323e-06,
      "loss": 1.6679306,
      "memory(GiB)": 97.17,
      "step": 20830,
      "train_speed(iter/s)": 1.637418
    },
    {
      "acc": 0.65198712,
      "epoch": 0.5285388127853882,
      "grad_norm": 6.96875,
      "learning_rate": 8.796446614297208e-06,
      "loss": 1.61573257,
      "memory(GiB)": 97.17,
      "step": 20835,
      "train_speed(iter/s)": 1.637459
    },
    {
      "acc": 0.65284791,
      "epoch": 0.5286656519533232,
      "grad_norm": 6.0625,
      "learning_rate": 8.795764135597421e-06,
      "loss": 1.62490902,
      "memory(GiB)": 97.17,
      "step": 20840,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.65560951,
      "epoch": 0.5287924911212583,
      "grad_norm": 4.78125,
      "learning_rate": 8.79508148994298e-06,
      "loss": 1.62684975,
      "memory(GiB)": 97.17,
      "step": 20845,
      "train_speed(iter/s)": 1.637539
    },
    {
      "acc": 0.64139962,
      "epoch": 0.5289193302891934,
      "grad_norm": 5.65625,
      "learning_rate": 8.794398677363913e-06,
      "loss": 1.642066,
      "memory(GiB)": 97.17,
      "step": 20850,
      "train_speed(iter/s)": 1.636889
    },
    {
      "acc": 0.66660652,
      "epoch": 0.5290461694571283,
      "grad_norm": 5.6875,
      "learning_rate": 8.79371569789025e-06,
      "loss": 1.55384274,
      "memory(GiB)": 97.17,
      "step": 20855,
      "train_speed(iter/s)": 1.63693
    },
    {
      "acc": 0.65368643,
      "epoch": 0.5291730086250634,
      "grad_norm": 6.15625,
      "learning_rate": 8.793032551552034e-06,
      "loss": 1.62286797,
      "memory(GiB)": 97.17,
      "step": 20860,
      "train_speed(iter/s)": 1.636973
    },
    {
      "acc": 0.6497963,
      "epoch": 0.5292998477929984,
      "grad_norm": 5.875,
      "learning_rate": 8.792349238379311e-06,
      "loss": 1.64692841,
      "memory(GiB)": 97.17,
      "step": 20865,
      "train_speed(iter/s)": 1.637015
    },
    {
      "acc": 0.63457546,
      "epoch": 0.5294266869609335,
      "grad_norm": 5.65625,
      "learning_rate": 8.791665758402137e-06,
      "loss": 1.6945919,
      "memory(GiB)": 97.17,
      "step": 20870,
      "train_speed(iter/s)": 1.637055
    },
    {
      "acc": 0.66274824,
      "epoch": 0.5295535261288686,
      "grad_norm": 5.875,
      "learning_rate": 8.790982111650574e-06,
      "loss": 1.60294247,
      "memory(GiB)": 97.17,
      "step": 20875,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.65766006,
      "epoch": 0.5296803652968036,
      "grad_norm": 5.09375,
      "learning_rate": 8.790298298154694e-06,
      "loss": 1.56191568,
      "memory(GiB)": 97.17,
      "step": 20880,
      "train_speed(iter/s)": 1.636424
    },
    {
      "acc": 0.64680061,
      "epoch": 0.5298072044647387,
      "grad_norm": 4.71875,
      "learning_rate": 8.78961431794457e-06,
      "loss": 1.63118896,
      "memory(GiB)": 97.17,
      "step": 20885,
      "train_speed(iter/s)": 1.635778
    },
    {
      "acc": 0.65813251,
      "epoch": 0.5299340436326738,
      "grad_norm": 5.375,
      "learning_rate": 8.78893017105029e-06,
      "loss": 1.59893961,
      "memory(GiB)": 97.17,
      "step": 20890,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.64986544,
      "epoch": 0.5300608828006088,
      "grad_norm": 5.6875,
      "learning_rate": 8.788245857501944e-06,
      "loss": 1.58806782,
      "memory(GiB)": 97.17,
      "step": 20895,
      "train_speed(iter/s)": 1.635863
    },
    {
      "acc": 0.64958639,
      "epoch": 0.5301877219685439,
      "grad_norm": 5.65625,
      "learning_rate": 8.787561377329633e-06,
      "loss": 1.66650715,
      "memory(GiB)": 97.17,
      "step": 20900,
      "train_speed(iter/s)": 1.635905
    },
    {
      "acc": 0.6603848,
      "epoch": 0.5303145611364789,
      "grad_norm": 5.5625,
      "learning_rate": 8.786876730563462e-06,
      "loss": 1.63459206,
      "memory(GiB)": 97.17,
      "step": 20905,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.65420365,
      "epoch": 0.530441400304414,
      "grad_norm": 7.3125,
      "learning_rate": 8.786191917233545e-06,
      "loss": 1.6224205,
      "memory(GiB)": 97.17,
      "step": 20910,
      "train_speed(iter/s)": 1.635989
    },
    {
      "acc": 0.66148334,
      "epoch": 0.5305682394723491,
      "grad_norm": 5.59375,
      "learning_rate": 8.785506937370003e-06,
      "loss": 1.57055511,
      "memory(GiB)": 97.17,
      "step": 20915,
      "train_speed(iter/s)": 1.636032
    },
    {
      "acc": 0.66252542,
      "epoch": 0.5306950786402841,
      "grad_norm": 6.84375,
      "learning_rate": 8.784821791002965e-06,
      "loss": 1.63754253,
      "memory(GiB)": 97.17,
      "step": 20920,
      "train_speed(iter/s)": 1.636074
    },
    {
      "acc": 0.6593853,
      "epoch": 0.5308219178082192,
      "grad_norm": 7.28125,
      "learning_rate": 8.784136478162567e-06,
      "loss": 1.6666605,
      "memory(GiB)": 97.17,
      "step": 20925,
      "train_speed(iter/s)": 1.636117
    },
    {
      "acc": 0.66064806,
      "epoch": 0.5309487569761543,
      "grad_norm": 6.40625,
      "learning_rate": 8.783450998878951e-06,
      "loss": 1.5791708,
      "memory(GiB)": 97.17,
      "step": 20930,
      "train_speed(iter/s)": 1.636159
    },
    {
      "acc": 0.65172501,
      "epoch": 0.5310755961440893,
      "grad_norm": 5.0,
      "learning_rate": 8.78276535318227e-06,
      "loss": 1.61124516,
      "memory(GiB)": 97.17,
      "step": 20935,
      "train_speed(iter/s)": 1.636199
    },
    {
      "acc": 0.63314943,
      "epoch": 0.5312024353120244,
      "grad_norm": 6.5625,
      "learning_rate": 8.782079541102678e-06,
      "loss": 1.68262787,
      "memory(GiB)": 97.17,
      "step": 20940,
      "train_speed(iter/s)": 1.63624
    },
    {
      "acc": 0.64838591,
      "epoch": 0.5313292744799594,
      "grad_norm": 4.96875,
      "learning_rate": 8.781393562670342e-06,
      "loss": 1.6325592,
      "memory(GiB)": 97.17,
      "step": 20945,
      "train_speed(iter/s)": 1.636279
    },
    {
      "acc": 0.63982205,
      "epoch": 0.5314561136478945,
      "grad_norm": 5.25,
      "learning_rate": 8.780707417915436e-06,
      "loss": 1.66564236,
      "memory(GiB)": 97.17,
      "step": 20950,
      "train_speed(iter/s)": 1.63632
    },
    {
      "acc": 0.66419077,
      "epoch": 0.5315829528158296,
      "grad_norm": 5.1875,
      "learning_rate": 8.780021106868138e-06,
      "loss": 1.50687103,
      "memory(GiB)": 97.17,
      "step": 20955,
      "train_speed(iter/s)": 1.636361
    },
    {
      "acc": 0.64624062,
      "epoch": 0.5317097919837646,
      "grad_norm": 6.71875,
      "learning_rate": 8.779334629558633e-06,
      "loss": 1.67514515,
      "memory(GiB)": 97.17,
      "step": 20960,
      "train_speed(iter/s)": 1.636399
    },
    {
      "acc": 0.65233459,
      "epoch": 0.5318366311516997,
      "grad_norm": 7.65625,
      "learning_rate": 8.77864798601712e-06,
      "loss": 1.63725204,
      "memory(GiB)": 97.17,
      "step": 20965,
      "train_speed(iter/s)": 1.63644
    },
    {
      "acc": 0.65923095,
      "epoch": 0.5319634703196348,
      "grad_norm": 5.1875,
      "learning_rate": 8.777961176273795e-06,
      "loss": 1.62402458,
      "memory(GiB)": 97.17,
      "step": 20970,
      "train_speed(iter/s)": 1.636479
    },
    {
      "acc": 0.65343065,
      "epoch": 0.5320903094875697,
      "grad_norm": 7.34375,
      "learning_rate": 8.777274200358873e-06,
      "loss": 1.66068497,
      "memory(GiB)": 97.17,
      "step": 20975,
      "train_speed(iter/s)": 1.636522
    },
    {
      "acc": 0.66354771,
      "epoch": 0.5322171486555048,
      "grad_norm": 5.40625,
      "learning_rate": 8.776587058302566e-06,
      "loss": 1.61062431,
      "memory(GiB)": 97.17,
      "step": 20980,
      "train_speed(iter/s)": 1.636564
    },
    {
      "acc": 0.65908747,
      "epoch": 0.5323439878234398,
      "grad_norm": 5.0,
      "learning_rate": 8.7758997501351e-06,
      "loss": 1.57970219,
      "memory(GiB)": 97.17,
      "step": 20985,
      "train_speed(iter/s)": 1.636606
    },
    {
      "acc": 0.66279473,
      "epoch": 0.5324708269913749,
      "grad_norm": 6.0,
      "learning_rate": 8.775212275886705e-06,
      "loss": 1.60897942,
      "memory(GiB)": 97.17,
      "step": 20990,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.63885632,
      "epoch": 0.53259766615931,
      "grad_norm": 6.5,
      "learning_rate": 8.774524635587617e-06,
      "loss": 1.66984577,
      "memory(GiB)": 97.17,
      "step": 20995,
      "train_speed(iter/s)": 1.636691
    },
    {
      "acc": 0.65384226,
      "epoch": 0.532724505327245,
      "grad_norm": 5.21875,
      "learning_rate": 8.773836829268084e-06,
      "loss": 1.63063908,
      "memory(GiB)": 97.17,
      "step": 21000,
      "train_speed(iter/s)": 1.636734
    },
    {
      "epoch": 0.532724505327245,
      "eval_acc": 0.6440223281689523,
      "eval_loss": 1.5857573747634888,
      "eval_runtime": 58.3644,
      "eval_samples_per_second": 109.142,
      "eval_steps_per_second": 27.294,
      "step": 21000
    },
    {
      "acc": 0.64876814,
      "epoch": 0.5328513444951801,
      "grad_norm": 5.46875,
      "learning_rate": 8.77314885695836e-06,
      "loss": 1.59493008,
      "memory(GiB)": 97.17,
      "step": 21005,
      "train_speed(iter/s)": 1.628842
    },
    {
      "acc": 0.65038333,
      "epoch": 0.5329781836631152,
      "grad_norm": 5.40625,
      "learning_rate": 8.772460718688702e-06,
      "loss": 1.59678936,
      "memory(GiB)": 97.17,
      "step": 21010,
      "train_speed(iter/s)": 1.628888
    },
    {
      "acc": 0.66392612,
      "epoch": 0.5331050228310502,
      "grad_norm": 6.1875,
      "learning_rate": 8.771772414489379e-06,
      "loss": 1.57642765,
      "memory(GiB)": 97.17,
      "step": 21015,
      "train_speed(iter/s)": 1.628936
    },
    {
      "acc": 0.64752407,
      "epoch": 0.5332318619989853,
      "grad_norm": 6.40625,
      "learning_rate": 8.771083944390665e-06,
      "loss": 1.64965706,
      "memory(GiB)": 97.17,
      "step": 21020,
      "train_speed(iter/s)": 1.628983
    },
    {
      "acc": 0.64618082,
      "epoch": 0.5333587011669203,
      "grad_norm": 4.84375,
      "learning_rate": 8.770395308422842e-06,
      "loss": 1.67151146,
      "memory(GiB)": 97.17,
      "step": 21025,
      "train_speed(iter/s)": 1.629026
    },
    {
      "acc": 0.65674653,
      "epoch": 0.5334855403348554,
      "grad_norm": 6.0625,
      "learning_rate": 8.769706506616201e-06,
      "loss": 1.6231432,
      "memory(GiB)": 97.17,
      "step": 21030,
      "train_speed(iter/s)": 1.629074
    },
    {
      "acc": 0.66036348,
      "epoch": 0.5336123795027905,
      "grad_norm": 4.3125,
      "learning_rate": 8.769017539001037e-06,
      "loss": 1.58847218,
      "memory(GiB)": 97.17,
      "step": 21035,
      "train_speed(iter/s)": 1.62912
    },
    {
      "acc": 0.65948958,
      "epoch": 0.5337392186707255,
      "grad_norm": 5.875,
      "learning_rate": 8.768328405607655e-06,
      "loss": 1.63342094,
      "memory(GiB)": 97.17,
      "step": 21040,
      "train_speed(iter/s)": 1.629166
    },
    {
      "acc": 0.64525099,
      "epoch": 0.5338660578386606,
      "grad_norm": 5.25,
      "learning_rate": 8.767639106466364e-06,
      "loss": 1.69602432,
      "memory(GiB)": 97.17,
      "step": 21045,
      "train_speed(iter/s)": 1.629213
    },
    {
      "acc": 0.65140467,
      "epoch": 0.5339928970065957,
      "grad_norm": 4.96875,
      "learning_rate": 8.766949641607484e-06,
      "loss": 1.62470818,
      "memory(GiB)": 97.17,
      "step": 21050,
      "train_speed(iter/s)": 1.629259
    },
    {
      "acc": 0.65899959,
      "epoch": 0.5341197361745307,
      "grad_norm": 6.125,
      "learning_rate": 8.76626001106134e-06,
      "loss": 1.61285095,
      "memory(GiB)": 97.17,
      "step": 21055,
      "train_speed(iter/s)": 1.629304
    },
    {
      "acc": 0.66049242,
      "epoch": 0.5342465753424658,
      "grad_norm": 5.46875,
      "learning_rate": 8.765570214858268e-06,
      "loss": 1.543892,
      "memory(GiB)": 97.17,
      "step": 21060,
      "train_speed(iter/s)": 1.62934
    },
    {
      "acc": 0.65370889,
      "epoch": 0.5343734145104008,
      "grad_norm": 6.0,
      "learning_rate": 8.764880253028604e-06,
      "loss": 1.57371101,
      "memory(GiB)": 97.17,
      "step": 21065,
      "train_speed(iter/s)": 1.629386
    },
    {
      "acc": 0.64696016,
      "epoch": 0.5345002536783359,
      "grad_norm": 4.4375,
      "learning_rate": 8.764190125602698e-06,
      "loss": 1.59636192,
      "memory(GiB)": 97.17,
      "step": 21070,
      "train_speed(iter/s)": 1.62943
    },
    {
      "acc": 0.65373049,
      "epoch": 0.534627092846271,
      "grad_norm": 5.875,
      "learning_rate": 8.763499832610904e-06,
      "loss": 1.64795647,
      "memory(GiB)": 97.17,
      "step": 21075,
      "train_speed(iter/s)": 1.629478
    },
    {
      "acc": 0.66816235,
      "epoch": 0.534753932014206,
      "grad_norm": 4.90625,
      "learning_rate": 8.762809374083585e-06,
      "loss": 1.54908638,
      "memory(GiB)": 97.17,
      "step": 21080,
      "train_speed(iter/s)": 1.629522
    },
    {
      "acc": 0.64169078,
      "epoch": 0.5348807711821411,
      "grad_norm": 7.5,
      "learning_rate": 8.76211875005111e-06,
      "loss": 1.65061417,
      "memory(GiB)": 97.17,
      "step": 21085,
      "train_speed(iter/s)": 1.62957
    },
    {
      "acc": 0.67778893,
      "epoch": 0.5350076103500762,
      "grad_norm": 7.15625,
      "learning_rate": 8.761427960543854e-06,
      "loss": 1.52431011,
      "memory(GiB)": 97.17,
      "step": 21090,
      "train_speed(iter/s)": 1.629617
    },
    {
      "acc": 0.65496731,
      "epoch": 0.5351344495180111,
      "grad_norm": 5.625,
      "learning_rate": 8.760737005592205e-06,
      "loss": 1.62793102,
      "memory(GiB)": 97.17,
      "step": 21095,
      "train_speed(iter/s)": 1.629663
    },
    {
      "acc": 0.68127284,
      "epoch": 0.5352612886859462,
      "grad_norm": 4.6875,
      "learning_rate": 8.760045885226551e-06,
      "loss": 1.5818656,
      "memory(GiB)": 97.17,
      "step": 21100,
      "train_speed(iter/s)": 1.629075
    },
    {
      "acc": 0.66091471,
      "epoch": 0.5353881278538812,
      "grad_norm": 5.875,
      "learning_rate": 8.759354599477293e-06,
      "loss": 1.62469769,
      "memory(GiB)": 97.17,
      "step": 21105,
      "train_speed(iter/s)": 1.629122
    },
    {
      "acc": 0.6668292,
      "epoch": 0.5355149670218163,
      "grad_norm": 5.75,
      "learning_rate": 8.758663148374833e-06,
      "loss": 1.53694038,
      "memory(GiB)": 97.17,
      "step": 21110,
      "train_speed(iter/s)": 1.629169
    },
    {
      "acc": 0.66185007,
      "epoch": 0.5356418061897514,
      "grad_norm": 5.40625,
      "learning_rate": 8.757971531949587e-06,
      "loss": 1.63506279,
      "memory(GiB)": 97.17,
      "step": 21115,
      "train_speed(iter/s)": 1.629214
    },
    {
      "acc": 0.65972338,
      "epoch": 0.5357686453576864,
      "grad_norm": 5.0,
      "learning_rate": 8.757279750231977e-06,
      "loss": 1.58701916,
      "memory(GiB)": 97.17,
      "step": 21120,
      "train_speed(iter/s)": 1.629261
    },
    {
      "acc": 0.66191249,
      "epoch": 0.5358954845256215,
      "grad_norm": 5.25,
      "learning_rate": 8.756587803252426e-06,
      "loss": 1.55750599,
      "memory(GiB)": 97.17,
      "step": 21125,
      "train_speed(iter/s)": 1.629307
    },
    {
      "acc": 0.65840745,
      "epoch": 0.5360223236935566,
      "grad_norm": 5.53125,
      "learning_rate": 8.755895691041373e-06,
      "loss": 1.61100273,
      "memory(GiB)": 97.17,
      "step": 21130,
      "train_speed(iter/s)": 1.629353
    },
    {
      "acc": 0.65723963,
      "epoch": 0.5361491628614916,
      "grad_norm": 4.96875,
      "learning_rate": 8.755203413629257e-06,
      "loss": 1.61247616,
      "memory(GiB)": 97.17,
      "step": 21135,
      "train_speed(iter/s)": 1.629398
    },
    {
      "acc": 0.66113396,
      "epoch": 0.5362760020294267,
      "grad_norm": 5.34375,
      "learning_rate": 8.75451097104653e-06,
      "loss": 1.5241684,
      "memory(GiB)": 97.17,
      "step": 21140,
      "train_speed(iter/s)": 1.629445
    },
    {
      "acc": 0.65825863,
      "epoch": 0.5364028411973617,
      "grad_norm": 5.40625,
      "learning_rate": 8.75381836332365e-06,
      "loss": 1.65647945,
      "memory(GiB)": 97.17,
      "step": 21145,
      "train_speed(iter/s)": 1.62949
    },
    {
      "acc": 0.63270392,
      "epoch": 0.5365296803652968,
      "grad_norm": 5.3125,
      "learning_rate": 8.753125590491077e-06,
      "loss": 1.64848175,
      "memory(GiB)": 97.17,
      "step": 21150,
      "train_speed(iter/s)": 1.629537
    },
    {
      "acc": 0.66235132,
      "epoch": 0.5366565195332319,
      "grad_norm": 4.78125,
      "learning_rate": 8.752432652579284e-06,
      "loss": 1.56671915,
      "memory(GiB)": 97.17,
      "step": 21155,
      "train_speed(iter/s)": 1.629583
    },
    {
      "acc": 0.64957347,
      "epoch": 0.5367833587011669,
      "grad_norm": 6.53125,
      "learning_rate": 8.751739549618749e-06,
      "loss": 1.64176674,
      "memory(GiB)": 97.17,
      "step": 21160,
      "train_speed(iter/s)": 1.629629
    },
    {
      "acc": 0.65781221,
      "epoch": 0.536910197869102,
      "grad_norm": 6.5,
      "learning_rate": 8.751046281639958e-06,
      "loss": 1.57351341,
      "memory(GiB)": 97.17,
      "step": 21165,
      "train_speed(iter/s)": 1.629675
    },
    {
      "acc": 0.65551076,
      "epoch": 0.5370370370370371,
      "grad_norm": 4.875,
      "learning_rate": 8.750352848673405e-06,
      "loss": 1.64728775,
      "memory(GiB)": 97.17,
      "step": 21170,
      "train_speed(iter/s)": 1.629722
    },
    {
      "acc": 0.64609766,
      "epoch": 0.5371638762049721,
      "grad_norm": 5.6875,
      "learning_rate": 8.749659250749589e-06,
      "loss": 1.6993351,
      "memory(GiB)": 97.17,
      "step": 21175,
      "train_speed(iter/s)": 1.629768
    },
    {
      "acc": 0.66735706,
      "epoch": 0.5372907153729072,
      "grad_norm": 6.78125,
      "learning_rate": 8.748965487899019e-06,
      "loss": 1.57400665,
      "memory(GiB)": 97.17,
      "step": 21180,
      "train_speed(iter/s)": 1.629815
    },
    {
      "acc": 0.65819178,
      "epoch": 0.5374175545408422,
      "grad_norm": 6.25,
      "learning_rate": 8.748271560152208e-06,
      "loss": 1.59758873,
      "memory(GiB)": 97.17,
      "step": 21185,
      "train_speed(iter/s)": 1.629863
    },
    {
      "acc": 0.6487606,
      "epoch": 0.5375443937087773,
      "grad_norm": 5.53125,
      "learning_rate": 8.74757746753968e-06,
      "loss": 1.64695549,
      "memory(GiB)": 97.17,
      "step": 21190,
      "train_speed(iter/s)": 1.629912
    },
    {
      "acc": 0.65843048,
      "epoch": 0.5376712328767124,
      "grad_norm": 6.375,
      "learning_rate": 8.746883210091963e-06,
      "loss": 1.52678165,
      "memory(GiB)": 97.17,
      "step": 21195,
      "train_speed(iter/s)": 1.62996
    },
    {
      "acc": 0.67876468,
      "epoch": 0.5377980720446474,
      "grad_norm": 4.875,
      "learning_rate": 8.746188787839593e-06,
      "loss": 1.49834757,
      "memory(GiB)": 97.17,
      "step": 21200,
      "train_speed(iter/s)": 1.630008
    },
    {
      "acc": 0.65346189,
      "epoch": 0.5379249112125825,
      "grad_norm": 5.125,
      "learning_rate": 8.745494200813116e-06,
      "loss": 1.62275333,
      "memory(GiB)": 97.17,
      "step": 21205,
      "train_speed(iter/s)": 1.630054
    },
    {
      "acc": 0.64756126,
      "epoch": 0.5380517503805176,
      "grad_norm": 7.375,
      "learning_rate": 8.74479944904308e-06,
      "loss": 1.62635345,
      "memory(GiB)": 97.17,
      "step": 21210,
      "train_speed(iter/s)": 1.6301
    },
    {
      "acc": 0.64901819,
      "epoch": 0.5381785895484525,
      "grad_norm": 5.59375,
      "learning_rate": 8.744104532560047e-06,
      "loss": 1.62624359,
      "memory(GiB)": 97.17,
      "step": 21215,
      "train_speed(iter/s)": 1.630145
    },
    {
      "acc": 0.64173245,
      "epoch": 0.5383054287163876,
      "grad_norm": 5.53125,
      "learning_rate": 8.74340945139458e-06,
      "loss": 1.72811394,
      "memory(GiB)": 97.17,
      "step": 21220,
      "train_speed(iter/s)": 1.630192
    },
    {
      "acc": 0.65192585,
      "epoch": 0.5384322678843226,
      "grad_norm": 8.375,
      "learning_rate": 8.742714205577251e-06,
      "loss": 1.63894882,
      "memory(GiB)": 97.17,
      "step": 21225,
      "train_speed(iter/s)": 1.630238
    },
    {
      "acc": 0.66877565,
      "epoch": 0.5385591070522577,
      "grad_norm": 4.96875,
      "learning_rate": 8.742018795138642e-06,
      "loss": 1.49626751,
      "memory(GiB)": 97.17,
      "step": 21230,
      "train_speed(iter/s)": 1.630284
    },
    {
      "acc": 0.65010915,
      "epoch": 0.5386859462201928,
      "grad_norm": 5.65625,
      "learning_rate": 8.74132322010934e-06,
      "loss": 1.6074913,
      "memory(GiB)": 97.17,
      "step": 21235,
      "train_speed(iter/s)": 1.630327
    },
    {
      "acc": 0.64931459,
      "epoch": 0.5388127853881278,
      "grad_norm": 5.1875,
      "learning_rate": 8.740627480519937e-06,
      "loss": 1.58404293,
      "memory(GiB)": 97.17,
      "step": 21240,
      "train_speed(iter/s)": 1.630372
    },
    {
      "acc": 0.64686995,
      "epoch": 0.5389396245560629,
      "grad_norm": 6.375,
      "learning_rate": 8.739931576401037e-06,
      "loss": 1.62725353,
      "memory(GiB)": 97.17,
      "step": 21245,
      "train_speed(iter/s)": 1.630417
    },
    {
      "acc": 0.65141277,
      "epoch": 0.539066463723998,
      "grad_norm": 5.3125,
      "learning_rate": 8.73923550778325e-06,
      "loss": 1.54246845,
      "memory(GiB)": 97.17,
      "step": 21250,
      "train_speed(iter/s)": 1.630464
    },
    {
      "acc": 0.66038904,
      "epoch": 0.539193302891933,
      "grad_norm": 5.84375,
      "learning_rate": 8.73853927469719e-06,
      "loss": 1.63263245,
      "memory(GiB)": 97.17,
      "step": 21255,
      "train_speed(iter/s)": 1.630507
    },
    {
      "acc": 0.65081282,
      "epoch": 0.5393201420598681,
      "grad_norm": 6.625,
      "learning_rate": 8.73784287717348e-06,
      "loss": 1.6154726,
      "memory(GiB)": 97.17,
      "step": 21260,
      "train_speed(iter/s)": 1.630554
    },
    {
      "acc": 0.62822676,
      "epoch": 0.5394469812278031,
      "grad_norm": 5.3125,
      "learning_rate": 8.737146315242755e-06,
      "loss": 1.71875343,
      "memory(GiB)": 97.17,
      "step": 21265,
      "train_speed(iter/s)": 1.630599
    },
    {
      "acc": 0.64272547,
      "epoch": 0.5395738203957382,
      "grad_norm": 5.4375,
      "learning_rate": 8.73644958893565e-06,
      "loss": 1.67719517,
      "memory(GiB)": 97.17,
      "step": 21270,
      "train_speed(iter/s)": 1.630648
    },
    {
      "acc": 0.66212773,
      "epoch": 0.5397006595636733,
      "grad_norm": 4.90625,
      "learning_rate": 8.735752698282807e-06,
      "loss": 1.57490864,
      "memory(GiB)": 97.17,
      "step": 21275,
      "train_speed(iter/s)": 1.630693
    },
    {
      "acc": 0.67078214,
      "epoch": 0.5398274987316083,
      "grad_norm": 5.25,
      "learning_rate": 8.735055643314883e-06,
      "loss": 1.66050434,
      "memory(GiB)": 97.17,
      "step": 21280,
      "train_speed(iter/s)": 1.63074
    },
    {
      "acc": 0.64633322,
      "epoch": 0.5399543378995434,
      "grad_norm": 5.3125,
      "learning_rate": 8.734358424062536e-06,
      "loss": 1.66605797,
      "memory(GiB)": 97.17,
      "step": 21285,
      "train_speed(iter/s)": 1.630782
    },
    {
      "acc": 0.66149521,
      "epoch": 0.5400811770674785,
      "grad_norm": 4.9375,
      "learning_rate": 8.733661040556433e-06,
      "loss": 1.60061684,
      "memory(GiB)": 97.17,
      "step": 21290,
      "train_speed(iter/s)": 1.630828
    },
    {
      "acc": 0.638094,
      "epoch": 0.5402080162354135,
      "grad_norm": 5.53125,
      "learning_rate": 8.732963492827248e-06,
      "loss": 1.64067459,
      "memory(GiB)": 97.17,
      "step": 21295,
      "train_speed(iter/s)": 1.630873
    },
    {
      "acc": 0.64380012,
      "epoch": 0.5403348554033486,
      "grad_norm": 5.09375,
      "learning_rate": 8.732265780905661e-06,
      "loss": 1.64288254,
      "memory(GiB)": 97.17,
      "step": 21300,
      "train_speed(iter/s)": 1.630917
    },
    {
      "acc": 0.64279027,
      "epoch": 0.5404616945712836,
      "grad_norm": 5.1875,
      "learning_rate": 8.731567904822362e-06,
      "loss": 1.6680933,
      "memory(GiB)": 97.17,
      "step": 21305,
      "train_speed(iter/s)": 1.630962
    },
    {
      "acc": 0.66126328,
      "epoch": 0.5405885337392187,
      "grad_norm": 5.28125,
      "learning_rate": 8.730869864608047e-06,
      "loss": 1.48987989,
      "memory(GiB)": 97.17,
      "step": 21310,
      "train_speed(iter/s)": 1.631008
    },
    {
      "acc": 0.64659724,
      "epoch": 0.5407153729071538,
      "grad_norm": 6.0625,
      "learning_rate": 8.730171660293418e-06,
      "loss": 1.70109501,
      "memory(GiB)": 97.17,
      "step": 21315,
      "train_speed(iter/s)": 1.631055
    },
    {
      "acc": 0.65225129,
      "epoch": 0.5408422120750888,
      "grad_norm": 4.625,
      "learning_rate": 8.729473291909185e-06,
      "loss": 1.60611572,
      "memory(GiB)": 97.17,
      "step": 21320,
      "train_speed(iter/s)": 1.631102
    },
    {
      "acc": 0.64276223,
      "epoch": 0.5409690512430239,
      "grad_norm": 5.90625,
      "learning_rate": 8.728774759486065e-06,
      "loss": 1.63817921,
      "memory(GiB)": 97.17,
      "step": 21325,
      "train_speed(iter/s)": 1.63115
    },
    {
      "acc": 0.64491348,
      "epoch": 0.541095890410959,
      "grad_norm": 5.09375,
      "learning_rate": 8.728076063054786e-06,
      "loss": 1.60837955,
      "memory(GiB)": 97.17,
      "step": 21330,
      "train_speed(iter/s)": 1.631196
    },
    {
      "acc": 0.65019789,
      "epoch": 0.541222729578894,
      "grad_norm": 6.03125,
      "learning_rate": 8.727377202646074e-06,
      "loss": 1.71281738,
      "memory(GiB)": 97.17,
      "step": 21335,
      "train_speed(iter/s)": 1.631244
    },
    {
      "acc": 0.66167102,
      "epoch": 0.541349568746829,
      "grad_norm": 7.15625,
      "learning_rate": 8.726678178290673e-06,
      "loss": 1.65932713,
      "memory(GiB)": 97.17,
      "step": 21340,
      "train_speed(iter/s)": 1.631289
    },
    {
      "acc": 0.6633606,
      "epoch": 0.541476407914764,
      "grad_norm": 6.3125,
      "learning_rate": 8.725978990019326e-06,
      "loss": 1.60574989,
      "memory(GiB)": 97.17,
      "step": 21345,
      "train_speed(iter/s)": 1.631338
    },
    {
      "acc": 0.6633574,
      "epoch": 0.5416032470826991,
      "grad_norm": 4.90625,
      "learning_rate": 8.72527963786279e-06,
      "loss": 1.58699722,
      "memory(GiB)": 97.17,
      "step": 21350,
      "train_speed(iter/s)": 1.631381
    },
    {
      "acc": 0.65025959,
      "epoch": 0.5417300862506342,
      "grad_norm": 4.90625,
      "learning_rate": 8.72458012185182e-06,
      "loss": 1.69510059,
      "memory(GiB)": 97.17,
      "step": 21355,
      "train_speed(iter/s)": 1.631423
    },
    {
      "acc": 0.65867424,
      "epoch": 0.5418569254185692,
      "grad_norm": 5.15625,
      "learning_rate": 8.72388044201719e-06,
      "loss": 1.57125511,
      "memory(GiB)": 97.17,
      "step": 21360,
      "train_speed(iter/s)": 1.631468
    },
    {
      "acc": 0.67092819,
      "epoch": 0.5419837645865043,
      "grad_norm": 5.90625,
      "learning_rate": 8.723180598389671e-06,
      "loss": 1.50119743,
      "memory(GiB)": 97.17,
      "step": 21365,
      "train_speed(iter/s)": 1.631512
    },
    {
      "acc": 0.64988117,
      "epoch": 0.5421106037544394,
      "grad_norm": 4.375,
      "learning_rate": 8.722480591000046e-06,
      "loss": 1.63654594,
      "memory(GiB)": 97.17,
      "step": 21370,
      "train_speed(iter/s)": 1.631556
    },
    {
      "acc": 0.67209129,
      "epoch": 0.5422374429223744,
      "grad_norm": 8.5,
      "learning_rate": 8.721780419879106e-06,
      "loss": 1.52114229,
      "memory(GiB)": 97.17,
      "step": 21375,
      "train_speed(iter/s)": 1.631603
    },
    {
      "acc": 0.65188808,
      "epoch": 0.5423642820903095,
      "grad_norm": 4.875,
      "learning_rate": 8.721080085057646e-06,
      "loss": 1.64187241,
      "memory(GiB)": 97.17,
      "step": 21380,
      "train_speed(iter/s)": 1.63165
    },
    {
      "acc": 0.6517755,
      "epoch": 0.5424911212582445,
      "grad_norm": 7.96875,
      "learning_rate": 8.72037958656647e-06,
      "loss": 1.70594139,
      "memory(GiB)": 97.17,
      "step": 21385,
      "train_speed(iter/s)": 1.631697
    },
    {
      "acc": 0.6375782,
      "epoch": 0.5426179604261796,
      "grad_norm": 5.125,
      "learning_rate": 8.71967892443639e-06,
      "loss": 1.64758358,
      "memory(GiB)": 97.17,
      "step": 21390,
      "train_speed(iter/s)": 1.631742
    },
    {
      "acc": 0.66278372,
      "epoch": 0.5427447995941147,
      "grad_norm": 6.34375,
      "learning_rate": 8.718978098698226e-06,
      "loss": 1.59483213,
      "memory(GiB)": 97.17,
      "step": 21395,
      "train_speed(iter/s)": 1.631787
    },
    {
      "acc": 0.65543542,
      "epoch": 0.5428716387620497,
      "grad_norm": 6.59375,
      "learning_rate": 8.718277109382799e-06,
      "loss": 1.64448128,
      "memory(GiB)": 97.17,
      "step": 21400,
      "train_speed(iter/s)": 1.631834
    },
    {
      "acc": 0.65426188,
      "epoch": 0.5429984779299848,
      "grad_norm": 6.75,
      "learning_rate": 8.717575956520942e-06,
      "loss": 1.66474266,
      "memory(GiB)": 97.17,
      "step": 21405,
      "train_speed(iter/s)": 1.631882
    },
    {
      "acc": 0.6654305,
      "epoch": 0.5431253170979199,
      "grad_norm": 4.78125,
      "learning_rate": 8.716874640143498e-06,
      "loss": 1.55772953,
      "memory(GiB)": 97.17,
      "step": 21410,
      "train_speed(iter/s)": 1.631927
    },
    {
      "acc": 0.65467491,
      "epoch": 0.5432521562658549,
      "grad_norm": 7.875,
      "learning_rate": 8.716173160281315e-06,
      "loss": 1.66262207,
      "memory(GiB)": 97.17,
      "step": 21415,
      "train_speed(iter/s)": 1.631971
    },
    {
      "acc": 0.63358483,
      "epoch": 0.54337899543379,
      "grad_norm": 7.21875,
      "learning_rate": 8.715471516965242e-06,
      "loss": 1.69256821,
      "memory(GiB)": 97.17,
      "step": 21420,
      "train_speed(iter/s)": 1.632016
    },
    {
      "acc": 0.64574442,
      "epoch": 0.543505834601725,
      "grad_norm": 6.15625,
      "learning_rate": 8.714769710226144e-06,
      "loss": 1.65064392,
      "memory(GiB)": 97.17,
      "step": 21425,
      "train_speed(iter/s)": 1.632061
    },
    {
      "acc": 0.6545187,
      "epoch": 0.5436326737696601,
      "grad_norm": 5.75,
      "learning_rate": 8.714067740094888e-06,
      "loss": 1.63890152,
      "memory(GiB)": 97.17,
      "step": 21430,
      "train_speed(iter/s)": 1.632107
    },
    {
      "acc": 0.67132926,
      "epoch": 0.5437595129375952,
      "grad_norm": 8.4375,
      "learning_rate": 8.713365606602353e-06,
      "loss": 1.58325291,
      "memory(GiB)": 97.17,
      "step": 21435,
      "train_speed(iter/s)": 1.632154
    },
    {
      "acc": 0.67222395,
      "epoch": 0.5438863521055302,
      "grad_norm": 5.9375,
      "learning_rate": 8.71266330977942e-06,
      "loss": 1.50973949,
      "memory(GiB)": 97.17,
      "step": 21440,
      "train_speed(iter/s)": 1.632196
    },
    {
      "acc": 0.67877855,
      "epoch": 0.5440131912734653,
      "grad_norm": 5.28125,
      "learning_rate": 8.711960849656975e-06,
      "loss": 1.52223024,
      "memory(GiB)": 97.17,
      "step": 21445,
      "train_speed(iter/s)": 1.63224
    },
    {
      "acc": 0.64687319,
      "epoch": 0.5441400304414004,
      "grad_norm": 6.75,
      "learning_rate": 8.711258226265922e-06,
      "loss": 1.62021103,
      "memory(GiB)": 97.17,
      "step": 21450,
      "train_speed(iter/s)": 1.632283
    },
    {
      "acc": 0.66055937,
      "epoch": 0.5442668696093353,
      "grad_norm": 5.875,
      "learning_rate": 8.710555439637163e-06,
      "loss": 1.65854111,
      "memory(GiB)": 97.17,
      "step": 21455,
      "train_speed(iter/s)": 1.63233
    },
    {
      "acc": 0.64457908,
      "epoch": 0.5443937087772704,
      "grad_norm": 6.65625,
      "learning_rate": 8.709852489801608e-06,
      "loss": 1.58159885,
      "memory(GiB)": 97.17,
      "step": 21460,
      "train_speed(iter/s)": 1.632375
    },
    {
      "acc": 0.6482645,
      "epoch": 0.5445205479452054,
      "grad_norm": 6.28125,
      "learning_rate": 8.709149376790177e-06,
      "loss": 1.64121399,
      "memory(GiB)": 97.17,
      "step": 21465,
      "train_speed(iter/s)": 1.632422
    },
    {
      "acc": 0.65086312,
      "epoch": 0.5446473871131405,
      "grad_norm": 5.71875,
      "learning_rate": 8.708446100633796e-06,
      "loss": 1.64037914,
      "memory(GiB)": 97.17,
      "step": 21470,
      "train_speed(iter/s)": 1.632467
    },
    {
      "acc": 0.68844948,
      "epoch": 0.5447742262810756,
      "grad_norm": 6.46875,
      "learning_rate": 8.707742661363401e-06,
      "loss": 1.47648354,
      "memory(GiB)": 97.17,
      "step": 21475,
      "train_speed(iter/s)": 1.632515
    },
    {
      "acc": 0.63630157,
      "epoch": 0.5449010654490106,
      "grad_norm": 6.40625,
      "learning_rate": 8.707039059009927e-06,
      "loss": 1.72131348,
      "memory(GiB)": 97.17,
      "step": 21480,
      "train_speed(iter/s)": 1.632558
    },
    {
      "acc": 0.66428332,
      "epoch": 0.5450279046169457,
      "grad_norm": 6.46875,
      "learning_rate": 8.706335293604326e-06,
      "loss": 1.60624733,
      "memory(GiB)": 97.17,
      "step": 21485,
      "train_speed(iter/s)": 1.632602
    },
    {
      "acc": 0.65580721,
      "epoch": 0.5451547437848808,
      "grad_norm": 5.34375,
      "learning_rate": 8.705631365177552e-06,
      "loss": 1.65646057,
      "memory(GiB)": 97.17,
      "step": 21490,
      "train_speed(iter/s)": 1.632649
    },
    {
      "acc": 0.65630984,
      "epoch": 0.5452815829528158,
      "grad_norm": 6.3125,
      "learning_rate": 8.704927273760563e-06,
      "loss": 1.58290529,
      "memory(GiB)": 97.17,
      "step": 21495,
      "train_speed(iter/s)": 1.632695
    },
    {
      "acc": 0.65003924,
      "epoch": 0.5454084221207509,
      "grad_norm": 5.125,
      "learning_rate": 8.704223019384334e-06,
      "loss": 1.63767433,
      "memory(GiB)": 97.17,
      "step": 21500,
      "train_speed(iter/s)": 1.632739
    },
    {
      "acc": 0.66435966,
      "epoch": 0.5455352612886859,
      "grad_norm": 6.375,
      "learning_rate": 8.703518602079836e-06,
      "loss": 1.57400694,
      "memory(GiB)": 97.17,
      "step": 21505,
      "train_speed(iter/s)": 1.632784
    },
    {
      "acc": 0.64868693,
      "epoch": 0.545662100456621,
      "grad_norm": 5.5625,
      "learning_rate": 8.702814021878057e-06,
      "loss": 1.6368927,
      "memory(GiB)": 97.17,
      "step": 21510,
      "train_speed(iter/s)": 1.63283
    },
    {
      "acc": 0.64857607,
      "epoch": 0.5457889396245561,
      "grad_norm": 5.21875,
      "learning_rate": 8.702109278809985e-06,
      "loss": 1.66303978,
      "memory(GiB)": 97.17,
      "step": 21515,
      "train_speed(iter/s)": 1.632876
    },
    {
      "acc": 0.66084538,
      "epoch": 0.5459157787924911,
      "grad_norm": 4.75,
      "learning_rate": 8.70140437290662e-06,
      "loss": 1.61717606,
      "memory(GiB)": 97.17,
      "step": 21520,
      "train_speed(iter/s)": 1.632918
    },
    {
      "acc": 0.64758983,
      "epoch": 0.5460426179604262,
      "grad_norm": 5.4375,
      "learning_rate": 8.700699304198963e-06,
      "loss": 1.68869362,
      "memory(GiB)": 97.17,
      "step": 21525,
      "train_speed(iter/s)": 1.632963
    },
    {
      "acc": 0.65050988,
      "epoch": 0.5461694571283613,
      "grad_norm": 5.25,
      "learning_rate": 8.699994072718026e-06,
      "loss": 1.65058823,
      "memory(GiB)": 97.17,
      "step": 21530,
      "train_speed(iter/s)": 1.633006
    },
    {
      "acc": 0.66107302,
      "epoch": 0.5462962962962963,
      "grad_norm": 5.96875,
      "learning_rate": 8.699288678494833e-06,
      "loss": 1.55940018,
      "memory(GiB)": 97.17,
      "step": 21535,
      "train_speed(iter/s)": 1.633052
    },
    {
      "acc": 0.64216099,
      "epoch": 0.5464231354642314,
      "grad_norm": 5.34375,
      "learning_rate": 8.698583121560407e-06,
      "loss": 1.6853466,
      "memory(GiB)": 97.17,
      "step": 21540,
      "train_speed(iter/s)": 1.633098
    },
    {
      "acc": 0.65122237,
      "epoch": 0.5465499746321664,
      "grad_norm": 5.625,
      "learning_rate": 8.697877401945784e-06,
      "loss": 1.62169094,
      "memory(GiB)": 97.17,
      "step": 21545,
      "train_speed(iter/s)": 1.633143
    },
    {
      "acc": 0.66671586,
      "epoch": 0.5466768138001015,
      "grad_norm": 5.40625,
      "learning_rate": 8.697171519682002e-06,
      "loss": 1.56995211,
      "memory(GiB)": 97.17,
      "step": 21550,
      "train_speed(iter/s)": 1.633189
    },
    {
      "acc": 0.65399394,
      "epoch": 0.5468036529680366,
      "grad_norm": 4.875,
      "learning_rate": 8.696465474800109e-06,
      "loss": 1.60602703,
      "memory(GiB)": 97.17,
      "step": 21555,
      "train_speed(iter/s)": 1.633232
    },
    {
      "acc": 0.65313334,
      "epoch": 0.5469304921359716,
      "grad_norm": 5.4375,
      "learning_rate": 8.695759267331162e-06,
      "loss": 1.63794708,
      "memory(GiB)": 97.17,
      "step": 21560,
      "train_speed(iter/s)": 1.633277
    },
    {
      "acc": 0.65162725,
      "epoch": 0.5470573313039067,
      "grad_norm": 4.84375,
      "learning_rate": 8.69505289730622e-06,
      "loss": 1.62994118,
      "memory(GiB)": 97.17,
      "step": 21565,
      "train_speed(iter/s)": 1.633321
    },
    {
      "acc": 0.66639128,
      "epoch": 0.5471841704718418,
      "grad_norm": 6.96875,
      "learning_rate": 8.694346364756356e-06,
      "loss": 1.604953,
      "memory(GiB)": 97.17,
      "step": 21570,
      "train_speed(iter/s)": 1.633365
    },
    {
      "acc": 0.65384355,
      "epoch": 0.5473110096397767,
      "grad_norm": 5.03125,
      "learning_rate": 8.693639669712645e-06,
      "loss": 1.63351345,
      "memory(GiB)": 97.17,
      "step": 21575,
      "train_speed(iter/s)": 1.633407
    },
    {
      "acc": 0.65933371,
      "epoch": 0.5474378488077118,
      "grad_norm": 6.53125,
      "learning_rate": 8.692932812206171e-06,
      "loss": 1.56650009,
      "memory(GiB)": 97.17,
      "step": 21580,
      "train_speed(iter/s)": 1.633449
    },
    {
      "acc": 0.64212232,
      "epoch": 0.5475646879756468,
      "grad_norm": 5.09375,
      "learning_rate": 8.692225792268023e-06,
      "loss": 1.62280388,
      "memory(GiB)": 97.17,
      "step": 21585,
      "train_speed(iter/s)": 1.633494
    },
    {
      "acc": 0.65291519,
      "epoch": 0.5476915271435819,
      "grad_norm": 6.4375,
      "learning_rate": 8.691518609929302e-06,
      "loss": 1.64678917,
      "memory(GiB)": 97.17,
      "step": 21590,
      "train_speed(iter/s)": 1.633538
    },
    {
      "acc": 0.6526032,
      "epoch": 0.547818366311517,
      "grad_norm": 5.03125,
      "learning_rate": 8.690811265221108e-06,
      "loss": 1.59640846,
      "memory(GiB)": 97.17,
      "step": 21595,
      "train_speed(iter/s)": 1.633579
    },
    {
      "acc": 0.66505475,
      "epoch": 0.547945205479452,
      "grad_norm": 5.375,
      "learning_rate": 8.690103758174558e-06,
      "loss": 1.60830574,
      "memory(GiB)": 97.17,
      "step": 21600,
      "train_speed(iter/s)": 1.633624
    },
    {
      "acc": 0.62413301,
      "epoch": 0.5480720446473871,
      "grad_norm": 6.09375,
      "learning_rate": 8.68939608882077e-06,
      "loss": 1.66306019,
      "memory(GiB)": 97.17,
      "step": 21605,
      "train_speed(iter/s)": 1.633669
    },
    {
      "acc": 0.64723406,
      "epoch": 0.5481988838153222,
      "grad_norm": 6.15625,
      "learning_rate": 8.688688257190869e-06,
      "loss": 1.63699303,
      "memory(GiB)": 97.17,
      "step": 21610,
      "train_speed(iter/s)": 1.633714
    },
    {
      "acc": 0.67335544,
      "epoch": 0.5483257229832572,
      "grad_norm": 7.6875,
      "learning_rate": 8.68798026331599e-06,
      "loss": 1.51924582,
      "memory(GiB)": 97.17,
      "step": 21615,
      "train_speed(iter/s)": 1.633759
    },
    {
      "acc": 0.65182371,
      "epoch": 0.5484525621511923,
      "grad_norm": 5.1875,
      "learning_rate": 8.687272107227274e-06,
      "loss": 1.6607935,
      "memory(GiB)": 97.17,
      "step": 21620,
      "train_speed(iter/s)": 1.633802
    },
    {
      "acc": 0.63714275,
      "epoch": 0.5485794013191273,
      "grad_norm": 4.53125,
      "learning_rate": 8.686563788955867e-06,
      "loss": 1.66709213,
      "memory(GiB)": 97.17,
      "step": 21625,
      "train_speed(iter/s)": 1.633847
    },
    {
      "acc": 0.65819426,
      "epoch": 0.5487062404870624,
      "grad_norm": 5.53125,
      "learning_rate": 8.685855308532926e-06,
      "loss": 1.58025532,
      "memory(GiB)": 97.17,
      "step": 21630,
      "train_speed(iter/s)": 1.633892
    },
    {
      "acc": 0.65978975,
      "epoch": 0.5488330796549975,
      "grad_norm": 6.8125,
      "learning_rate": 8.685146665989613e-06,
      "loss": 1.60376511,
      "memory(GiB)": 97.17,
      "step": 21635,
      "train_speed(iter/s)": 1.633937
    },
    {
      "acc": 0.64683123,
      "epoch": 0.5489599188229325,
      "grad_norm": 5.6875,
      "learning_rate": 8.684437861357095e-06,
      "loss": 1.60137978,
      "memory(GiB)": 97.17,
      "step": 21640,
      "train_speed(iter/s)": 1.633983
    },
    {
      "acc": 0.65049801,
      "epoch": 0.5490867579908676,
      "grad_norm": 5.46875,
      "learning_rate": 8.683728894666551e-06,
      "loss": 1.56815739,
      "memory(GiB)": 97.17,
      "step": 21645,
      "train_speed(iter/s)": 1.634029
    },
    {
      "acc": 0.65585585,
      "epoch": 0.5492135971588027,
      "grad_norm": 5.46875,
      "learning_rate": 8.683019765949163e-06,
      "loss": 1.62692013,
      "memory(GiB)": 97.17,
      "step": 21650,
      "train_speed(iter/s)": 1.634074
    },
    {
      "acc": 0.64181685,
      "epoch": 0.5493404363267377,
      "grad_norm": 5.84375,
      "learning_rate": 8.682310475236123e-06,
      "loss": 1.61168747,
      "memory(GiB)": 97.17,
      "step": 21655,
      "train_speed(iter/s)": 1.634118
    },
    {
      "acc": 0.64873557,
      "epoch": 0.5494672754946728,
      "grad_norm": 6.15625,
      "learning_rate": 8.681601022558628e-06,
      "loss": 1.64048157,
      "memory(GiB)": 97.17,
      "step": 21660,
      "train_speed(iter/s)": 1.634163
    },
    {
      "acc": 0.65478134,
      "epoch": 0.5495941146626078,
      "grad_norm": 5.28125,
      "learning_rate": 8.680891407947882e-06,
      "loss": 1.59257851,
      "memory(GiB)": 97.17,
      "step": 21665,
      "train_speed(iter/s)": 1.634208
    },
    {
      "acc": 0.64225292,
      "epoch": 0.5497209538305429,
      "grad_norm": 5.375,
      "learning_rate": 8.680181631435098e-06,
      "loss": 1.72277775,
      "memory(GiB)": 97.17,
      "step": 21670,
      "train_speed(iter/s)": 1.634253
    },
    {
      "acc": 0.65724702,
      "epoch": 0.549847792998478,
      "grad_norm": 6.90625,
      "learning_rate": 8.679471693051495e-06,
      "loss": 1.61637383,
      "memory(GiB)": 97.17,
      "step": 21675,
      "train_speed(iter/s)": 1.634297
    },
    {
      "acc": 0.66019516,
      "epoch": 0.549974632166413,
      "grad_norm": 5.6875,
      "learning_rate": 8.678761592828301e-06,
      "loss": 1.59520149,
      "memory(GiB)": 97.17,
      "step": 21680,
      "train_speed(iter/s)": 1.63434
    },
    {
      "acc": 0.65677004,
      "epoch": 0.5501014713343481,
      "grad_norm": 5.5,
      "learning_rate": 8.678051330796746e-06,
      "loss": 1.56796055,
      "memory(GiB)": 97.17,
      "step": 21685,
      "train_speed(iter/s)": 1.634378
    },
    {
      "acc": 0.63619041,
      "epoch": 0.5502283105022832,
      "grad_norm": 6.1875,
      "learning_rate": 8.677340906988072e-06,
      "loss": 1.62516232,
      "memory(GiB)": 97.17,
      "step": 21690,
      "train_speed(iter/s)": 1.634422
    },
    {
      "acc": 0.65493727,
      "epoch": 0.5503551496702181,
      "grad_norm": 5.8125,
      "learning_rate": 8.676630321433528e-06,
      "loss": 1.58001223,
      "memory(GiB)": 97.17,
      "step": 21695,
      "train_speed(iter/s)": 1.634466
    },
    {
      "acc": 0.65300012,
      "epoch": 0.5504819888381532,
      "grad_norm": 5.3125,
      "learning_rate": 8.675919574164366e-06,
      "loss": 1.62814541,
      "memory(GiB)": 97.17,
      "step": 21700,
      "train_speed(iter/s)": 1.634509
    },
    {
      "acc": 0.65359364,
      "epoch": 0.5506088280060882,
      "grad_norm": 4.75,
      "learning_rate": 8.675208665211851e-06,
      "loss": 1.60846596,
      "memory(GiB)": 97.17,
      "step": 21705,
      "train_speed(iter/s)": 1.634552
    },
    {
      "acc": 0.64589133,
      "epoch": 0.5507356671740233,
      "grad_norm": 8.375,
      "learning_rate": 8.674497594607249e-06,
      "loss": 1.63216724,
      "memory(GiB)": 97.17,
      "step": 21710,
      "train_speed(iter/s)": 1.634595
    },
    {
      "acc": 0.64513469,
      "epoch": 0.5508625063419584,
      "grad_norm": 7.03125,
      "learning_rate": 8.673786362381837e-06,
      "loss": 1.67052593,
      "memory(GiB)": 97.17,
      "step": 21715,
      "train_speed(iter/s)": 1.63464
    },
    {
      "acc": 0.64713788,
      "epoch": 0.5509893455098934,
      "grad_norm": 4.6875,
      "learning_rate": 8.673074968566899e-06,
      "loss": 1.58162365,
      "memory(GiB)": 97.17,
      "step": 21720,
      "train_speed(iter/s)": 1.634684
    },
    {
      "acc": 0.63161049,
      "epoch": 0.5511161846778285,
      "grad_norm": 6.21875,
      "learning_rate": 8.672363413193724e-06,
      "loss": 1.70378265,
      "memory(GiB)": 97.17,
      "step": 21725,
      "train_speed(iter/s)": 1.634727
    },
    {
      "acc": 0.65538349,
      "epoch": 0.5512430238457636,
      "grad_norm": 4.625,
      "learning_rate": 8.671651696293613e-06,
      "loss": 1.59071178,
      "memory(GiB)": 97.17,
      "step": 21730,
      "train_speed(iter/s)": 1.634771
    },
    {
      "acc": 0.64878416,
      "epoch": 0.5513698630136986,
      "grad_norm": 5.9375,
      "learning_rate": 8.670939817897865e-06,
      "loss": 1.66455212,
      "memory(GiB)": 97.17,
      "step": 21735,
      "train_speed(iter/s)": 1.634816
    },
    {
      "acc": 0.64152088,
      "epoch": 0.5514967021816337,
      "grad_norm": 4.53125,
      "learning_rate": 8.670227778037796e-06,
      "loss": 1.64212399,
      "memory(GiB)": 97.17,
      "step": 21740,
      "train_speed(iter/s)": 1.634858
    },
    {
      "acc": 0.65355515,
      "epoch": 0.5516235413495687,
      "grad_norm": 6.40625,
      "learning_rate": 8.669515576744722e-06,
      "loss": 1.6184082,
      "memory(GiB)": 97.17,
      "step": 21745,
      "train_speed(iter/s)": 1.6349
    },
    {
      "acc": 0.6453692,
      "epoch": 0.5517503805175038,
      "grad_norm": 5.8125,
      "learning_rate": 8.66880321404997e-06,
      "loss": 1.64840527,
      "memory(GiB)": 97.17,
      "step": 21750,
      "train_speed(iter/s)": 1.634944
    },
    {
      "acc": 0.66755161,
      "epoch": 0.5518772196854389,
      "grad_norm": 6.75,
      "learning_rate": 8.668090689984872e-06,
      "loss": 1.5847641,
      "memory(GiB)": 97.17,
      "step": 21755,
      "train_speed(iter/s)": 1.634989
    },
    {
      "acc": 0.64707861,
      "epoch": 0.5520040588533739,
      "grad_norm": 5.75,
      "learning_rate": 8.667378004580769e-06,
      "loss": 1.59607029,
      "memory(GiB)": 97.17,
      "step": 21760,
      "train_speed(iter/s)": 1.635033
    },
    {
      "acc": 0.66107101,
      "epoch": 0.552130898021309,
      "grad_norm": 5.4375,
      "learning_rate": 8.666665157869007e-06,
      "loss": 1.58932552,
      "memory(GiB)": 97.17,
      "step": 21765,
      "train_speed(iter/s)": 1.635079
    },
    {
      "acc": 0.65089655,
      "epoch": 0.5522577371892441,
      "grad_norm": 5.4375,
      "learning_rate": 8.665952149880942e-06,
      "loss": 1.66242371,
      "memory(GiB)": 97.17,
      "step": 21770,
      "train_speed(iter/s)": 1.635122
    },
    {
      "acc": 0.666045,
      "epoch": 0.5523845763571791,
      "grad_norm": 5.90625,
      "learning_rate": 8.665238980647934e-06,
      "loss": 1.5191741,
      "memory(GiB)": 97.17,
      "step": 21775,
      "train_speed(iter/s)": 1.635165
    },
    {
      "acc": 0.64814553,
      "epoch": 0.5525114155251142,
      "grad_norm": 6.03125,
      "learning_rate": 8.66452565020135e-06,
      "loss": 1.65093784,
      "memory(GiB)": 97.17,
      "step": 21780,
      "train_speed(iter/s)": 1.635207
    },
    {
      "acc": 0.65193038,
      "epoch": 0.5526382546930492,
      "grad_norm": 6.25,
      "learning_rate": 8.663812158572568e-06,
      "loss": 1.63222961,
      "memory(GiB)": 97.17,
      "step": 21785,
      "train_speed(iter/s)": 1.63525
    },
    {
      "acc": 0.6616415,
      "epoch": 0.5527650938609843,
      "grad_norm": 7.09375,
      "learning_rate": 8.663098505792971e-06,
      "loss": 1.56274929,
      "memory(GiB)": 97.17,
      "step": 21790,
      "train_speed(iter/s)": 1.635294
    },
    {
      "acc": 0.64735346,
      "epoch": 0.5528919330289194,
      "grad_norm": 5.25,
      "learning_rate": 8.662384691893947e-06,
      "loss": 1.59049015,
      "memory(GiB)": 97.17,
      "step": 21795,
      "train_speed(iter/s)": 1.635337
    },
    {
      "acc": 0.65642405,
      "epoch": 0.5530187721968544,
      "grad_norm": 5.40625,
      "learning_rate": 8.661670716906889e-06,
      "loss": 1.52714348,
      "memory(GiB)": 97.17,
      "step": 21800,
      "train_speed(iter/s)": 1.63538
    },
    {
      "acc": 0.65531926,
      "epoch": 0.5531456113647895,
      "grad_norm": 5.96875,
      "learning_rate": 8.66095658086321e-06,
      "loss": 1.67893257,
      "memory(GiB)": 97.17,
      "step": 21805,
      "train_speed(iter/s)": 1.635422
    },
    {
      "acc": 0.64629855,
      "epoch": 0.5532724505327246,
      "grad_norm": 5.71875,
      "learning_rate": 8.660242283794312e-06,
      "loss": 1.60194397,
      "memory(GiB)": 97.17,
      "step": 21810,
      "train_speed(iter/s)": 1.635466
    },
    {
      "acc": 0.64400001,
      "epoch": 0.5533992897006595,
      "grad_norm": 6.5625,
      "learning_rate": 8.659527825731617e-06,
      "loss": 1.66130409,
      "memory(GiB)": 97.17,
      "step": 21815,
      "train_speed(iter/s)": 1.635508
    },
    {
      "acc": 0.64950542,
      "epoch": 0.5535261288685946,
      "grad_norm": 6.21875,
      "learning_rate": 8.65881320670655e-06,
      "loss": 1.60069313,
      "memory(GiB)": 97.17,
      "step": 21820,
      "train_speed(iter/s)": 1.635549
    },
    {
      "acc": 0.65981803,
      "epoch": 0.5536529680365296,
      "grad_norm": 5.375,
      "learning_rate": 8.658098426750543e-06,
      "loss": 1.61864071,
      "memory(GiB)": 97.17,
      "step": 21825,
      "train_speed(iter/s)": 1.635593
    },
    {
      "acc": 0.64138765,
      "epoch": 0.5537798072044647,
      "grad_norm": 4.9375,
      "learning_rate": 8.657383485895034e-06,
      "loss": 1.70235882,
      "memory(GiB)": 97.17,
      "step": 21830,
      "train_speed(iter/s)": 1.635638
    },
    {
      "acc": 0.6707552,
      "epoch": 0.5539066463723998,
      "grad_norm": 5.90625,
      "learning_rate": 8.656668384171472e-06,
      "loss": 1.6362215,
      "memory(GiB)": 97.17,
      "step": 21835,
      "train_speed(iter/s)": 1.635681
    },
    {
      "acc": 0.67066364,
      "epoch": 0.5540334855403348,
      "grad_norm": 7.125,
      "learning_rate": 8.655953121611307e-06,
      "loss": 1.58275146,
      "memory(GiB)": 97.17,
      "step": 21840,
      "train_speed(iter/s)": 1.635725
    },
    {
      "acc": 0.65124965,
      "epoch": 0.5541603247082699,
      "grad_norm": 5.125,
      "learning_rate": 8.655237698246002e-06,
      "loss": 1.62097054,
      "memory(GiB)": 97.17,
      "step": 21845,
      "train_speed(iter/s)": 1.63577
    },
    {
      "acc": 0.65502911,
      "epoch": 0.554287163876205,
      "grad_norm": 6.375,
      "learning_rate": 8.654522114107024e-06,
      "loss": 1.56840305,
      "memory(GiB)": 97.17,
      "step": 21850,
      "train_speed(iter/s)": 1.635812
    },
    {
      "acc": 0.65515871,
      "epoch": 0.55441400304414,
      "grad_norm": 5.53125,
      "learning_rate": 8.653806369225846e-06,
      "loss": 1.60781307,
      "memory(GiB)": 97.17,
      "step": 21855,
      "train_speed(iter/s)": 1.635855
    },
    {
      "acc": 0.64324269,
      "epoch": 0.5545408422120751,
      "grad_norm": 5.875,
      "learning_rate": 8.65309046363395e-06,
      "loss": 1.60527706,
      "memory(GiB)": 97.17,
      "step": 21860,
      "train_speed(iter/s)": 1.635897
    },
    {
      "acc": 0.6480732,
      "epoch": 0.5546676813800101,
      "grad_norm": 5.53125,
      "learning_rate": 8.652374397362828e-06,
      "loss": 1.59619026,
      "memory(GiB)": 97.17,
      "step": 21865,
      "train_speed(iter/s)": 1.635937
    },
    {
      "acc": 0.65826082,
      "epoch": 0.5547945205479452,
      "grad_norm": 4.65625,
      "learning_rate": 8.651658170443972e-06,
      "loss": 1.51217985,
      "memory(GiB)": 97.17,
      "step": 21870,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.65649519,
      "epoch": 0.5549213597158803,
      "grad_norm": 6.15625,
      "learning_rate": 8.650941782908886e-06,
      "loss": 1.58350449,
      "memory(GiB)": 97.17,
      "step": 21875,
      "train_speed(iter/s)": 1.636023
    },
    {
      "acc": 0.66670685,
      "epoch": 0.5550481988838153,
      "grad_norm": 4.34375,
      "learning_rate": 8.65022523478908e-06,
      "loss": 1.53149109,
      "memory(GiB)": 97.17,
      "step": 21880,
      "train_speed(iter/s)": 1.636066
    },
    {
      "acc": 0.65694299,
      "epoch": 0.5551750380517504,
      "grad_norm": 5.5,
      "learning_rate": 8.649508526116073e-06,
      "loss": 1.57991695,
      "memory(GiB)": 97.17,
      "step": 21885,
      "train_speed(iter/s)": 1.63611
    },
    {
      "acc": 0.64611058,
      "epoch": 0.5553018772196855,
      "grad_norm": 6.03125,
      "learning_rate": 8.648791656921384e-06,
      "loss": 1.57871151,
      "memory(GiB)": 97.17,
      "step": 21890,
      "train_speed(iter/s)": 1.636155
    },
    {
      "acc": 0.65757208,
      "epoch": 0.5554287163876205,
      "grad_norm": 5.6875,
      "learning_rate": 8.648074627236549e-06,
      "loss": 1.60997257,
      "memory(GiB)": 97.17,
      "step": 21895,
      "train_speed(iter/s)": 1.636195
    },
    {
      "acc": 0.65856309,
      "epoch": 0.5555555555555556,
      "grad_norm": 7.34375,
      "learning_rate": 8.647357437093104e-06,
      "loss": 1.60625763,
      "memory(GiB)": 97.17,
      "step": 21900,
      "train_speed(iter/s)": 1.636238
    },
    {
      "acc": 0.66049991,
      "epoch": 0.5556823947234906,
      "grad_norm": 5.1875,
      "learning_rate": 8.646640086522595e-06,
      "loss": 1.63476467,
      "memory(GiB)": 97.17,
      "step": 21905,
      "train_speed(iter/s)": 1.636282
    },
    {
      "acc": 0.65205097,
      "epoch": 0.5558092338914257,
      "grad_norm": 5.90625,
      "learning_rate": 8.645922575556575e-06,
      "loss": 1.68087444,
      "memory(GiB)": 97.17,
      "step": 21910,
      "train_speed(iter/s)": 1.636327
    },
    {
      "acc": 0.65282979,
      "epoch": 0.5559360730593608,
      "grad_norm": 4.9375,
      "learning_rate": 8.645204904226601e-06,
      "loss": 1.63369827,
      "memory(GiB)": 97.17,
      "step": 21915,
      "train_speed(iter/s)": 1.63637
    },
    {
      "acc": 0.66559925,
      "epoch": 0.5560629122272958,
      "grad_norm": 6.375,
      "learning_rate": 8.64448707256424e-06,
      "loss": 1.54760084,
      "memory(GiB)": 97.17,
      "step": 21920,
      "train_speed(iter/s)": 1.636416
    },
    {
      "acc": 0.66131949,
      "epoch": 0.5561897513952309,
      "grad_norm": 4.65625,
      "learning_rate": 8.643769080601067e-06,
      "loss": 1.56847315,
      "memory(GiB)": 97.17,
      "step": 21925,
      "train_speed(iter/s)": 1.63646
    },
    {
      "acc": 0.66901631,
      "epoch": 0.556316590563166,
      "grad_norm": 6.03125,
      "learning_rate": 8.643050928368661e-06,
      "loss": 1.54112282,
      "memory(GiB)": 97.17,
      "step": 21930,
      "train_speed(iter/s)": 1.636504
    },
    {
      "acc": 0.6561614,
      "epoch": 0.556443429731101,
      "grad_norm": 6.21875,
      "learning_rate": 8.642332615898611e-06,
      "loss": 1.59263697,
      "memory(GiB)": 97.17,
      "step": 21935,
      "train_speed(iter/s)": 1.636545
    },
    {
      "acc": 0.65573463,
      "epoch": 0.556570268899036,
      "grad_norm": 5.53125,
      "learning_rate": 8.64161414322251e-06,
      "loss": 1.5521946,
      "memory(GiB)": 97.17,
      "step": 21940,
      "train_speed(iter/s)": 1.636586
    },
    {
      "acc": 0.67051821,
      "epoch": 0.556697108066971,
      "grad_norm": 6.0,
      "learning_rate": 8.64089551037196e-06,
      "loss": 1.55594463,
      "memory(GiB)": 97.17,
      "step": 21945,
      "train_speed(iter/s)": 1.636628
    },
    {
      "acc": 0.65225635,
      "epoch": 0.5568239472349061,
      "grad_norm": 4.6875,
      "learning_rate": 8.640176717378573e-06,
      "loss": 1.58434238,
      "memory(GiB)": 97.17,
      "step": 21950,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.66207571,
      "epoch": 0.5569507864028412,
      "grad_norm": 5.9375,
      "learning_rate": 8.639457764273957e-06,
      "loss": 1.63933716,
      "memory(GiB)": 97.17,
      "step": 21955,
      "train_speed(iter/s)": 1.636711
    },
    {
      "acc": 0.63854728,
      "epoch": 0.5570776255707762,
      "grad_norm": 6.53125,
      "learning_rate": 8.638738651089744e-06,
      "loss": 1.63071098,
      "memory(GiB)": 97.17,
      "step": 21960,
      "train_speed(iter/s)": 1.636754
    },
    {
      "acc": 0.64369378,
      "epoch": 0.5572044647387113,
      "grad_norm": 5.9375,
      "learning_rate": 8.638019377857555e-06,
      "loss": 1.6409647,
      "memory(GiB)": 97.17,
      "step": 21965,
      "train_speed(iter/s)": 1.636796
    },
    {
      "acc": 0.65234113,
      "epoch": 0.5573313039066464,
      "grad_norm": 7.8125,
      "learning_rate": 8.637299944609034e-06,
      "loss": 1.69292946,
      "memory(GiB)": 97.17,
      "step": 21970,
      "train_speed(iter/s)": 1.636837
    },
    {
      "acc": 0.66183348,
      "epoch": 0.5574581430745814,
      "grad_norm": 6.9375,
      "learning_rate": 8.636580351375821e-06,
      "loss": 1.62541199,
      "memory(GiB)": 97.17,
      "step": 21975,
      "train_speed(iter/s)": 1.636879
    },
    {
      "acc": 0.64304957,
      "epoch": 0.5575849822425165,
      "grad_norm": 6.5625,
      "learning_rate": 8.635860598189569e-06,
      "loss": 1.70774117,
      "memory(GiB)": 97.17,
      "step": 21980,
      "train_speed(iter/s)": 1.63692
    },
    {
      "acc": 0.64546146,
      "epoch": 0.5577118214104515,
      "grad_norm": 5.1875,
      "learning_rate": 8.635140685081936e-06,
      "loss": 1.66176224,
      "memory(GiB)": 97.17,
      "step": 21985,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.65537148,
      "epoch": 0.5578386605783866,
      "grad_norm": 6.125,
      "learning_rate": 8.634420612084583e-06,
      "loss": 1.57952919,
      "memory(GiB)": 97.17,
      "step": 21990,
      "train_speed(iter/s)": 1.637004
    },
    {
      "acc": 0.64952412,
      "epoch": 0.5579654997463217,
      "grad_norm": 5.25,
      "learning_rate": 8.633700379229187e-06,
      "loss": 1.63313599,
      "memory(GiB)": 97.17,
      "step": 21995,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.65133219,
      "epoch": 0.5580923389142567,
      "grad_norm": 6.125,
      "learning_rate": 8.632979986547423e-06,
      "loss": 1.69516029,
      "memory(GiB)": 97.17,
      "step": 22000,
      "train_speed(iter/s)": 1.63709
    },
    {
      "epoch": 0.5580923389142567,
      "eval_acc": 0.6442837840942076,
      "eval_loss": 1.584275245666504,
      "eval_runtime": 57.9033,
      "eval_samples_per_second": 110.011,
      "eval_steps_per_second": 27.511,
      "step": 22000
    },
    {
      "acc": 0.64766521,
      "epoch": 0.5582191780821918,
      "grad_norm": 5.21875,
      "learning_rate": 8.632259434070982e-06,
      "loss": 1.65357399,
      "memory(GiB)": 97.17,
      "step": 22005,
      "train_speed(iter/s)": 1.629594
    },
    {
      "acc": 0.67265396,
      "epoch": 0.5583460172501269,
      "grad_norm": 5.46875,
      "learning_rate": 8.631538721831551e-06,
      "loss": 1.59529648,
      "memory(GiB)": 97.17,
      "step": 22010,
      "train_speed(iter/s)": 1.62964
    },
    {
      "acc": 0.64182444,
      "epoch": 0.5584728564180619,
      "grad_norm": 4.9375,
      "learning_rate": 8.630817849860835e-06,
      "loss": 1.67003822,
      "memory(GiB)": 97.17,
      "step": 22015,
      "train_speed(iter/s)": 1.629678
    },
    {
      "acc": 0.65431232,
      "epoch": 0.558599695585997,
      "grad_norm": 6.65625,
      "learning_rate": 8.63009681819054e-06,
      "loss": 1.59051275,
      "memory(GiB)": 97.17,
      "step": 22020,
      "train_speed(iter/s)": 1.629718
    },
    {
      "acc": 0.64806523,
      "epoch": 0.558726534753932,
      "grad_norm": 5.34375,
      "learning_rate": 8.629375626852378e-06,
      "loss": 1.58648911,
      "memory(GiB)": 97.17,
      "step": 22025,
      "train_speed(iter/s)": 1.629757
    },
    {
      "acc": 0.64907637,
      "epoch": 0.5588533739218671,
      "grad_norm": 4.5,
      "learning_rate": 8.628654275878074e-06,
      "loss": 1.64522209,
      "memory(GiB)": 97.17,
      "step": 22030,
      "train_speed(iter/s)": 1.629796
    },
    {
      "acc": 0.65537271,
      "epoch": 0.5589802130898022,
      "grad_norm": 5.03125,
      "learning_rate": 8.627932765299353e-06,
      "loss": 1.58960819,
      "memory(GiB)": 97.17,
      "step": 22035,
      "train_speed(iter/s)": 1.629835
    },
    {
      "acc": 0.65957298,
      "epoch": 0.5591070522577372,
      "grad_norm": 5.3125,
      "learning_rate": 8.627211095147952e-06,
      "loss": 1.56123638,
      "memory(GiB)": 97.17,
      "step": 22040,
      "train_speed(iter/s)": 1.62988
    },
    {
      "acc": 0.66828861,
      "epoch": 0.5592338914256723,
      "grad_norm": 6.15625,
      "learning_rate": 8.626489265455614e-06,
      "loss": 1.67097893,
      "memory(GiB)": 97.17,
      "step": 22045,
      "train_speed(iter/s)": 1.629919
    },
    {
      "acc": 0.6471066,
      "epoch": 0.5593607305936074,
      "grad_norm": 6.4375,
      "learning_rate": 8.625767276254084e-06,
      "loss": 1.64611778,
      "memory(GiB)": 97.17,
      "step": 22050,
      "train_speed(iter/s)": 1.629961
    },
    {
      "acc": 0.64427881,
      "epoch": 0.5594875697615423,
      "grad_norm": 5.4375,
      "learning_rate": 8.625045127575123e-06,
      "loss": 1.65598946,
      "memory(GiB)": 97.17,
      "step": 22055,
      "train_speed(iter/s)": 1.63
    },
    {
      "acc": 0.66052151,
      "epoch": 0.5596144089294774,
      "grad_norm": 8.8125,
      "learning_rate": 8.624322819450493e-06,
      "loss": 1.65792542,
      "memory(GiB)": 97.17,
      "step": 22060,
      "train_speed(iter/s)": 1.630041
    },
    {
      "acc": 0.67088552,
      "epoch": 0.5597412480974124,
      "grad_norm": 8.625,
      "learning_rate": 8.623600351911962e-06,
      "loss": 1.54815693,
      "memory(GiB)": 97.17,
      "step": 22065,
      "train_speed(iter/s)": 1.630082
    },
    {
      "acc": 0.66115246,
      "epoch": 0.5598680872653475,
      "grad_norm": 5.0,
      "learning_rate": 8.622877724991312e-06,
      "loss": 1.62382908,
      "memory(GiB)": 97.17,
      "step": 22070,
      "train_speed(iter/s)": 1.630107
    },
    {
      "acc": 0.66171041,
      "epoch": 0.5599949264332826,
      "grad_norm": 4.90625,
      "learning_rate": 8.622154938720323e-06,
      "loss": 1.56196642,
      "memory(GiB)": 97.17,
      "step": 22075,
      "train_speed(iter/s)": 1.630146
    },
    {
      "acc": 0.64730992,
      "epoch": 0.5601217656012176,
      "grad_norm": 5.125,
      "learning_rate": 8.621431993130787e-06,
      "loss": 1.64873505,
      "memory(GiB)": 97.17,
      "step": 22080,
      "train_speed(iter/s)": 1.630185
    },
    {
      "acc": 0.66908636,
      "epoch": 0.5602486047691527,
      "grad_norm": 4.96875,
      "learning_rate": 8.620708888254506e-06,
      "loss": 1.57481318,
      "memory(GiB)": 97.17,
      "step": 22085,
      "train_speed(iter/s)": 1.630224
    },
    {
      "acc": 0.66095982,
      "epoch": 0.5603754439370878,
      "grad_norm": 5.65625,
      "learning_rate": 8.619985624123282e-06,
      "loss": 1.5818306,
      "memory(GiB)": 97.17,
      "step": 22090,
      "train_speed(iter/s)": 1.630265
    },
    {
      "acc": 0.66841831,
      "epoch": 0.5605022831050228,
      "grad_norm": 4.8125,
      "learning_rate": 8.619262200768928e-06,
      "loss": 1.53226871,
      "memory(GiB)": 97.17,
      "step": 22095,
      "train_speed(iter/s)": 1.630305
    },
    {
      "acc": 0.66509643,
      "epoch": 0.5606291222729579,
      "grad_norm": 5.03125,
      "learning_rate": 8.618538618223262e-06,
      "loss": 1.56957045,
      "memory(GiB)": 97.17,
      "step": 22100,
      "train_speed(iter/s)": 1.630342
    },
    {
      "acc": 0.66304598,
      "epoch": 0.5607559614408929,
      "grad_norm": 5.84375,
      "learning_rate": 8.617814876518114e-06,
      "loss": 1.59483376,
      "memory(GiB)": 97.17,
      "step": 22105,
      "train_speed(iter/s)": 1.630384
    },
    {
      "acc": 0.65765972,
      "epoch": 0.560882800608828,
      "grad_norm": 8.25,
      "learning_rate": 8.617090975685314e-06,
      "loss": 1.65395584,
      "memory(GiB)": 97.17,
      "step": 22110,
      "train_speed(iter/s)": 1.630425
    },
    {
      "acc": 0.65512762,
      "epoch": 0.5610096397767631,
      "grad_norm": 5.59375,
      "learning_rate": 8.616366915756704e-06,
      "loss": 1.61534538,
      "memory(GiB)": 97.17,
      "step": 22115,
      "train_speed(iter/s)": 1.630465
    },
    {
      "acc": 0.65322418,
      "epoch": 0.5611364789446981,
      "grad_norm": 5.0,
      "learning_rate": 8.615642696764131e-06,
      "loss": 1.62809296,
      "memory(GiB)": 97.17,
      "step": 22120,
      "train_speed(iter/s)": 1.630508
    },
    {
      "acc": 0.65729551,
      "epoch": 0.5612633181126332,
      "grad_norm": 5.34375,
      "learning_rate": 8.614918318739452e-06,
      "loss": 1.62419853,
      "memory(GiB)": 97.17,
      "step": 22125,
      "train_speed(iter/s)": 1.630547
    },
    {
      "acc": 0.66303067,
      "epoch": 0.5613901572805683,
      "grad_norm": 5.875,
      "learning_rate": 8.614193781714522e-06,
      "loss": 1.61998672,
      "memory(GiB)": 97.17,
      "step": 22130,
      "train_speed(iter/s)": 1.630588
    },
    {
      "acc": 0.65157919,
      "epoch": 0.5615169964485033,
      "grad_norm": 5.5625,
      "learning_rate": 8.613469085721215e-06,
      "loss": 1.61684551,
      "memory(GiB)": 97.17,
      "step": 22135,
      "train_speed(iter/s)": 1.630632
    },
    {
      "acc": 0.65204177,
      "epoch": 0.5616438356164384,
      "grad_norm": 5.53125,
      "learning_rate": 8.612744230791406e-06,
      "loss": 1.67990074,
      "memory(GiB)": 97.17,
      "step": 22140,
      "train_speed(iter/s)": 1.63067
    },
    {
      "acc": 0.66377344,
      "epoch": 0.5617706747843734,
      "grad_norm": 8.125,
      "learning_rate": 8.612019216956975e-06,
      "loss": 1.55788307,
      "memory(GiB)": 97.17,
      "step": 22145,
      "train_speed(iter/s)": 1.63071
    },
    {
      "acc": 0.64535675,
      "epoch": 0.5618975139523085,
      "grad_norm": 5.3125,
      "learning_rate": 8.611294044249811e-06,
      "loss": 1.60758018,
      "memory(GiB)": 97.17,
      "step": 22150,
      "train_speed(iter/s)": 1.630748
    },
    {
      "acc": 0.65348024,
      "epoch": 0.5620243531202436,
      "grad_norm": 5.8125,
      "learning_rate": 8.610568712701814e-06,
      "loss": 1.60966167,
      "memory(GiB)": 97.17,
      "step": 22155,
      "train_speed(iter/s)": 1.630786
    },
    {
      "acc": 0.64837885,
      "epoch": 0.5621511922881786,
      "grad_norm": 5.09375,
      "learning_rate": 8.609843222344883e-06,
      "loss": 1.61449966,
      "memory(GiB)": 97.17,
      "step": 22160,
      "train_speed(iter/s)": 1.63083
    },
    {
      "acc": 0.66407118,
      "epoch": 0.5622780314561137,
      "grad_norm": 5.25,
      "learning_rate": 8.609117573210931e-06,
      "loss": 1.52226009,
      "memory(GiB)": 97.17,
      "step": 22165,
      "train_speed(iter/s)": 1.63087
    },
    {
      "acc": 0.6528976,
      "epoch": 0.5624048706240488,
      "grad_norm": 5.59375,
      "learning_rate": 8.608391765331875e-06,
      "loss": 1.58448124,
      "memory(GiB)": 97.17,
      "step": 22170,
      "train_speed(iter/s)": 1.630915
    },
    {
      "acc": 0.66446767,
      "epoch": 0.5625317097919837,
      "grad_norm": 5.15625,
      "learning_rate": 8.607665798739638e-06,
      "loss": 1.60372181,
      "memory(GiB)": 97.17,
      "step": 22175,
      "train_speed(iter/s)": 1.630954
    },
    {
      "acc": 0.6596478,
      "epoch": 0.5626585489599188,
      "grad_norm": 5.78125,
      "learning_rate": 8.606939673466153e-06,
      "loss": 1.59313192,
      "memory(GiB)": 97.17,
      "step": 22180,
      "train_speed(iter/s)": 1.630994
    },
    {
      "acc": 0.65518103,
      "epoch": 0.5627853881278538,
      "grad_norm": 6.875,
      "learning_rate": 8.606213389543356e-06,
      "loss": 1.58872013,
      "memory(GiB)": 97.17,
      "step": 22185,
      "train_speed(iter/s)": 1.631035
    },
    {
      "acc": 0.65328484,
      "epoch": 0.5629122272957889,
      "grad_norm": 6.40625,
      "learning_rate": 8.605486947003194e-06,
      "loss": 1.65926418,
      "memory(GiB)": 97.17,
      "step": 22190,
      "train_speed(iter/s)": 1.631077
    },
    {
      "acc": 0.64201903,
      "epoch": 0.563039066463724,
      "grad_norm": 4.875,
      "learning_rate": 8.60476034587762e-06,
      "loss": 1.68698311,
      "memory(GiB)": 97.17,
      "step": 22195,
      "train_speed(iter/s)": 1.631113
    },
    {
      "acc": 0.66494994,
      "epoch": 0.563165905631659,
      "grad_norm": 5.40625,
      "learning_rate": 8.604033586198592e-06,
      "loss": 1.57686749,
      "memory(GiB)": 97.17,
      "step": 22200,
      "train_speed(iter/s)": 1.631154
    },
    {
      "acc": 0.63865528,
      "epoch": 0.5632927447995941,
      "grad_norm": 5.96875,
      "learning_rate": 8.603306667998074e-06,
      "loss": 1.67061462,
      "memory(GiB)": 97.17,
      "step": 22205,
      "train_speed(iter/s)": 1.631194
    },
    {
      "acc": 0.6500452,
      "epoch": 0.5634195839675292,
      "grad_norm": 6.03125,
      "learning_rate": 8.602579591308043e-06,
      "loss": 1.64475994,
      "memory(GiB)": 97.17,
      "step": 22210,
      "train_speed(iter/s)": 1.631233
    },
    {
      "acc": 0.63216305,
      "epoch": 0.5635464231354642,
      "grad_norm": 5.34375,
      "learning_rate": 8.601852356160476e-06,
      "loss": 1.68564835,
      "memory(GiB)": 97.17,
      "step": 22215,
      "train_speed(iter/s)": 1.631273
    },
    {
      "acc": 0.66152973,
      "epoch": 0.5636732623033993,
      "grad_norm": 5.28125,
      "learning_rate": 8.60112496258736e-06,
      "loss": 1.59963875,
      "memory(GiB)": 97.17,
      "step": 22220,
      "train_speed(iter/s)": 1.631314
    },
    {
      "acc": 0.64364147,
      "epoch": 0.5638001014713343,
      "grad_norm": 5.125,
      "learning_rate": 8.600397410620693e-06,
      "loss": 1.65571842,
      "memory(GiB)": 97.17,
      "step": 22225,
      "train_speed(iter/s)": 1.631352
    },
    {
      "acc": 0.64903145,
      "epoch": 0.5639269406392694,
      "grad_norm": 6.03125,
      "learning_rate": 8.599669700292472e-06,
      "loss": 1.62756386,
      "memory(GiB)": 97.17,
      "step": 22230,
      "train_speed(iter/s)": 1.631392
    },
    {
      "acc": 0.66847501,
      "epoch": 0.5640537798072045,
      "grad_norm": 5.40625,
      "learning_rate": 8.598941831634707e-06,
      "loss": 1.64775505,
      "memory(GiB)": 97.17,
      "step": 22235,
      "train_speed(iter/s)": 1.631432
    },
    {
      "acc": 0.63823862,
      "epoch": 0.5641806189751395,
      "grad_norm": 5.25,
      "learning_rate": 8.598213804679412e-06,
      "loss": 1.65832157,
      "memory(GiB)": 97.17,
      "step": 22240,
      "train_speed(iter/s)": 1.631473
    },
    {
      "acc": 0.64000854,
      "epoch": 0.5643074581430746,
      "grad_norm": 5.28125,
      "learning_rate": 8.597485619458609e-06,
      "loss": 1.67096252,
      "memory(GiB)": 97.17,
      "step": 22245,
      "train_speed(iter/s)": 1.631503
    },
    {
      "acc": 0.65146627,
      "epoch": 0.5644342973110097,
      "grad_norm": 5.4375,
      "learning_rate": 8.596757276004327e-06,
      "loss": 1.6283865,
      "memory(GiB)": 97.17,
      "step": 22250,
      "train_speed(iter/s)": 1.631543
    },
    {
      "acc": 0.66809855,
      "epoch": 0.5645611364789447,
      "grad_norm": 6.40625,
      "learning_rate": 8.5960287743486e-06,
      "loss": 1.66621056,
      "memory(GiB)": 97.17,
      "step": 22255,
      "train_speed(iter/s)": 1.631588
    },
    {
      "acc": 0.6549943,
      "epoch": 0.5646879756468798,
      "grad_norm": 5.15625,
      "learning_rate": 8.595300114523473e-06,
      "loss": 1.57259054,
      "memory(GiB)": 97.17,
      "step": 22260,
      "train_speed(iter/s)": 1.631631
    },
    {
      "acc": 0.65657454,
      "epoch": 0.5648148148148148,
      "grad_norm": 5.3125,
      "learning_rate": 8.594571296560997e-06,
      "loss": 1.67733383,
      "memory(GiB)": 97.17,
      "step": 22265,
      "train_speed(iter/s)": 1.631672
    },
    {
      "acc": 0.64633541,
      "epoch": 0.5649416539827499,
      "grad_norm": 4.90625,
      "learning_rate": 8.593842320493224e-06,
      "loss": 1.68772202,
      "memory(GiB)": 97.17,
      "step": 22270,
      "train_speed(iter/s)": 1.631712
    },
    {
      "acc": 0.64857774,
      "epoch": 0.565068493150685,
      "grad_norm": 5.625,
      "learning_rate": 8.593113186352222e-06,
      "loss": 1.68307953,
      "memory(GiB)": 97.17,
      "step": 22275,
      "train_speed(iter/s)": 1.631755
    },
    {
      "acc": 0.63205538,
      "epoch": 0.56519533231862,
      "grad_norm": 5.625,
      "learning_rate": 8.592383894170059e-06,
      "loss": 1.68476067,
      "memory(GiB)": 97.17,
      "step": 22280,
      "train_speed(iter/s)": 1.631799
    },
    {
      "acc": 0.66157036,
      "epoch": 0.5653221714865551,
      "grad_norm": 4.625,
      "learning_rate": 8.591654443978815e-06,
      "loss": 1.59444284,
      "memory(GiB)": 97.17,
      "step": 22285,
      "train_speed(iter/s)": 1.63184
    },
    {
      "acc": 0.65190382,
      "epoch": 0.5654490106544902,
      "grad_norm": 6.125,
      "learning_rate": 8.590924835810572e-06,
      "loss": 1.61812744,
      "memory(GiB)": 97.17,
      "step": 22290,
      "train_speed(iter/s)": 1.631885
    },
    {
      "acc": 0.65776253,
      "epoch": 0.5655758498224251,
      "grad_norm": 6.25,
      "learning_rate": 8.590195069697423e-06,
      "loss": 1.56847477,
      "memory(GiB)": 97.17,
      "step": 22295,
      "train_speed(iter/s)": 1.631928
    },
    {
      "acc": 0.64089284,
      "epoch": 0.5657026889903602,
      "grad_norm": 5.875,
      "learning_rate": 8.589465145671465e-06,
      "loss": 1.63800068,
      "memory(GiB)": 97.17,
      "step": 22300,
      "train_speed(iter/s)": 1.631969
    },
    {
      "acc": 0.6443018,
      "epoch": 0.5658295281582952,
      "grad_norm": 5.21875,
      "learning_rate": 8.588735063764803e-06,
      "loss": 1.61475258,
      "memory(GiB)": 97.17,
      "step": 22305,
      "train_speed(iter/s)": 1.63201
    },
    {
      "acc": 0.65383434,
      "epoch": 0.5659563673262303,
      "grad_norm": 4.96875,
      "learning_rate": 8.588004824009552e-06,
      "loss": 1.68520279,
      "memory(GiB)": 97.17,
      "step": 22310,
      "train_speed(iter/s)": 1.632051
    },
    {
      "acc": 0.6525898,
      "epoch": 0.5660832064941654,
      "grad_norm": 6.0625,
      "learning_rate": 8.58727442643783e-06,
      "loss": 1.67151756,
      "memory(GiB)": 97.17,
      "step": 22315,
      "train_speed(iter/s)": 1.632089
    },
    {
      "acc": 0.64363141,
      "epoch": 0.5662100456621004,
      "grad_norm": 6.40625,
      "learning_rate": 8.586543871081764e-06,
      "loss": 1.63221016,
      "memory(GiB)": 97.17,
      "step": 22320,
      "train_speed(iter/s)": 1.63213
    },
    {
      "acc": 0.67659674,
      "epoch": 0.5663368848300355,
      "grad_norm": 5.1875,
      "learning_rate": 8.585813157973482e-06,
      "loss": 1.53684492,
      "memory(GiB)": 97.17,
      "step": 22325,
      "train_speed(iter/s)": 1.632171
    },
    {
      "acc": 0.64077759,
      "epoch": 0.5664637239979706,
      "grad_norm": 6.53125,
      "learning_rate": 8.58508228714513e-06,
      "loss": 1.59774857,
      "memory(GiB)": 97.17,
      "step": 22330,
      "train_speed(iter/s)": 1.632214
    },
    {
      "acc": 0.65861969,
      "epoch": 0.5665905631659056,
      "grad_norm": 6.40625,
      "learning_rate": 8.584351258628852e-06,
      "loss": 1.58166103,
      "memory(GiB)": 97.17,
      "step": 22335,
      "train_speed(iter/s)": 1.632254
    },
    {
      "acc": 0.6759655,
      "epoch": 0.5667174023338407,
      "grad_norm": 5.375,
      "learning_rate": 8.583620072456803e-06,
      "loss": 1.48119345,
      "memory(GiB)": 97.17,
      "step": 22340,
      "train_speed(iter/s)": 1.632295
    },
    {
      "acc": 0.66174827,
      "epoch": 0.5668442415017757,
      "grad_norm": 5.34375,
      "learning_rate": 8.582888728661142e-06,
      "loss": 1.59675293,
      "memory(GiB)": 97.17,
      "step": 22345,
      "train_speed(iter/s)": 1.632337
    },
    {
      "acc": 0.64102669,
      "epoch": 0.5669710806697108,
      "grad_norm": 5.875,
      "learning_rate": 8.582157227274042e-06,
      "loss": 1.68126278,
      "memory(GiB)": 97.17,
      "step": 22350,
      "train_speed(iter/s)": 1.632378
    },
    {
      "acc": 0.65328808,
      "epoch": 0.5670979198376459,
      "grad_norm": 6.75,
      "learning_rate": 8.581425568327671e-06,
      "loss": 1.62424183,
      "memory(GiB)": 97.17,
      "step": 22355,
      "train_speed(iter/s)": 1.632416
    },
    {
      "acc": 0.64292893,
      "epoch": 0.5672247590055809,
      "grad_norm": 5.34375,
      "learning_rate": 8.580693751854215e-06,
      "loss": 1.65654449,
      "memory(GiB)": 97.17,
      "step": 22360,
      "train_speed(iter/s)": 1.632459
    },
    {
      "acc": 0.64939899,
      "epoch": 0.567351598173516,
      "grad_norm": 5.0,
      "learning_rate": 8.57996177788586e-06,
      "loss": 1.5958786,
      "memory(GiB)": 97.17,
      "step": 22365,
      "train_speed(iter/s)": 1.632501
    },
    {
      "acc": 0.64354606,
      "epoch": 0.5674784373414511,
      "grad_norm": 7.09375,
      "learning_rate": 8.579229646454803e-06,
      "loss": 1.63148613,
      "memory(GiB)": 97.17,
      "step": 22370,
      "train_speed(iter/s)": 1.632543
    },
    {
      "acc": 0.67264624,
      "epoch": 0.5676052765093861,
      "grad_norm": 8.4375,
      "learning_rate": 8.578497357593246e-06,
      "loss": 1.5287075,
      "memory(GiB)": 97.17,
      "step": 22375,
      "train_speed(iter/s)": 1.632585
    },
    {
      "acc": 0.66717358,
      "epoch": 0.5677321156773212,
      "grad_norm": 5.34375,
      "learning_rate": 8.5777649113334e-06,
      "loss": 1.49966593,
      "memory(GiB)": 97.17,
      "step": 22380,
      "train_speed(iter/s)": 1.632623
    },
    {
      "acc": 0.65953608,
      "epoch": 0.5678589548452562,
      "grad_norm": 5.28125,
      "learning_rate": 8.577032307707476e-06,
      "loss": 1.57445078,
      "memory(GiB)": 97.17,
      "step": 22385,
      "train_speed(iter/s)": 1.632665
    },
    {
      "acc": 0.63723788,
      "epoch": 0.5679857940131913,
      "grad_norm": 5.75,
      "learning_rate": 8.576299546747704e-06,
      "loss": 1.70777283,
      "memory(GiB)": 97.17,
      "step": 22390,
      "train_speed(iter/s)": 1.632708
    },
    {
      "acc": 0.65935545,
      "epoch": 0.5681126331811264,
      "grad_norm": 5.625,
      "learning_rate": 8.575566628486309e-06,
      "loss": 1.63568611,
      "memory(GiB)": 97.17,
      "step": 22395,
      "train_speed(iter/s)": 1.632748
    },
    {
      "acc": 0.63334737,
      "epoch": 0.5682394723490614,
      "grad_norm": 5.5,
      "learning_rate": 8.574833552955532e-06,
      "loss": 1.65445709,
      "memory(GiB)": 97.17,
      "step": 22400,
      "train_speed(iter/s)": 1.63279
    },
    {
      "acc": 0.6508224,
      "epoch": 0.5683663115169965,
      "grad_norm": 4.71875,
      "learning_rate": 8.574100320187612e-06,
      "loss": 1.64084568,
      "memory(GiB)": 97.17,
      "step": 22405,
      "train_speed(iter/s)": 1.63283
    },
    {
      "acc": 0.65611391,
      "epoch": 0.5684931506849316,
      "grad_norm": 5.8125,
      "learning_rate": 8.573366930214807e-06,
      "loss": 1.66153507,
      "memory(GiB)": 97.17,
      "step": 22410,
      "train_speed(iter/s)": 1.632872
    },
    {
      "acc": 0.63149695,
      "epoch": 0.5686199898528665,
      "grad_norm": 6.71875,
      "learning_rate": 8.572633383069366e-06,
      "loss": 1.67252502,
      "memory(GiB)": 97.17,
      "step": 22415,
      "train_speed(iter/s)": 1.632916
    },
    {
      "acc": 0.64451799,
      "epoch": 0.5687468290208016,
      "grad_norm": 12.4375,
      "learning_rate": 8.571899678783561e-06,
      "loss": 1.63048859,
      "memory(GiB)": 97.17,
      "step": 22420,
      "train_speed(iter/s)": 1.632956
    },
    {
      "acc": 0.65323577,
      "epoch": 0.5688736681887366,
      "grad_norm": 6.3125,
      "learning_rate": 8.57116581738966e-06,
      "loss": 1.62588692,
      "memory(GiB)": 97.17,
      "step": 22425,
      "train_speed(iter/s)": 1.632997
    },
    {
      "acc": 0.66419601,
      "epoch": 0.5690005073566717,
      "grad_norm": 5.21875,
      "learning_rate": 8.570431798919941e-06,
      "loss": 1.57818089,
      "memory(GiB)": 97.17,
      "step": 22430,
      "train_speed(iter/s)": 1.633037
    },
    {
      "acc": 0.6607265,
      "epoch": 0.5691273465246068,
      "grad_norm": 6.5625,
      "learning_rate": 8.569697623406692e-06,
      "loss": 1.56275806,
      "memory(GiB)": 97.17,
      "step": 22435,
      "train_speed(iter/s)": 1.633076
    },
    {
      "acc": 0.64556789,
      "epoch": 0.5692541856925418,
      "grad_norm": 5.09375,
      "learning_rate": 8.568963290882204e-06,
      "loss": 1.62211933,
      "memory(GiB)": 97.17,
      "step": 22440,
      "train_speed(iter/s)": 1.633116
    },
    {
      "acc": 0.66603861,
      "epoch": 0.5693810248604769,
      "grad_norm": 5.375,
      "learning_rate": 8.568228801378775e-06,
      "loss": 1.53854589,
      "memory(GiB)": 97.17,
      "step": 22445,
      "train_speed(iter/s)": 1.633155
    },
    {
      "acc": 0.65906096,
      "epoch": 0.569507864028412,
      "grad_norm": 4.96875,
      "learning_rate": 8.567494154928713e-06,
      "loss": 1.58619699,
      "memory(GiB)": 97.17,
      "step": 22450,
      "train_speed(iter/s)": 1.633195
    },
    {
      "acc": 0.63740587,
      "epoch": 0.569634703196347,
      "grad_norm": 5.125,
      "learning_rate": 8.566759351564332e-06,
      "loss": 1.61090679,
      "memory(GiB)": 97.17,
      "step": 22455,
      "train_speed(iter/s)": 1.633233
    },
    {
      "acc": 0.64091873,
      "epoch": 0.5697615423642821,
      "grad_norm": 5.40625,
      "learning_rate": 8.566024391317947e-06,
      "loss": 1.65458221,
      "memory(GiB)": 97.17,
      "step": 22460,
      "train_speed(iter/s)": 1.633272
    },
    {
      "acc": 0.65131426,
      "epoch": 0.5698883815322171,
      "grad_norm": 6.90625,
      "learning_rate": 8.565289274221891e-06,
      "loss": 1.61053772,
      "memory(GiB)": 97.17,
      "step": 22465,
      "train_speed(iter/s)": 1.633312
    },
    {
      "acc": 0.66191001,
      "epoch": 0.5700152207001522,
      "grad_norm": 4.96875,
      "learning_rate": 8.564554000308493e-06,
      "loss": 1.5993679,
      "memory(GiB)": 97.17,
      "step": 22470,
      "train_speed(iter/s)": 1.633354
    },
    {
      "acc": 0.6503726,
      "epoch": 0.5701420598680873,
      "grad_norm": 6.15625,
      "learning_rate": 8.563818569610096e-06,
      "loss": 1.62359886,
      "memory(GiB)": 97.17,
      "step": 22475,
      "train_speed(iter/s)": 1.633396
    },
    {
      "acc": 0.64764481,
      "epoch": 0.5702688990360223,
      "grad_norm": 4.96875,
      "learning_rate": 8.563082982159048e-06,
      "loss": 1.59813824,
      "memory(GiB)": 97.17,
      "step": 22480,
      "train_speed(iter/s)": 1.633435
    },
    {
      "acc": 0.66095142,
      "epoch": 0.5703957382039574,
      "grad_norm": 6.09375,
      "learning_rate": 8.562347237987701e-06,
      "loss": 1.62702637,
      "memory(GiB)": 97.17,
      "step": 22485,
      "train_speed(iter/s)": 1.633474
    },
    {
      "acc": 0.63881736,
      "epoch": 0.5705225773718925,
      "grad_norm": 6.0625,
      "learning_rate": 8.561611337128418e-06,
      "loss": 1.67207108,
      "memory(GiB)": 97.17,
      "step": 22490,
      "train_speed(iter/s)": 1.633511
    },
    {
      "acc": 0.6611743,
      "epoch": 0.5706494165398275,
      "grad_norm": 5.71875,
      "learning_rate": 8.560875279613568e-06,
      "loss": 1.57725773,
      "memory(GiB)": 97.17,
      "step": 22495,
      "train_speed(iter/s)": 1.633549
    },
    {
      "acc": 0.64042716,
      "epoch": 0.5707762557077626,
      "grad_norm": 5.15625,
      "learning_rate": 8.560139065475523e-06,
      "loss": 1.66806068,
      "memory(GiB)": 97.17,
      "step": 22500,
      "train_speed(iter/s)": 1.633587
    },
    {
      "acc": 0.66801863,
      "epoch": 0.5709030948756976,
      "grad_norm": 6.40625,
      "learning_rate": 8.559402694746671e-06,
      "loss": 1.6008194,
      "memory(GiB)": 97.17,
      "step": 22505,
      "train_speed(iter/s)": 1.633629
    },
    {
      "acc": 0.66383257,
      "epoch": 0.5710299340436327,
      "grad_norm": 5.5625,
      "learning_rate": 8.558666167459393e-06,
      "loss": 1.52449493,
      "memory(GiB)": 97.17,
      "step": 22510,
      "train_speed(iter/s)": 1.633674
    },
    {
      "acc": 0.64486103,
      "epoch": 0.5711567732115678,
      "grad_norm": 5.03125,
      "learning_rate": 8.55792948364609e-06,
      "loss": 1.73176956,
      "memory(GiB)": 97.17,
      "step": 22515,
      "train_speed(iter/s)": 1.633716
    },
    {
      "acc": 0.66222563,
      "epoch": 0.5712836123795028,
      "grad_norm": 5.96875,
      "learning_rate": 8.557192643339164e-06,
      "loss": 1.65326118,
      "memory(GiB)": 97.17,
      "step": 22520,
      "train_speed(iter/s)": 1.633754
    },
    {
      "acc": 0.66218786,
      "epoch": 0.5714104515474379,
      "grad_norm": 4.59375,
      "learning_rate": 8.556455646571022e-06,
      "loss": 1.61443806,
      "memory(GiB)": 97.17,
      "step": 22525,
      "train_speed(iter/s)": 1.633794
    },
    {
      "acc": 0.66605749,
      "epoch": 0.571537290715373,
      "grad_norm": 5.5625,
      "learning_rate": 8.555718493374084e-06,
      "loss": 1.64509468,
      "memory(GiB)": 97.17,
      "step": 22530,
      "train_speed(iter/s)": 1.633831
    },
    {
      "acc": 0.66166363,
      "epoch": 0.571664129883308,
      "grad_norm": 5.5,
      "learning_rate": 8.55498118378077e-06,
      "loss": 1.63057213,
      "memory(GiB)": 97.17,
      "step": 22535,
      "train_speed(iter/s)": 1.63387
    },
    {
      "acc": 0.6782208,
      "epoch": 0.571790969051243,
      "grad_norm": 7.65625,
      "learning_rate": 8.554243717823512e-06,
      "loss": 1.50373659,
      "memory(GiB)": 97.17,
      "step": 22540,
      "train_speed(iter/s)": 1.633912
    },
    {
      "acc": 0.65800934,
      "epoch": 0.571917808219178,
      "grad_norm": 5.65625,
      "learning_rate": 8.553506095534747e-06,
      "loss": 1.64103966,
      "memory(GiB)": 97.17,
      "step": 22545,
      "train_speed(iter/s)": 1.633952
    },
    {
      "acc": 0.65278802,
      "epoch": 0.5720446473871131,
      "grad_norm": 5.625,
      "learning_rate": 8.55276831694692e-06,
      "loss": 1.55767307,
      "memory(GiB)": 97.17,
      "step": 22550,
      "train_speed(iter/s)": 1.633994
    },
    {
      "acc": 0.6616468,
      "epoch": 0.5721714865550482,
      "grad_norm": 5.59375,
      "learning_rate": 8.552030382092477e-06,
      "loss": 1.61699657,
      "memory(GiB)": 97.17,
      "step": 22555,
      "train_speed(iter/s)": 1.634033
    },
    {
      "acc": 0.66322451,
      "epoch": 0.5722983257229832,
      "grad_norm": 5.625,
      "learning_rate": 8.551292291003884e-06,
      "loss": 1.62340508,
      "memory(GiB)": 97.17,
      "step": 22560,
      "train_speed(iter/s)": 1.634074
    },
    {
      "acc": 0.65315456,
      "epoch": 0.5724251648909183,
      "grad_norm": 8.0625,
      "learning_rate": 8.550554043713597e-06,
      "loss": 1.63757877,
      "memory(GiB)": 97.17,
      "step": 22565,
      "train_speed(iter/s)": 1.634115
    },
    {
      "acc": 0.66580782,
      "epoch": 0.5725520040588534,
      "grad_norm": 5.53125,
      "learning_rate": 8.549815640254092e-06,
      "loss": 1.56148205,
      "memory(GiB)": 97.17,
      "step": 22570,
      "train_speed(iter/s)": 1.634157
    },
    {
      "acc": 0.65758343,
      "epoch": 0.5726788432267884,
      "grad_norm": 6.25,
      "learning_rate": 8.549077080657846e-06,
      "loss": 1.59915657,
      "memory(GiB)": 97.17,
      "step": 22575,
      "train_speed(iter/s)": 1.634199
    },
    {
      "acc": 0.63893986,
      "epoch": 0.5728056823947235,
      "grad_norm": 6.5,
      "learning_rate": 8.548338364957345e-06,
      "loss": 1.68260727,
      "memory(GiB)": 97.17,
      "step": 22580,
      "train_speed(iter/s)": 1.634238
    },
    {
      "acc": 0.64738617,
      "epoch": 0.5729325215626585,
      "grad_norm": 5.5,
      "learning_rate": 8.54759949318508e-06,
      "loss": 1.64903946,
      "memory(GiB)": 97.17,
      "step": 22585,
      "train_speed(iter/s)": 1.634278
    },
    {
      "acc": 0.65078812,
      "epoch": 0.5730593607305936,
      "grad_norm": 4.78125,
      "learning_rate": 8.546860465373552e-06,
      "loss": 1.5880373,
      "memory(GiB)": 97.17,
      "step": 22590,
      "train_speed(iter/s)": 1.634317
    },
    {
      "acc": 0.65682545,
      "epoch": 0.5731861998985287,
      "grad_norm": 5.71875,
      "learning_rate": 8.546121281555265e-06,
      "loss": 1.57653236,
      "memory(GiB)": 97.17,
      "step": 22595,
      "train_speed(iter/s)": 1.634359
    },
    {
      "acc": 0.66010704,
      "epoch": 0.5733130390664637,
      "grad_norm": 5.40625,
      "learning_rate": 8.54538194176273e-06,
      "loss": 1.56372452,
      "memory(GiB)": 97.17,
      "step": 22600,
      "train_speed(iter/s)": 1.634396
    },
    {
      "acc": 0.6544301,
      "epoch": 0.5734398782343988,
      "grad_norm": 6.5625,
      "learning_rate": 8.544642446028469e-06,
      "loss": 1.60253029,
      "memory(GiB)": 97.17,
      "step": 22605,
      "train_speed(iter/s)": 1.634438
    },
    {
      "acc": 0.6301075,
      "epoch": 0.5735667174023339,
      "grad_norm": 8.1875,
      "learning_rate": 8.543902794385008e-06,
      "loss": 1.73892765,
      "memory(GiB)": 97.17,
      "step": 22610,
      "train_speed(iter/s)": 1.634477
    },
    {
      "acc": 0.6411973,
      "epoch": 0.5736935565702689,
      "grad_norm": 5.71875,
      "learning_rate": 8.543162986864879e-06,
      "loss": 1.66419106,
      "memory(GiB)": 97.17,
      "step": 22615,
      "train_speed(iter/s)": 1.634515
    },
    {
      "acc": 0.67401361,
      "epoch": 0.573820395738204,
      "grad_norm": 5.5625,
      "learning_rate": 8.542423023500623e-06,
      "loss": 1.54650631,
      "memory(GiB)": 97.17,
      "step": 22620,
      "train_speed(iter/s)": 1.634553
    },
    {
      "acc": 0.64794912,
      "epoch": 0.573947234906139,
      "grad_norm": 6.21875,
      "learning_rate": 8.541682904324786e-06,
      "loss": 1.64625969,
      "memory(GiB)": 97.17,
      "step": 22625,
      "train_speed(iter/s)": 1.63459
    },
    {
      "acc": 0.64722643,
      "epoch": 0.5740740740740741,
      "grad_norm": 5.5625,
      "learning_rate": 8.540942629369923e-06,
      "loss": 1.65565987,
      "memory(GiB)": 97.17,
      "step": 22630,
      "train_speed(iter/s)": 1.634628
    },
    {
      "acc": 0.63343868,
      "epoch": 0.5742009132420092,
      "grad_norm": 5.21875,
      "learning_rate": 8.540202198668595e-06,
      "loss": 1.63734856,
      "memory(GiB)": 97.17,
      "step": 22635,
      "train_speed(iter/s)": 1.634667
    },
    {
      "acc": 0.65869322,
      "epoch": 0.5743277524099442,
      "grad_norm": 5.28125,
      "learning_rate": 8.539461612253368e-06,
      "loss": 1.5885767,
      "memory(GiB)": 97.17,
      "step": 22640,
      "train_speed(iter/s)": 1.634706
    },
    {
      "acc": 0.65664492,
      "epoch": 0.5744545915778793,
      "grad_norm": 5.46875,
      "learning_rate": 8.538720870156816e-06,
      "loss": 1.59492931,
      "memory(GiB)": 97.17,
      "step": 22645,
      "train_speed(iter/s)": 1.634745
    },
    {
      "acc": 0.65455604,
      "epoch": 0.5745814307458144,
      "grad_norm": 5.21875,
      "learning_rate": 8.53797997241152e-06,
      "loss": 1.57693586,
      "memory(GiB)": 97.17,
      "step": 22650,
      "train_speed(iter/s)": 1.634784
    },
    {
      "acc": 0.66212964,
      "epoch": 0.5747082699137493,
      "grad_norm": 7.03125,
      "learning_rate": 8.537238919050071e-06,
      "loss": 1.59262638,
      "memory(GiB)": 97.17,
      "step": 22655,
      "train_speed(iter/s)": 1.634826
    },
    {
      "acc": 0.6656559,
      "epoch": 0.5748351090816844,
      "grad_norm": 5.0625,
      "learning_rate": 8.53649771010506e-06,
      "loss": 1.56936579,
      "memory(GiB)": 97.17,
      "step": 22660,
      "train_speed(iter/s)": 1.634864
    },
    {
      "acc": 0.65250397,
      "epoch": 0.5749619482496194,
      "grad_norm": 5.375,
      "learning_rate": 8.535756345609092e-06,
      "loss": 1.6102972,
      "memory(GiB)": 97.17,
      "step": 22665,
      "train_speed(iter/s)": 1.634906
    },
    {
      "acc": 0.66066313,
      "epoch": 0.5750887874175545,
      "grad_norm": 5.96875,
      "learning_rate": 8.535014825594772e-06,
      "loss": 1.59062881,
      "memory(GiB)": 97.17,
      "step": 22670,
      "train_speed(iter/s)": 1.634941
    },
    {
      "acc": 0.65874252,
      "epoch": 0.5752156265854896,
      "grad_norm": 5.03125,
      "learning_rate": 8.534273150094718e-06,
      "loss": 1.59291534,
      "memory(GiB)": 97.17,
      "step": 22675,
      "train_speed(iter/s)": 1.634978
    },
    {
      "acc": 0.64439807,
      "epoch": 0.5753424657534246,
      "grad_norm": 7.28125,
      "learning_rate": 8.533531319141552e-06,
      "loss": 1.57150679,
      "memory(GiB)": 97.17,
      "step": 22680,
      "train_speed(iter/s)": 1.635014
    },
    {
      "acc": 0.63908682,
      "epoch": 0.5754693049213597,
      "grad_norm": 6.28125,
      "learning_rate": 8.532789332767902e-06,
      "loss": 1.69566898,
      "memory(GiB)": 97.17,
      "step": 22685,
      "train_speed(iter/s)": 1.635049
    },
    {
      "acc": 0.65716305,
      "epoch": 0.5755961440892948,
      "grad_norm": 5.625,
      "learning_rate": 8.532047191006405e-06,
      "loss": 1.58032408,
      "memory(GiB)": 97.17,
      "step": 22690,
      "train_speed(iter/s)": 1.635088
    },
    {
      "acc": 0.64233589,
      "epoch": 0.5757229832572298,
      "grad_norm": 4.375,
      "learning_rate": 8.531304893889702e-06,
      "loss": 1.63509083,
      "memory(GiB)": 97.17,
      "step": 22695,
      "train_speed(iter/s)": 1.635126
    },
    {
      "acc": 0.66026225,
      "epoch": 0.5758498224251649,
      "grad_norm": 5.5625,
      "learning_rate": 8.530562441450445e-06,
      "loss": 1.63984337,
      "memory(GiB)": 97.17,
      "step": 22700,
      "train_speed(iter/s)": 1.635163
    },
    {
      "acc": 0.65600543,
      "epoch": 0.5759766615930999,
      "grad_norm": 5.65625,
      "learning_rate": 8.529819833721289e-06,
      "loss": 1.63429909,
      "memory(GiB)": 97.17,
      "step": 22705,
      "train_speed(iter/s)": 1.635203
    },
    {
      "acc": 0.65073228,
      "epoch": 0.576103500761035,
      "grad_norm": 5.96875,
      "learning_rate": 8.529077070734896e-06,
      "loss": 1.65823822,
      "memory(GiB)": 97.17,
      "step": 22710,
      "train_speed(iter/s)": 1.635242
    },
    {
      "acc": 0.65583858,
      "epoch": 0.5762303399289701,
      "grad_norm": 6.28125,
      "learning_rate": 8.528334152523938e-06,
      "loss": 1.6523138,
      "memory(GiB)": 97.17,
      "step": 22715,
      "train_speed(iter/s)": 1.635281
    },
    {
      "acc": 0.67079296,
      "epoch": 0.5763571790969051,
      "grad_norm": 5.75,
      "learning_rate": 8.52759107912109e-06,
      "loss": 1.5369688,
      "memory(GiB)": 97.17,
      "step": 22720,
      "train_speed(iter/s)": 1.635321
    },
    {
      "acc": 0.66434717,
      "epoch": 0.5764840182648402,
      "grad_norm": 5.4375,
      "learning_rate": 8.526847850559037e-06,
      "loss": 1.59597836,
      "memory(GiB)": 97.17,
      "step": 22725,
      "train_speed(iter/s)": 1.635359
    },
    {
      "acc": 0.65048466,
      "epoch": 0.5766108574327753,
      "grad_norm": 4.78125,
      "learning_rate": 8.526104466870472e-06,
      "loss": 1.64152584,
      "memory(GiB)": 97.17,
      "step": 22730,
      "train_speed(iter/s)": 1.635396
    },
    {
      "acc": 0.66162724,
      "epoch": 0.5767376966007103,
      "grad_norm": 5.09375,
      "learning_rate": 8.525360928088087e-06,
      "loss": 1.57986431,
      "memory(GiB)": 97.17,
      "step": 22735,
      "train_speed(iter/s)": 1.635434
    },
    {
      "acc": 0.65751557,
      "epoch": 0.5768645357686454,
      "grad_norm": 5.5,
      "learning_rate": 8.524617234244588e-06,
      "loss": 1.59875603,
      "memory(GiB)": 97.17,
      "step": 22740,
      "train_speed(iter/s)": 1.635471
    },
    {
      "acc": 0.65876346,
      "epoch": 0.5769913749365804,
      "grad_norm": 5.71875,
      "learning_rate": 8.523873385372687e-06,
      "loss": 1.50989676,
      "memory(GiB)": 97.17,
      "step": 22745,
      "train_speed(iter/s)": 1.635509
    },
    {
      "acc": 0.64990516,
      "epoch": 0.5771182141045155,
      "grad_norm": 6.28125,
      "learning_rate": 8.523129381505104e-06,
      "loss": 1.69517059,
      "memory(GiB)": 97.17,
      "step": 22750,
      "train_speed(iter/s)": 1.635546
    },
    {
      "acc": 0.67025118,
      "epoch": 0.5772450532724506,
      "grad_norm": 5.71875,
      "learning_rate": 8.522385222674559e-06,
      "loss": 1.55408869,
      "memory(GiB)": 97.17,
      "step": 22755,
      "train_speed(iter/s)": 1.635585
    },
    {
      "acc": 0.660672,
      "epoch": 0.5773718924403856,
      "grad_norm": 5.21875,
      "learning_rate": 8.521640908913787e-06,
      "loss": 1.54079924,
      "memory(GiB)": 97.17,
      "step": 22760,
      "train_speed(iter/s)": 1.635621
    },
    {
      "acc": 0.66065736,
      "epoch": 0.5774987316083207,
      "grad_norm": 5.125,
      "learning_rate": 8.520896440255524e-06,
      "loss": 1.63035107,
      "memory(GiB)": 97.17,
      "step": 22765,
      "train_speed(iter/s)": 1.635659
    },
    {
      "acc": 0.66402907,
      "epoch": 0.5776255707762558,
      "grad_norm": 5.96875,
      "learning_rate": 8.520151816732517e-06,
      "loss": 1.56851654,
      "memory(GiB)": 97.17,
      "step": 22770,
      "train_speed(iter/s)": 1.6357
    },
    {
      "acc": 0.63997631,
      "epoch": 0.5777524099441907,
      "grad_norm": 6.0625,
      "learning_rate": 8.519407038377515e-06,
      "loss": 1.70682335,
      "memory(GiB)": 97.17,
      "step": 22775,
      "train_speed(iter/s)": 1.635741
    },
    {
      "acc": 0.64541235,
      "epoch": 0.5778792491121258,
      "grad_norm": 5.28125,
      "learning_rate": 8.518662105223279e-06,
      "loss": 1.66096973,
      "memory(GiB)": 97.17,
      "step": 22780,
      "train_speed(iter/s)": 1.63578
    },
    {
      "acc": 0.65397167,
      "epoch": 0.5780060882800608,
      "grad_norm": 5.6875,
      "learning_rate": 8.517917017302574e-06,
      "loss": 1.63193169,
      "memory(GiB)": 97.17,
      "step": 22785,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.65304327,
      "epoch": 0.5781329274479959,
      "grad_norm": 5.15625,
      "learning_rate": 8.517171774648172e-06,
      "loss": 1.65042076,
      "memory(GiB)": 97.17,
      "step": 22790,
      "train_speed(iter/s)": 1.635854
    },
    {
      "acc": 0.6451889,
      "epoch": 0.578259766615931,
      "grad_norm": 5.40625,
      "learning_rate": 8.516426377292854e-06,
      "loss": 1.6351757,
      "memory(GiB)": 97.17,
      "step": 22795,
      "train_speed(iter/s)": 1.635892
    },
    {
      "acc": 0.65759945,
      "epoch": 0.578386605783866,
      "grad_norm": 5.0625,
      "learning_rate": 8.515680825269404e-06,
      "loss": 1.57706127,
      "memory(GiB)": 97.17,
      "step": 22800,
      "train_speed(iter/s)": 1.635931
    },
    {
      "acc": 0.65407743,
      "epoch": 0.5785134449518011,
      "grad_norm": 5.75,
      "learning_rate": 8.514935118610613e-06,
      "loss": 1.67828522,
      "memory(GiB)": 97.17,
      "step": 22805,
      "train_speed(iter/s)": 1.635969
    },
    {
      "acc": 0.65324202,
      "epoch": 0.5786402841197362,
      "grad_norm": 5.125,
      "learning_rate": 8.514189257349283e-06,
      "loss": 1.62905502,
      "memory(GiB)": 97.17,
      "step": 22810,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.66132698,
      "epoch": 0.5787671232876712,
      "grad_norm": 5.375,
      "learning_rate": 8.51344324151822e-06,
      "loss": 1.57477112,
      "memory(GiB)": 97.17,
      "step": 22815,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.64995775,
      "epoch": 0.5788939624556063,
      "grad_norm": 6.09375,
      "learning_rate": 8.512697071150235e-06,
      "loss": 1.61292915,
      "memory(GiB)": 97.17,
      "step": 22820,
      "train_speed(iter/s)": 1.636074
    },
    {
      "acc": 0.65169153,
      "epoch": 0.5790208016235413,
      "grad_norm": 5.875,
      "learning_rate": 8.511950746278152e-06,
      "loss": 1.49183826,
      "memory(GiB)": 97.17,
      "step": 22825,
      "train_speed(iter/s)": 1.636112
    },
    {
      "acc": 0.65585341,
      "epoch": 0.5791476407914764,
      "grad_norm": 6.75,
      "learning_rate": 8.511204266934797e-06,
      "loss": 1.62278996,
      "memory(GiB)": 97.17,
      "step": 22830,
      "train_speed(iter/s)": 1.636148
    },
    {
      "acc": 0.6411128,
      "epoch": 0.5792744799594115,
      "grad_norm": 5.65625,
      "learning_rate": 8.510457633152998e-06,
      "loss": 1.70634956,
      "memory(GiB)": 97.17,
      "step": 22835,
      "train_speed(iter/s)": 1.636185
    },
    {
      "acc": 0.64888144,
      "epoch": 0.5794013191273465,
      "grad_norm": 4.9375,
      "learning_rate": 8.509710844965602e-06,
      "loss": 1.612883,
      "memory(GiB)": 97.17,
      "step": 22840,
      "train_speed(iter/s)": 1.636222
    },
    {
      "acc": 0.64987893,
      "epoch": 0.5795281582952816,
      "grad_norm": 7.71875,
      "learning_rate": 8.508963902405451e-06,
      "loss": 1.69131012,
      "memory(GiB)": 97.17,
      "step": 22845,
      "train_speed(iter/s)": 1.636258
    },
    {
      "acc": 0.6553628,
      "epoch": 0.5796549974632167,
      "grad_norm": 6.5,
      "learning_rate": 8.508216805505403e-06,
      "loss": 1.58818827,
      "memory(GiB)": 97.17,
      "step": 22850,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.6612669,
      "epoch": 0.5797818366311517,
      "grad_norm": 4.5625,
      "learning_rate": 8.507469554298318e-06,
      "loss": 1.63925438,
      "memory(GiB)": 97.17,
      "step": 22855,
      "train_speed(iter/s)": 1.63633
    },
    {
      "acc": 0.66252761,
      "epoch": 0.5799086757990868,
      "grad_norm": 5.625,
      "learning_rate": 8.506722148817061e-06,
      "loss": 1.58173122,
      "memory(GiB)": 97.17,
      "step": 22860,
      "train_speed(iter/s)": 1.636367
    },
    {
      "acc": 0.66712084,
      "epoch": 0.5800355149670218,
      "grad_norm": 5.40625,
      "learning_rate": 8.505974589094505e-06,
      "loss": 1.55390739,
      "memory(GiB)": 97.17,
      "step": 22865,
      "train_speed(iter/s)": 1.636403
    },
    {
      "acc": 0.66790862,
      "epoch": 0.5801623541349569,
      "grad_norm": 4.75,
      "learning_rate": 8.505226875163537e-06,
      "loss": 1.62106781,
      "memory(GiB)": 97.17,
      "step": 22870,
      "train_speed(iter/s)": 1.63644
    },
    {
      "acc": 0.65671058,
      "epoch": 0.580289193302892,
      "grad_norm": 6.375,
      "learning_rate": 8.50447900705704e-06,
      "loss": 1.59848099,
      "memory(GiB)": 97.17,
      "step": 22875,
      "train_speed(iter/s)": 1.636481
    },
    {
      "acc": 0.66197948,
      "epoch": 0.580416032470827,
      "grad_norm": 4.375,
      "learning_rate": 8.503730984807911e-06,
      "loss": 1.58144665,
      "memory(GiB)": 97.17,
      "step": 22880,
      "train_speed(iter/s)": 1.636519
    },
    {
      "acc": 0.67121081,
      "epoch": 0.5805428716387621,
      "grad_norm": 5.21875,
      "learning_rate": 8.502982808449049e-06,
      "loss": 1.55431309,
      "memory(GiB)": 97.17,
      "step": 22885,
      "train_speed(iter/s)": 1.636556
    },
    {
      "acc": 0.64571066,
      "epoch": 0.5806697108066972,
      "grad_norm": 5.0,
      "learning_rate": 8.502234478013363e-06,
      "loss": 1.67499714,
      "memory(GiB)": 97.17,
      "step": 22890,
      "train_speed(iter/s)": 1.636596
    },
    {
      "acc": 0.64868345,
      "epoch": 0.5807965499746321,
      "grad_norm": 6.9375,
      "learning_rate": 8.501485993533769e-06,
      "loss": 1.64944916,
      "memory(GiB)": 97.17,
      "step": 22895,
      "train_speed(iter/s)": 1.636634
    },
    {
      "acc": 0.65402269,
      "epoch": 0.5809233891425672,
      "grad_norm": 5.71875,
      "learning_rate": 8.500737355043188e-06,
      "loss": 1.63367748,
      "memory(GiB)": 97.17,
      "step": 22900,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.65983505,
      "epoch": 0.5810502283105022,
      "grad_norm": 5.21875,
      "learning_rate": 8.499988562574549e-06,
      "loss": 1.6229641,
      "memory(GiB)": 97.17,
      "step": 22905,
      "train_speed(iter/s)": 1.636714
    },
    {
      "acc": 0.66664605,
      "epoch": 0.5811770674784373,
      "grad_norm": 8.1875,
      "learning_rate": 8.499239616160787e-06,
      "loss": 1.64632587,
      "memory(GiB)": 97.17,
      "step": 22910,
      "train_speed(iter/s)": 1.63675
    },
    {
      "acc": 0.66269217,
      "epoch": 0.5813039066463724,
      "grad_norm": 6.09375,
      "learning_rate": 8.498490515834841e-06,
      "loss": 1.62370987,
      "memory(GiB)": 97.17,
      "step": 22915,
      "train_speed(iter/s)": 1.636789
    },
    {
      "acc": 0.64714632,
      "epoch": 0.5814307458143074,
      "grad_norm": 4.75,
      "learning_rate": 8.497741261629664e-06,
      "loss": 1.62714901,
      "memory(GiB)": 97.17,
      "step": 22920,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.64612112,
      "epoch": 0.5815575849822425,
      "grad_norm": 5.28125,
      "learning_rate": 8.496991853578212e-06,
      "loss": 1.62256603,
      "memory(GiB)": 97.17,
      "step": 22925,
      "train_speed(iter/s)": 1.636865
    },
    {
      "acc": 0.65187206,
      "epoch": 0.5816844241501776,
      "grad_norm": 5.375,
      "learning_rate": 8.496242291713444e-06,
      "loss": 1.62594223,
      "memory(GiB)": 97.17,
      "step": 22930,
      "train_speed(iter/s)": 1.636904
    },
    {
      "acc": 0.66642528,
      "epoch": 0.5818112633181126,
      "grad_norm": 6.6875,
      "learning_rate": 8.495492576068329e-06,
      "loss": 1.62148819,
      "memory(GiB)": 97.17,
      "step": 22935,
      "train_speed(iter/s)": 1.636943
    },
    {
      "acc": 0.65136757,
      "epoch": 0.5819381024860477,
      "grad_norm": 4.84375,
      "learning_rate": 8.494742706675844e-06,
      "loss": 1.65615921,
      "memory(GiB)": 97.17,
      "step": 22940,
      "train_speed(iter/s)": 1.636981
    },
    {
      "acc": 0.6433547,
      "epoch": 0.5820649416539827,
      "grad_norm": 5.78125,
      "learning_rate": 8.493992683568975e-06,
      "loss": 1.70139961,
      "memory(GiB)": 97.17,
      "step": 22945,
      "train_speed(iter/s)": 1.63702
    },
    {
      "acc": 0.65435853,
      "epoch": 0.5821917808219178,
      "grad_norm": 7.84375,
      "learning_rate": 8.493242506780705e-06,
      "loss": 1.59594765,
      "memory(GiB)": 97.17,
      "step": 22950,
      "train_speed(iter/s)": 1.637059
    },
    {
      "acc": 0.64634666,
      "epoch": 0.5823186199898529,
      "grad_norm": 7.03125,
      "learning_rate": 8.492492176344035e-06,
      "loss": 1.69558048,
      "memory(GiB)": 97.17,
      "step": 22955,
      "train_speed(iter/s)": 1.637099
    },
    {
      "acc": 0.66459703,
      "epoch": 0.5824454591577879,
      "grad_norm": 4.78125,
      "learning_rate": 8.491741692291967e-06,
      "loss": 1.59905815,
      "memory(GiB)": 97.17,
      "step": 22960,
      "train_speed(iter/s)": 1.637137
    },
    {
      "acc": 0.64039135,
      "epoch": 0.582572298325723,
      "grad_norm": 6.84375,
      "learning_rate": 8.490991054657507e-06,
      "loss": 1.68901482,
      "memory(GiB)": 97.17,
      "step": 22965,
      "train_speed(iter/s)": 1.637175
    },
    {
      "acc": 0.65616636,
      "epoch": 0.5826991374936581,
      "grad_norm": 6.21875,
      "learning_rate": 8.490240263473677e-06,
      "loss": 1.6567852,
      "memory(GiB)": 97.17,
      "step": 22970,
      "train_speed(iter/s)": 1.637213
    },
    {
      "acc": 0.64903359,
      "epoch": 0.5828259766615931,
      "grad_norm": 7.46875,
      "learning_rate": 8.489489318773496e-06,
      "loss": 1.64451199,
      "memory(GiB)": 97.17,
      "step": 22975,
      "train_speed(iter/s)": 1.637253
    },
    {
      "acc": 0.64575744,
      "epoch": 0.5829528158295282,
      "grad_norm": 5.9375,
      "learning_rate": 8.488738220589996e-06,
      "loss": 1.68805923,
      "memory(GiB)": 97.17,
      "step": 22980,
      "train_speed(iter/s)": 1.637291
    },
    {
      "acc": 0.66106524,
      "epoch": 0.5830796549974632,
      "grad_norm": 5.46875,
      "learning_rate": 8.487986968956212e-06,
      "loss": 1.56896286,
      "memory(GiB)": 97.17,
      "step": 22985,
      "train_speed(iter/s)": 1.637329
    },
    {
      "acc": 0.65841284,
      "epoch": 0.5832064941653983,
      "grad_norm": 5.28125,
      "learning_rate": 8.487235563905191e-06,
      "loss": 1.61763344,
      "memory(GiB)": 97.17,
      "step": 22990,
      "train_speed(iter/s)": 1.637368
    },
    {
      "acc": 0.65555654,
      "epoch": 0.5833333333333334,
      "grad_norm": 5.8125,
      "learning_rate": 8.486484005469977e-06,
      "loss": 1.60066929,
      "memory(GiB)": 97.17,
      "step": 22995,
      "train_speed(iter/s)": 1.637405
    },
    {
      "acc": 0.657195,
      "epoch": 0.5834601725012684,
      "grad_norm": 5.25,
      "learning_rate": 8.485732293683633e-06,
      "loss": 1.58927183,
      "memory(GiB)": 97.17,
      "step": 23000,
      "train_speed(iter/s)": 1.637443
    },
    {
      "epoch": 0.5834601725012684,
      "eval_acc": 0.6443664810162533,
      "eval_loss": 1.584184169769287,
      "eval_runtime": 58.432,
      "eval_samples_per_second": 109.016,
      "eval_steps_per_second": 27.262,
      "step": 23000
    },
    {
      "acc": 0.65083895,
      "epoch": 0.5835870116692035,
      "grad_norm": 6.75,
      "learning_rate": 8.48498042857922e-06,
      "loss": 1.6316246,
      "memory(GiB)": 97.17,
      "step": 23005,
      "train_speed(iter/s)": 1.630208
    },
    {
      "acc": 0.64590511,
      "epoch": 0.5837138508371386,
      "grad_norm": 6.09375,
      "learning_rate": 8.484228410189807e-06,
      "loss": 1.63669891,
      "memory(GiB)": 97.17,
      "step": 23010,
      "train_speed(iter/s)": 1.630245
    },
    {
      "acc": 0.63782701,
      "epoch": 0.5838406900050735,
      "grad_norm": 5.96875,
      "learning_rate": 8.483476238548473e-06,
      "loss": 1.70055771,
      "memory(GiB)": 97.17,
      "step": 23015,
      "train_speed(iter/s)": 1.630282
    },
    {
      "acc": 0.65304079,
      "epoch": 0.5839675291730086,
      "grad_norm": 4.96875,
      "learning_rate": 8.482723913688301e-06,
      "loss": 1.66720009,
      "memory(GiB)": 97.17,
      "step": 23020,
      "train_speed(iter/s)": 1.630319
    },
    {
      "acc": 0.66204033,
      "epoch": 0.5840943683409436,
      "grad_norm": 6.34375,
      "learning_rate": 8.481971435642382e-06,
      "loss": 1.60161018,
      "memory(GiB)": 97.17,
      "step": 23025,
      "train_speed(iter/s)": 1.630357
    },
    {
      "acc": 0.66512694,
      "epoch": 0.5842212075088787,
      "grad_norm": 5.28125,
      "learning_rate": 8.481218804443814e-06,
      "loss": 1.57576027,
      "memory(GiB)": 97.17,
      "step": 23030,
      "train_speed(iter/s)": 1.630395
    },
    {
      "acc": 0.64378653,
      "epoch": 0.5843480466768138,
      "grad_norm": 6.5,
      "learning_rate": 8.480466020125701e-06,
      "loss": 1.6231741,
      "memory(GiB)": 97.17,
      "step": 23035,
      "train_speed(iter/s)": 1.630433
    },
    {
      "acc": 0.64961295,
      "epoch": 0.5844748858447488,
      "grad_norm": 4.75,
      "learning_rate": 8.479713082721153e-06,
      "loss": 1.65742702,
      "memory(GiB)": 97.17,
      "step": 23040,
      "train_speed(iter/s)": 1.630469
    },
    {
      "acc": 0.67278495,
      "epoch": 0.5846017250126839,
      "grad_norm": 6.15625,
      "learning_rate": 8.478959992263288e-06,
      "loss": 1.54251881,
      "memory(GiB)": 97.17,
      "step": 23045,
      "train_speed(iter/s)": 1.630505
    },
    {
      "acc": 0.64318628,
      "epoch": 0.584728564180619,
      "grad_norm": 5.53125,
      "learning_rate": 8.478206748785229e-06,
      "loss": 1.67044601,
      "memory(GiB)": 97.17,
      "step": 23050,
      "train_speed(iter/s)": 1.630545
    },
    {
      "acc": 0.65506592,
      "epoch": 0.584855403348554,
      "grad_norm": 5.625,
      "learning_rate": 8.477453352320108e-06,
      "loss": 1.58791437,
      "memory(GiB)": 97.17,
      "step": 23055,
      "train_speed(iter/s)": 1.630582
    },
    {
      "acc": 0.6587451,
      "epoch": 0.5849822425164891,
      "grad_norm": 5.03125,
      "learning_rate": 8.476699802901066e-06,
      "loss": 1.60707893,
      "memory(GiB)": 97.17,
      "step": 23060,
      "train_speed(iter/s)": 1.630619
    },
    {
      "acc": 0.65943556,
      "epoch": 0.5851090816844241,
      "grad_norm": 5.6875,
      "learning_rate": 8.47594610056124e-06,
      "loss": 1.56292238,
      "memory(GiB)": 97.17,
      "step": 23065,
      "train_speed(iter/s)": 1.630658
    },
    {
      "acc": 0.64979725,
      "epoch": 0.5852359208523592,
      "grad_norm": 5.875,
      "learning_rate": 8.475192245333787e-06,
      "loss": 1.61738892,
      "memory(GiB)": 97.17,
      "step": 23070,
      "train_speed(iter/s)": 1.630696
    },
    {
      "acc": 0.65580435,
      "epoch": 0.5853627600202943,
      "grad_norm": 6.875,
      "learning_rate": 8.474438237251864e-06,
      "loss": 1.56334438,
      "memory(GiB)": 97.17,
      "step": 23075,
      "train_speed(iter/s)": 1.630732
    },
    {
      "acc": 0.66290398,
      "epoch": 0.5854895991882293,
      "grad_norm": 5.15625,
      "learning_rate": 8.473684076348635e-06,
      "loss": 1.54914141,
      "memory(GiB)": 97.17,
      "step": 23080,
      "train_speed(iter/s)": 1.630769
    },
    {
      "acc": 0.65610676,
      "epoch": 0.5856164383561644,
      "grad_norm": 6.0,
      "learning_rate": 8.472929762657272e-06,
      "loss": 1.61477509,
      "memory(GiB)": 97.17,
      "step": 23085,
      "train_speed(iter/s)": 1.630807
    },
    {
      "acc": 0.65753269,
      "epoch": 0.5857432775240995,
      "grad_norm": 5.59375,
      "learning_rate": 8.472175296210952e-06,
      "loss": 1.58797789,
      "memory(GiB)": 97.17,
      "step": 23090,
      "train_speed(iter/s)": 1.630846
    },
    {
      "acc": 0.65164843,
      "epoch": 0.5858701166920345,
      "grad_norm": 6.0625,
      "learning_rate": 8.471420677042858e-06,
      "loss": 1.6396883,
      "memory(GiB)": 97.17,
      "step": 23095,
      "train_speed(iter/s)": 1.630884
    },
    {
      "acc": 0.66111522,
      "epoch": 0.5859969558599696,
      "grad_norm": 5.78125,
      "learning_rate": 8.470665905186188e-06,
      "loss": 1.57690125,
      "memory(GiB)": 97.17,
      "step": 23100,
      "train_speed(iter/s)": 1.630923
    },
    {
      "acc": 0.65413284,
      "epoch": 0.5861237950279046,
      "grad_norm": 5.09375,
      "learning_rate": 8.469910980674134e-06,
      "loss": 1.65788784,
      "memory(GiB)": 97.17,
      "step": 23105,
      "train_speed(iter/s)": 1.630961
    },
    {
      "acc": 0.67559872,
      "epoch": 0.5862506341958397,
      "grad_norm": 5.34375,
      "learning_rate": 8.469155903539903e-06,
      "loss": 1.48754435,
      "memory(GiB)": 97.17,
      "step": 23110,
      "train_speed(iter/s)": 1.630998
    },
    {
      "acc": 0.64601326,
      "epoch": 0.5863774733637748,
      "grad_norm": 7.65625,
      "learning_rate": 8.468400673816705e-06,
      "loss": 1.64300041,
      "memory(GiB)": 97.17,
      "step": 23115,
      "train_speed(iter/s)": 1.631034
    },
    {
      "acc": 0.65507307,
      "epoch": 0.5865043125317098,
      "grad_norm": 6.09375,
      "learning_rate": 8.467645291537763e-06,
      "loss": 1.67484093,
      "memory(GiB)": 97.17,
      "step": 23120,
      "train_speed(iter/s)": 1.631073
    },
    {
      "acc": 0.64951906,
      "epoch": 0.5866311516996449,
      "grad_norm": 5.53125,
      "learning_rate": 8.466889756736298e-06,
      "loss": 1.57847137,
      "memory(GiB)": 97.17,
      "step": 23125,
      "train_speed(iter/s)": 1.631114
    },
    {
      "acc": 0.64380484,
      "epoch": 0.58675799086758,
      "grad_norm": 4.96875,
      "learning_rate": 8.466134069445544e-06,
      "loss": 1.59181137,
      "memory(GiB)": 97.17,
      "step": 23130,
      "train_speed(iter/s)": 1.631152
    },
    {
      "acc": 0.65185699,
      "epoch": 0.586884830035515,
      "grad_norm": 5.21875,
      "learning_rate": 8.465378229698737e-06,
      "loss": 1.66399231,
      "memory(GiB)": 97.17,
      "step": 23135,
      "train_speed(iter/s)": 1.63119
    },
    {
      "acc": 0.64459105,
      "epoch": 0.58701166920345,
      "grad_norm": 5.59375,
      "learning_rate": 8.464622237529123e-06,
      "loss": 1.66215572,
      "memory(GiB)": 97.17,
      "step": 23140,
      "train_speed(iter/s)": 1.631228
    },
    {
      "acc": 0.66941447,
      "epoch": 0.587138508371385,
      "grad_norm": 5.9375,
      "learning_rate": 8.463866092969958e-06,
      "loss": 1.54132729,
      "memory(GiB)": 97.17,
      "step": 23145,
      "train_speed(iter/s)": 1.631266
    },
    {
      "acc": 0.66022167,
      "epoch": 0.5872653475393201,
      "grad_norm": 5.5,
      "learning_rate": 8.463109796054495e-06,
      "loss": 1.61912155,
      "memory(GiB)": 97.17,
      "step": 23150,
      "train_speed(iter/s)": 1.631303
    },
    {
      "acc": 0.65391273,
      "epoch": 0.5873921867072552,
      "grad_norm": 6.28125,
      "learning_rate": 8.462353346815999e-06,
      "loss": 1.6016737,
      "memory(GiB)": 97.17,
      "step": 23155,
      "train_speed(iter/s)": 1.631332
    },
    {
      "acc": 0.65436144,
      "epoch": 0.5875190258751902,
      "grad_norm": 5.96875,
      "learning_rate": 8.461596745287747e-06,
      "loss": 1.5995245,
      "memory(GiB)": 97.17,
      "step": 23160,
      "train_speed(iter/s)": 1.631372
    },
    {
      "acc": 0.64279499,
      "epoch": 0.5876458650431253,
      "grad_norm": 5.9375,
      "learning_rate": 8.460839991503016e-06,
      "loss": 1.673703,
      "memory(GiB)": 97.17,
      "step": 23165,
      "train_speed(iter/s)": 1.631411
    },
    {
      "acc": 0.66852322,
      "epoch": 0.5877727042110604,
      "grad_norm": 6.6875,
      "learning_rate": 8.46008308549509e-06,
      "loss": 1.54446135,
      "memory(GiB)": 97.17,
      "step": 23170,
      "train_speed(iter/s)": 1.631449
    },
    {
      "acc": 0.65555248,
      "epoch": 0.5878995433789954,
      "grad_norm": 5.625,
      "learning_rate": 8.459326027297261e-06,
      "loss": 1.59135494,
      "memory(GiB)": 97.17,
      "step": 23175,
      "train_speed(iter/s)": 1.631488
    },
    {
      "acc": 0.65714149,
      "epoch": 0.5880263825469305,
      "grad_norm": 6.03125,
      "learning_rate": 8.45856881694283e-06,
      "loss": 1.59860697,
      "memory(GiB)": 97.17,
      "step": 23180,
      "train_speed(iter/s)": 1.631528
    },
    {
      "acc": 0.65539322,
      "epoch": 0.5881532217148655,
      "grad_norm": 5.6875,
      "learning_rate": 8.4578114544651e-06,
      "loss": 1.58906803,
      "memory(GiB)": 97.17,
      "step": 23185,
      "train_speed(iter/s)": 1.631563
    },
    {
      "acc": 0.65719471,
      "epoch": 0.5882800608828006,
      "grad_norm": 6.53125,
      "learning_rate": 8.457053939897385e-06,
      "loss": 1.62150211,
      "memory(GiB)": 97.17,
      "step": 23190,
      "train_speed(iter/s)": 1.631602
    },
    {
      "acc": 0.66549149,
      "epoch": 0.5884069000507357,
      "grad_norm": 5.09375,
      "learning_rate": 8.456296273273e-06,
      "loss": 1.58194494,
      "memory(GiB)": 97.17,
      "step": 23195,
      "train_speed(iter/s)": 1.631641
    },
    {
      "acc": 0.65814161,
      "epoch": 0.5885337392186707,
      "grad_norm": 6.125,
      "learning_rate": 8.455538454625276e-06,
      "loss": 1.60630798,
      "memory(GiB)": 97.17,
      "step": 23200,
      "train_speed(iter/s)": 1.631681
    },
    {
      "acc": 0.65053844,
      "epoch": 0.5886605783866058,
      "grad_norm": 5.90625,
      "learning_rate": 8.454780483987544e-06,
      "loss": 1.63077755,
      "memory(GiB)": 97.17,
      "step": 23205,
      "train_speed(iter/s)": 1.631719
    },
    {
      "acc": 0.66433411,
      "epoch": 0.5887874175545409,
      "grad_norm": 8.5,
      "learning_rate": 8.45402236139314e-06,
      "loss": 1.56801281,
      "memory(GiB)": 97.17,
      "step": 23210,
      "train_speed(iter/s)": 1.63176
    },
    {
      "acc": 0.65297689,
      "epoch": 0.5889142567224759,
      "grad_norm": 5.1875,
      "learning_rate": 8.453264086875411e-06,
      "loss": 1.6130888,
      "memory(GiB)": 97.17,
      "step": 23215,
      "train_speed(iter/s)": 1.631799
    },
    {
      "acc": 0.65417309,
      "epoch": 0.589041095890411,
      "grad_norm": 6.1875,
      "learning_rate": 8.452505660467713e-06,
      "loss": 1.53566599,
      "memory(GiB)": 97.17,
      "step": 23220,
      "train_speed(iter/s)": 1.631839
    },
    {
      "acc": 0.66245646,
      "epoch": 0.589167935058346,
      "grad_norm": 6.90625,
      "learning_rate": 8.451747082203398e-06,
      "loss": 1.62393475,
      "memory(GiB)": 97.17,
      "step": 23225,
      "train_speed(iter/s)": 1.631878
    },
    {
      "acc": 0.66943555,
      "epoch": 0.5892947742262811,
      "grad_norm": 7.625,
      "learning_rate": 8.450988352115838e-06,
      "loss": 1.55491238,
      "memory(GiB)": 97.17,
      "step": 23230,
      "train_speed(iter/s)": 1.631918
    },
    {
      "acc": 0.65429621,
      "epoch": 0.5894216133942162,
      "grad_norm": 5.25,
      "learning_rate": 8.450229470238401e-06,
      "loss": 1.58933592,
      "memory(GiB)": 97.17,
      "step": 23235,
      "train_speed(iter/s)": 1.631957
    },
    {
      "acc": 0.64573588,
      "epoch": 0.5895484525621512,
      "grad_norm": 7.75,
      "learning_rate": 8.44947043660447e-06,
      "loss": 1.62904663,
      "memory(GiB)": 97.17,
      "step": 23240,
      "train_speed(iter/s)": 1.631996
    },
    {
      "acc": 0.67295542,
      "epoch": 0.5896752917300863,
      "grad_norm": 6.125,
      "learning_rate": 8.448711251247425e-06,
      "loss": 1.56359625,
      "memory(GiB)": 97.17,
      "step": 23245,
      "train_speed(iter/s)": 1.632037
    },
    {
      "acc": 0.655369,
      "epoch": 0.5898021308980214,
      "grad_norm": 6.40625,
      "learning_rate": 8.447951914200665e-06,
      "loss": 1.62867508,
      "memory(GiB)": 97.17,
      "step": 23250,
      "train_speed(iter/s)": 1.632074
    },
    {
      "acc": 0.66799998,
      "epoch": 0.5899289700659563,
      "grad_norm": 5.5625,
      "learning_rate": 8.447192425497583e-06,
      "loss": 1.56641235,
      "memory(GiB)": 97.17,
      "step": 23255,
      "train_speed(iter/s)": 1.632112
    },
    {
      "acc": 0.6660368,
      "epoch": 0.5900558092338914,
      "grad_norm": 6.21875,
      "learning_rate": 8.44643278517159e-06,
      "loss": 1.56193514,
      "memory(GiB)": 97.17,
      "step": 23260,
      "train_speed(iter/s)": 1.63215
    },
    {
      "acc": 0.63970351,
      "epoch": 0.5901826484018264,
      "grad_norm": 5.15625,
      "learning_rate": 8.445672993256095e-06,
      "loss": 1.71108532,
      "memory(GiB)": 97.17,
      "step": 23265,
      "train_speed(iter/s)": 1.632187
    },
    {
      "acc": 0.64268646,
      "epoch": 0.5903094875697615,
      "grad_norm": 5.9375,
      "learning_rate": 8.444913049784517e-06,
      "loss": 1.67581463,
      "memory(GiB)": 97.17,
      "step": 23270,
      "train_speed(iter/s)": 1.632226
    },
    {
      "acc": 0.67111602,
      "epoch": 0.5904363267376966,
      "grad_norm": 5.96875,
      "learning_rate": 8.444152954790285e-06,
      "loss": 1.57265453,
      "memory(GiB)": 97.17,
      "step": 23275,
      "train_speed(iter/s)": 1.632264
    },
    {
      "acc": 0.65613003,
      "epoch": 0.5905631659056316,
      "grad_norm": 5.53125,
      "learning_rate": 8.443392708306827e-06,
      "loss": 1.63115654,
      "memory(GiB)": 97.17,
      "step": 23280,
      "train_speed(iter/s)": 1.632301
    },
    {
      "acc": 0.66380405,
      "epoch": 0.5906900050735667,
      "grad_norm": 7.21875,
      "learning_rate": 8.442632310367585e-06,
      "loss": 1.57424507,
      "memory(GiB)": 97.17,
      "step": 23285,
      "train_speed(iter/s)": 1.632342
    },
    {
      "acc": 0.65739794,
      "epoch": 0.5908168442415018,
      "grad_norm": 5.75,
      "learning_rate": 8.441871761006001e-06,
      "loss": 1.5503912,
      "memory(GiB)": 97.17,
      "step": 23290,
      "train_speed(iter/s)": 1.632381
    },
    {
      "acc": 0.6517333,
      "epoch": 0.5909436834094368,
      "grad_norm": 6.5625,
      "learning_rate": 8.441111060255533e-06,
      "loss": 1.66126156,
      "memory(GiB)": 97.17,
      "step": 23295,
      "train_speed(iter/s)": 1.63242
    },
    {
      "acc": 0.62545977,
      "epoch": 0.5910705225773719,
      "grad_norm": 6.125,
      "learning_rate": 8.440350208149637e-06,
      "loss": 1.72088184,
      "memory(GiB)": 97.17,
      "step": 23300,
      "train_speed(iter/s)": 1.632458
    },
    {
      "acc": 0.65832124,
      "epoch": 0.5911973617453069,
      "grad_norm": 7.875,
      "learning_rate": 8.43958920472178e-06,
      "loss": 1.58804693,
      "memory(GiB)": 97.17,
      "step": 23305,
      "train_speed(iter/s)": 1.632498
    },
    {
      "acc": 0.65824728,
      "epoch": 0.591324200913242,
      "grad_norm": 4.9375,
      "learning_rate": 8.43882805000543e-06,
      "loss": 1.61714001,
      "memory(GiB)": 97.17,
      "step": 23310,
      "train_speed(iter/s)": 1.632533
    },
    {
      "acc": 0.64468489,
      "epoch": 0.5914510400811771,
      "grad_norm": 5.5625,
      "learning_rate": 8.43806674403407e-06,
      "loss": 1.67449455,
      "memory(GiB)": 97.17,
      "step": 23315,
      "train_speed(iter/s)": 1.632573
    },
    {
      "acc": 0.66782417,
      "epoch": 0.5915778792491121,
      "grad_norm": 6.1875,
      "learning_rate": 8.437305286841187e-06,
      "loss": 1.52976942,
      "memory(GiB)": 97.17,
      "step": 23320,
      "train_speed(iter/s)": 1.632613
    },
    {
      "acc": 0.67024045,
      "epoch": 0.5917047184170472,
      "grad_norm": 5.28125,
      "learning_rate": 8.436543678460269e-06,
      "loss": 1.58486519,
      "memory(GiB)": 97.17,
      "step": 23325,
      "train_speed(iter/s)": 1.632651
    },
    {
      "acc": 0.66463103,
      "epoch": 0.5918315575849823,
      "grad_norm": 7.0,
      "learning_rate": 8.435781918924817e-06,
      "loss": 1.59040165,
      "memory(GiB)": 97.17,
      "step": 23330,
      "train_speed(iter/s)": 1.632688
    },
    {
      "acc": 0.66600232,
      "epoch": 0.5919583967529173,
      "grad_norm": 5.71875,
      "learning_rate": 8.435020008268335e-06,
      "loss": 1.57260599,
      "memory(GiB)": 97.17,
      "step": 23335,
      "train_speed(iter/s)": 1.632725
    },
    {
      "acc": 0.65437956,
      "epoch": 0.5920852359208524,
      "grad_norm": 6.75,
      "learning_rate": 8.43425794652434e-06,
      "loss": 1.60292473,
      "memory(GiB)": 97.17,
      "step": 23340,
      "train_speed(iter/s)": 1.632762
    },
    {
      "acc": 0.64356432,
      "epoch": 0.5922120750887874,
      "grad_norm": 5.125,
      "learning_rate": 8.433495733726345e-06,
      "loss": 1.665802,
      "memory(GiB)": 97.17,
      "step": 23345,
      "train_speed(iter/s)": 1.632801
    },
    {
      "acc": 0.64941578,
      "epoch": 0.5923389142567225,
      "grad_norm": 4.8125,
      "learning_rate": 8.43273336990788e-06,
      "loss": 1.66789627,
      "memory(GiB)": 97.17,
      "step": 23350,
      "train_speed(iter/s)": 1.632839
    },
    {
      "acc": 0.66172667,
      "epoch": 0.5924657534246576,
      "grad_norm": 5.5625,
      "learning_rate": 8.431970855102475e-06,
      "loss": 1.59821272,
      "memory(GiB)": 97.17,
      "step": 23355,
      "train_speed(iter/s)": 1.63288
    },
    {
      "acc": 0.66407843,
      "epoch": 0.5925925925925926,
      "grad_norm": 4.46875,
      "learning_rate": 8.43120818934367e-06,
      "loss": 1.57593365,
      "memory(GiB)": 97.17,
      "step": 23360,
      "train_speed(iter/s)": 1.632921
    },
    {
      "acc": 0.65724597,
      "epoch": 0.5927194317605277,
      "grad_norm": 5.6875,
      "learning_rate": 8.430445372665008e-06,
      "loss": 1.64276714,
      "memory(GiB)": 97.17,
      "step": 23365,
      "train_speed(iter/s)": 1.632961
    },
    {
      "acc": 0.6491847,
      "epoch": 0.5928462709284628,
      "grad_norm": 5.40625,
      "learning_rate": 8.429682405100042e-06,
      "loss": 1.67211685,
      "memory(GiB)": 97.17,
      "step": 23370,
      "train_speed(iter/s)": 1.632999
    },
    {
      "acc": 0.66703463,
      "epoch": 0.5929731100963977,
      "grad_norm": 5.6875,
      "learning_rate": 8.428919286682333e-06,
      "loss": 1.55302324,
      "memory(GiB)": 97.17,
      "step": 23375,
      "train_speed(iter/s)": 1.633038
    },
    {
      "acc": 0.66369238,
      "epoch": 0.5930999492643328,
      "grad_norm": 5.46875,
      "learning_rate": 8.428156017445443e-06,
      "loss": 1.57563419,
      "memory(GiB)": 97.17,
      "step": 23380,
      "train_speed(iter/s)": 1.633074
    },
    {
      "acc": 0.65099092,
      "epoch": 0.5932267884322678,
      "grad_norm": 5.875,
      "learning_rate": 8.427392597422947e-06,
      "loss": 1.62383461,
      "memory(GiB)": 97.17,
      "step": 23385,
      "train_speed(iter/s)": 1.633113
    },
    {
      "acc": 0.65516863,
      "epoch": 0.5933536276002029,
      "grad_norm": 5.25,
      "learning_rate": 8.426629026648423e-06,
      "loss": 1.63778954,
      "memory(GiB)": 97.17,
      "step": 23390,
      "train_speed(iter/s)": 1.633152
    },
    {
      "acc": 0.64695163,
      "epoch": 0.593480466768138,
      "grad_norm": 6.5,
      "learning_rate": 8.425865305155455e-06,
      "loss": 1.65957737,
      "memory(GiB)": 97.17,
      "step": 23395,
      "train_speed(iter/s)": 1.633191
    },
    {
      "acc": 0.65745635,
      "epoch": 0.593607305936073,
      "grad_norm": 7.03125,
      "learning_rate": 8.425101432977636e-06,
      "loss": 1.64048576,
      "memory(GiB)": 97.17,
      "step": 23400,
      "train_speed(iter/s)": 1.63323
    },
    {
      "acc": 0.62996144,
      "epoch": 0.5937341451040081,
      "grad_norm": 5.625,
      "learning_rate": 8.424337410148562e-06,
      "loss": 1.66695919,
      "memory(GiB)": 97.17,
      "step": 23405,
      "train_speed(iter/s)": 1.633269
    },
    {
      "acc": 0.64446592,
      "epoch": 0.5938609842719432,
      "grad_norm": 5.84375,
      "learning_rate": 8.423573236701842e-06,
      "loss": 1.65895329,
      "memory(GiB)": 97.17,
      "step": 23410,
      "train_speed(iter/s)": 1.633309
    },
    {
      "acc": 0.64963346,
      "epoch": 0.5939878234398782,
      "grad_norm": 5.5625,
      "learning_rate": 8.422808912671086e-06,
      "loss": 1.66036854,
      "memory(GiB)": 97.17,
      "step": 23415,
      "train_speed(iter/s)": 1.633351
    },
    {
      "acc": 0.66884909,
      "epoch": 0.5941146626078133,
      "grad_norm": 4.84375,
      "learning_rate": 8.422044438089911e-06,
      "loss": 1.62320824,
      "memory(GiB)": 97.17,
      "step": 23420,
      "train_speed(iter/s)": 1.633386
    },
    {
      "acc": 0.65858469,
      "epoch": 0.5942415017757483,
      "grad_norm": 7.375,
      "learning_rate": 8.421279812991944e-06,
      "loss": 1.58954277,
      "memory(GiB)": 97.17,
      "step": 23425,
      "train_speed(iter/s)": 1.633426
    },
    {
      "acc": 0.65962644,
      "epoch": 0.5943683409436834,
      "grad_norm": 5.0625,
      "learning_rate": 8.420515037410817e-06,
      "loss": 1.59507351,
      "memory(GiB)": 97.17,
      "step": 23430,
      "train_speed(iter/s)": 1.633466
    },
    {
      "acc": 0.64386396,
      "epoch": 0.5944951801116185,
      "grad_norm": 5.1875,
      "learning_rate": 8.419750111380166e-06,
      "loss": 1.60042381,
      "memory(GiB)": 97.17,
      "step": 23435,
      "train_speed(iter/s)": 1.633504
    },
    {
      "acc": 0.65020213,
      "epoch": 0.5946220192795535,
      "grad_norm": 5.5,
      "learning_rate": 8.418985034933637e-06,
      "loss": 1.64524841,
      "memory(GiB)": 97.17,
      "step": 23440,
      "train_speed(iter/s)": 1.633543
    },
    {
      "acc": 0.66250248,
      "epoch": 0.5947488584474886,
      "grad_norm": 5.5,
      "learning_rate": 8.418219808104882e-06,
      "loss": 1.57158031,
      "memory(GiB)": 97.17,
      "step": 23445,
      "train_speed(iter/s)": 1.633582
    },
    {
      "acc": 0.66387591,
      "epoch": 0.5948756976154237,
      "grad_norm": 5.8125,
      "learning_rate": 8.417454430927559e-06,
      "loss": 1.60943375,
      "memory(GiB)": 97.17,
      "step": 23450,
      "train_speed(iter/s)": 1.633623
    },
    {
      "acc": 0.6533473,
      "epoch": 0.5950025367833587,
      "grad_norm": 5.59375,
      "learning_rate": 8.41668890343533e-06,
      "loss": 1.66412411,
      "memory(GiB)": 97.17,
      "step": 23455,
      "train_speed(iter/s)": 1.633662
    },
    {
      "acc": 0.64161644,
      "epoch": 0.5951293759512938,
      "grad_norm": 4.90625,
      "learning_rate": 8.41592322566187e-06,
      "loss": 1.67894611,
      "memory(GiB)": 97.17,
      "step": 23460,
      "train_speed(iter/s)": 1.633699
    },
    {
      "acc": 0.65543036,
      "epoch": 0.5952562151192288,
      "grad_norm": 5.96875,
      "learning_rate": 8.415157397640857e-06,
      "loss": 1.52906284,
      "memory(GiB)": 97.17,
      "step": 23465,
      "train_speed(iter/s)": 1.633737
    },
    {
      "acc": 0.64938507,
      "epoch": 0.5953830542871639,
      "grad_norm": 5.5,
      "learning_rate": 8.414391419405972e-06,
      "loss": 1.61706276,
      "memory(GiB)": 97.17,
      "step": 23470,
      "train_speed(iter/s)": 1.633776
    },
    {
      "acc": 0.65292044,
      "epoch": 0.595509893455099,
      "grad_norm": 6.5625,
      "learning_rate": 8.413625290990909e-06,
      "loss": 1.66275444,
      "memory(GiB)": 97.17,
      "step": 23475,
      "train_speed(iter/s)": 1.633817
    },
    {
      "acc": 0.67137198,
      "epoch": 0.595636732623034,
      "grad_norm": 5.59375,
      "learning_rate": 8.412859012429365e-06,
      "loss": 1.52991858,
      "memory(GiB)": 97.17,
      "step": 23480,
      "train_speed(iter/s)": 1.633854
    },
    {
      "acc": 0.66181145,
      "epoch": 0.5957635717909691,
      "grad_norm": 5.59375,
      "learning_rate": 8.412092583755043e-06,
      "loss": 1.57894573,
      "memory(GiB)": 97.17,
      "step": 23485,
      "train_speed(iter/s)": 1.633893
    },
    {
      "acc": 0.64901962,
      "epoch": 0.5958904109589042,
      "grad_norm": 5.4375,
      "learning_rate": 8.411326005001658e-06,
      "loss": 1.61250839,
      "memory(GiB)": 97.17,
      "step": 23490,
      "train_speed(iter/s)": 1.633934
    },
    {
      "acc": 0.64099665,
      "epoch": 0.5960172501268391,
      "grad_norm": 5.5,
      "learning_rate": 8.410559276202922e-06,
      "loss": 1.64035416,
      "memory(GiB)": 97.17,
      "step": 23495,
      "train_speed(iter/s)": 1.633971
    },
    {
      "acc": 0.64370995,
      "epoch": 0.5961440892947742,
      "grad_norm": 4.71875,
      "learning_rate": 8.409792397392565e-06,
      "loss": 1.61438026,
      "memory(GiB)": 97.17,
      "step": 23500,
      "train_speed(iter/s)": 1.634008
    },
    {
      "acc": 0.6530561,
      "epoch": 0.5962709284627092,
      "grad_norm": 6.4375,
      "learning_rate": 8.40902536860431e-06,
      "loss": 1.59955921,
      "memory(GiB)": 97.17,
      "step": 23505,
      "train_speed(iter/s)": 1.634046
    },
    {
      "acc": 0.65695734,
      "epoch": 0.5963977676306443,
      "grad_norm": 5.5625,
      "learning_rate": 8.408258189871904e-06,
      "loss": 1.58472672,
      "memory(GiB)": 97.17,
      "step": 23510,
      "train_speed(iter/s)": 1.634084
    },
    {
      "acc": 0.66635165,
      "epoch": 0.5965246067985794,
      "grad_norm": 5.15625,
      "learning_rate": 8.407490861229084e-06,
      "loss": 1.61341419,
      "memory(GiB)": 97.17,
      "step": 23515,
      "train_speed(iter/s)": 1.634123
    },
    {
      "acc": 0.64310079,
      "epoch": 0.5966514459665144,
      "grad_norm": 6.84375,
      "learning_rate": 8.406723382709603e-06,
      "loss": 1.66445427,
      "memory(GiB)": 97.17,
      "step": 23520,
      "train_speed(iter/s)": 1.634162
    },
    {
      "acc": 0.66659174,
      "epoch": 0.5967782851344495,
      "grad_norm": 5.3125,
      "learning_rate": 8.405955754347216e-06,
      "loss": 1.61462631,
      "memory(GiB)": 97.17,
      "step": 23525,
      "train_speed(iter/s)": 1.634199
    },
    {
      "acc": 0.6579628,
      "epoch": 0.5969051243023846,
      "grad_norm": 5.375,
      "learning_rate": 8.40518797617569e-06,
      "loss": 1.60541725,
      "memory(GiB)": 97.17,
      "step": 23530,
      "train_speed(iter/s)": 1.634237
    },
    {
      "acc": 0.67792006,
      "epoch": 0.5970319634703196,
      "grad_norm": 5.625,
      "learning_rate": 8.404420048228794e-06,
      "loss": 1.46336565,
      "memory(GiB)": 97.17,
      "step": 23535,
      "train_speed(iter/s)": 1.634272
    },
    {
      "acc": 0.6444315,
      "epoch": 0.5971588026382547,
      "grad_norm": 5.21875,
      "learning_rate": 8.403651970540305e-06,
      "loss": 1.65791664,
      "memory(GiB)": 97.17,
      "step": 23540,
      "train_speed(iter/s)": 1.63431
    },
    {
      "acc": 0.6475204,
      "epoch": 0.5972856418061897,
      "grad_norm": 5.71875,
      "learning_rate": 8.402883743144005e-06,
      "loss": 1.5997447,
      "memory(GiB)": 97.17,
      "step": 23545,
      "train_speed(iter/s)": 1.63435
    },
    {
      "acc": 0.6533288,
      "epoch": 0.5974124809741248,
      "grad_norm": 5.65625,
      "learning_rate": 8.402115366073686e-06,
      "loss": 1.65870399,
      "memory(GiB)": 97.17,
      "step": 23550,
      "train_speed(iter/s)": 1.63439
    },
    {
      "acc": 0.64768128,
      "epoch": 0.5975393201420599,
      "grad_norm": 5.5,
      "learning_rate": 8.401346839363143e-06,
      "loss": 1.60726089,
      "memory(GiB)": 97.17,
      "step": 23555,
      "train_speed(iter/s)": 1.634427
    },
    {
      "acc": 0.65291977,
      "epoch": 0.5976661593099949,
      "grad_norm": 6.15625,
      "learning_rate": 8.40057816304618e-06,
      "loss": 1.63162308,
      "memory(GiB)": 97.17,
      "step": 23560,
      "train_speed(iter/s)": 1.634468
    },
    {
      "acc": 0.64753909,
      "epoch": 0.59779299847793,
      "grad_norm": 6.28125,
      "learning_rate": 8.399809337156608e-06,
      "loss": 1.67615204,
      "memory(GiB)": 97.17,
      "step": 23565,
      "train_speed(iter/s)": 1.634507
    },
    {
      "acc": 0.65012894,
      "epoch": 0.5979198376458651,
      "grad_norm": 5.15625,
      "learning_rate": 8.39904036172824e-06,
      "loss": 1.61763802,
      "memory(GiB)": 97.17,
      "step": 23570,
      "train_speed(iter/s)": 1.634544
    },
    {
      "acc": 0.65089827,
      "epoch": 0.5980466768138001,
      "grad_norm": 5.3125,
      "learning_rate": 8.398271236794904e-06,
      "loss": 1.56932459,
      "memory(GiB)": 97.17,
      "step": 23575,
      "train_speed(iter/s)": 1.634582
    },
    {
      "acc": 0.64946094,
      "epoch": 0.5981735159817352,
      "grad_norm": 5.96875,
      "learning_rate": 8.397501962390427e-06,
      "loss": 1.61197071,
      "memory(GiB)": 97.17,
      "step": 23580,
      "train_speed(iter/s)": 1.63462
    },
    {
      "acc": 0.64305458,
      "epoch": 0.5983003551496702,
      "grad_norm": 5.375,
      "learning_rate": 8.396732538548642e-06,
      "loss": 1.64893017,
      "memory(GiB)": 97.17,
      "step": 23585,
      "train_speed(iter/s)": 1.634656
    },
    {
      "acc": 0.65778427,
      "epoch": 0.5984271943176053,
      "grad_norm": 5.1875,
      "learning_rate": 8.395962965303397e-06,
      "loss": 1.68108139,
      "memory(GiB)": 97.17,
      "step": 23590,
      "train_speed(iter/s)": 1.634693
    },
    {
      "acc": 0.67590961,
      "epoch": 0.5985540334855404,
      "grad_norm": 5.0,
      "learning_rate": 8.395193242688537e-06,
      "loss": 1.58161917,
      "memory(GiB)": 97.17,
      "step": 23595,
      "train_speed(iter/s)": 1.634729
    },
    {
      "acc": 0.65937762,
      "epoch": 0.5986808726534754,
      "grad_norm": 5.65625,
      "learning_rate": 8.394423370737922e-06,
      "loss": 1.53476868,
      "memory(GiB)": 97.17,
      "step": 23600,
      "train_speed(iter/s)": 1.634769
    },
    {
      "acc": 0.66086016,
      "epoch": 0.5988077118214105,
      "grad_norm": 5.3125,
      "learning_rate": 8.393653349485412e-06,
      "loss": 1.5682476,
      "memory(GiB)": 97.17,
      "step": 23605,
      "train_speed(iter/s)": 1.634806
    },
    {
      "acc": 0.65274711,
      "epoch": 0.5989345509893456,
      "grad_norm": 6.8125,
      "learning_rate": 8.392883178964874e-06,
      "loss": 1.64301605,
      "memory(GiB)": 97.17,
      "step": 23610,
      "train_speed(iter/s)": 1.634843
    },
    {
      "acc": 0.6575932,
      "epoch": 0.5990613901572805,
      "grad_norm": 6.75,
      "learning_rate": 8.392112859210186e-06,
      "loss": 1.59404621,
      "memory(GiB)": 97.17,
      "step": 23615,
      "train_speed(iter/s)": 1.634882
    },
    {
      "acc": 0.65138531,
      "epoch": 0.5991882293252156,
      "grad_norm": 5.53125,
      "learning_rate": 8.391342390255232e-06,
      "loss": 1.6398407,
      "memory(GiB)": 97.17,
      "step": 23620,
      "train_speed(iter/s)": 1.63492
    },
    {
      "acc": 0.65016856,
      "epoch": 0.5993150684931506,
      "grad_norm": 4.8125,
      "learning_rate": 8.390571772133896e-06,
      "loss": 1.62908974,
      "memory(GiB)": 97.17,
      "step": 23625,
      "train_speed(iter/s)": 1.634959
    },
    {
      "acc": 0.66987467,
      "epoch": 0.5994419076610857,
      "grad_norm": 5.84375,
      "learning_rate": 8.389801004880077e-06,
      "loss": 1.63829155,
      "memory(GiB)": 97.17,
      "step": 23630,
      "train_speed(iter/s)": 1.634996
    },
    {
      "acc": 0.64644132,
      "epoch": 0.5995687468290208,
      "grad_norm": 5.40625,
      "learning_rate": 8.389030088527675e-06,
      "loss": 1.63240185,
      "memory(GiB)": 97.17,
      "step": 23635,
      "train_speed(iter/s)": 1.635035
    },
    {
      "acc": 0.64512405,
      "epoch": 0.5996955859969558,
      "grad_norm": 6.03125,
      "learning_rate": 8.388259023110598e-06,
      "loss": 1.6434124,
      "memory(GiB)": 97.17,
      "step": 23640,
      "train_speed(iter/s)": 1.635073
    },
    {
      "acc": 0.65086021,
      "epoch": 0.5998224251648909,
      "grad_norm": 6.5,
      "learning_rate": 8.387487808662765e-06,
      "loss": 1.62117882,
      "memory(GiB)": 97.17,
      "step": 23645,
      "train_speed(iter/s)": 1.635112
    },
    {
      "acc": 0.64057388,
      "epoch": 0.599949264332826,
      "grad_norm": 5.375,
      "learning_rate": 8.38671644521809e-06,
      "loss": 1.66967812,
      "memory(GiB)": 97.17,
      "step": 23650,
      "train_speed(iter/s)": 1.635149
    },
    {
      "acc": 0.65469522,
      "epoch": 0.600076103500761,
      "grad_norm": 5.21875,
      "learning_rate": 8.385944932810508e-06,
      "loss": 1.63358097,
      "memory(GiB)": 97.17,
      "step": 23655,
      "train_speed(iter/s)": 1.635187
    },
    {
      "acc": 0.6675519,
      "epoch": 0.6002029426686961,
      "grad_norm": 5.5625,
      "learning_rate": 8.385173271473948e-06,
      "loss": 1.58354616,
      "memory(GiB)": 97.17,
      "step": 23660,
      "train_speed(iter/s)": 1.635226
    },
    {
      "acc": 0.64779515,
      "epoch": 0.6003297818366311,
      "grad_norm": 5.8125,
      "learning_rate": 8.384401461242355e-06,
      "loss": 1.67138214,
      "memory(GiB)": 97.17,
      "step": 23665,
      "train_speed(iter/s)": 1.635266
    },
    {
      "acc": 0.65686722,
      "epoch": 0.6004566210045662,
      "grad_norm": 6.40625,
      "learning_rate": 8.383629502149678e-06,
      "loss": 1.59346199,
      "memory(GiB)": 97.17,
      "step": 23670,
      "train_speed(iter/s)": 1.635304
    },
    {
      "acc": 0.65006852,
      "epoch": 0.6005834601725013,
      "grad_norm": 6.21875,
      "learning_rate": 8.382857394229865e-06,
      "loss": 1.61142731,
      "memory(GiB)": 97.17,
      "step": 23675,
      "train_speed(iter/s)": 1.635344
    },
    {
      "acc": 0.65049267,
      "epoch": 0.6007102993404363,
      "grad_norm": 5.25,
      "learning_rate": 8.382085137516883e-06,
      "loss": 1.64968128,
      "memory(GiB)": 97.17,
      "step": 23680,
      "train_speed(iter/s)": 1.63538
    },
    {
      "acc": 0.66970272,
      "epoch": 0.6008371385083714,
      "grad_norm": 5.25,
      "learning_rate": 8.381312732044696e-06,
      "loss": 1.60367165,
      "memory(GiB)": 97.17,
      "step": 23685,
      "train_speed(iter/s)": 1.635418
    },
    {
      "acc": 0.67198534,
      "epoch": 0.6009639776763065,
      "grad_norm": 6.21875,
      "learning_rate": 8.380540177847278e-06,
      "loss": 1.55842972,
      "memory(GiB)": 97.17,
      "step": 23690,
      "train_speed(iter/s)": 1.635456
    },
    {
      "acc": 0.63949499,
      "epoch": 0.6010908168442415,
      "grad_norm": 5.53125,
      "learning_rate": 8.37976747495861e-06,
      "loss": 1.68784389,
      "memory(GiB)": 97.17,
      "step": 23695,
      "train_speed(iter/s)": 1.635492
    },
    {
      "acc": 0.65733333,
      "epoch": 0.6012176560121766,
      "grad_norm": 8.375,
      "learning_rate": 8.378994623412679e-06,
      "loss": 1.54904194,
      "memory(GiB)": 97.17,
      "step": 23700,
      "train_speed(iter/s)": 1.635533
    },
    {
      "acc": 0.63535132,
      "epoch": 0.6013444951801116,
      "grad_norm": 5.8125,
      "learning_rate": 8.378221623243478e-06,
      "loss": 1.5994071,
      "memory(GiB)": 97.17,
      "step": 23705,
      "train_speed(iter/s)": 1.635571
    },
    {
      "acc": 0.65157075,
      "epoch": 0.6014713343480467,
      "grad_norm": 5.09375,
      "learning_rate": 8.377448474485008e-06,
      "loss": 1.57085342,
      "memory(GiB)": 97.17,
      "step": 23710,
      "train_speed(iter/s)": 1.63561
    },
    {
      "acc": 0.6623836,
      "epoch": 0.6015981735159818,
      "grad_norm": 4.90625,
      "learning_rate": 8.376675177171273e-06,
      "loss": 1.6030447,
      "memory(GiB)": 97.17,
      "step": 23715,
      "train_speed(iter/s)": 1.635647
    },
    {
      "acc": 0.66628366,
      "epoch": 0.6017250126839168,
      "grad_norm": 5.5625,
      "learning_rate": 8.375901731336292e-06,
      "loss": 1.566998,
      "memory(GiB)": 97.17,
      "step": 23720,
      "train_speed(iter/s)": 1.635685
    },
    {
      "acc": 0.65826263,
      "epoch": 0.6018518518518519,
      "grad_norm": 6.84375,
      "learning_rate": 8.375128137014076e-06,
      "loss": 1.5863224,
      "memory(GiB)": 97.17,
      "step": 23725,
      "train_speed(iter/s)": 1.635722
    },
    {
      "acc": 0.64467607,
      "epoch": 0.601978691019787,
      "grad_norm": 6.96875,
      "learning_rate": 8.374354394238658e-06,
      "loss": 1.7122797,
      "memory(GiB)": 97.17,
      "step": 23730,
      "train_speed(iter/s)": 1.63576
    },
    {
      "acc": 0.65857096,
      "epoch": 0.602105530187722,
      "grad_norm": 5.75,
      "learning_rate": 8.373580503044068e-06,
      "loss": 1.58004427,
      "memory(GiB)": 97.17,
      "step": 23735,
      "train_speed(iter/s)": 1.635794
    },
    {
      "acc": 0.66265678,
      "epoch": 0.602232369355657,
      "grad_norm": 7.15625,
      "learning_rate": 8.372806463464347e-06,
      "loss": 1.61926918,
      "memory(GiB)": 97.17,
      "step": 23740,
      "train_speed(iter/s)": 1.635828
    },
    {
      "acc": 0.67056332,
      "epoch": 0.602359208523592,
      "grad_norm": 5.8125,
      "learning_rate": 8.372032275533538e-06,
      "loss": 1.51040268,
      "memory(GiB)": 97.17,
      "step": 23745,
      "train_speed(iter/s)": 1.635866
    },
    {
      "acc": 0.6574666,
      "epoch": 0.6024860476915271,
      "grad_norm": 5.84375,
      "learning_rate": 8.371257939285692e-06,
      "loss": 1.63043423,
      "memory(GiB)": 97.17,
      "step": 23750,
      "train_speed(iter/s)": 1.635903
    },
    {
      "acc": 0.67122121,
      "epoch": 0.6026128868594622,
      "grad_norm": 6.15625,
      "learning_rate": 8.370483454754873e-06,
      "loss": 1.61019058,
      "memory(GiB)": 97.17,
      "step": 23755,
      "train_speed(iter/s)": 1.635941
    },
    {
      "acc": 0.66039429,
      "epoch": 0.6027397260273972,
      "grad_norm": 5.1875,
      "learning_rate": 8.369708821975144e-06,
      "loss": 1.58683586,
      "memory(GiB)": 97.17,
      "step": 23760,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.67080631,
      "epoch": 0.6028665651953323,
      "grad_norm": 5.875,
      "learning_rate": 8.368934040980576e-06,
      "loss": 1.52552214,
      "memory(GiB)": 97.17,
      "step": 23765,
      "train_speed(iter/s)": 1.636014
    },
    {
      "acc": 0.66372862,
      "epoch": 0.6029934043632674,
      "grad_norm": 6.25,
      "learning_rate": 8.368159111805246e-06,
      "loss": 1.68078499,
      "memory(GiB)": 97.17,
      "step": 23770,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.66894069,
      "epoch": 0.6031202435312024,
      "grad_norm": 5.09375,
      "learning_rate": 8.367384034483242e-06,
      "loss": 1.57246742,
      "memory(GiB)": 97.17,
      "step": 23775,
      "train_speed(iter/s)": 1.636088
    },
    {
      "acc": 0.65181723,
      "epoch": 0.6032470826991375,
      "grad_norm": 5.8125,
      "learning_rate": 8.366608809048653e-06,
      "loss": 1.65746231,
      "memory(GiB)": 97.17,
      "step": 23780,
      "train_speed(iter/s)": 1.636123
    },
    {
      "acc": 0.64440804,
      "epoch": 0.6033739218670725,
      "grad_norm": 6.75,
      "learning_rate": 8.365833435535579e-06,
      "loss": 1.61723785,
      "memory(GiB)": 97.17,
      "step": 23785,
      "train_speed(iter/s)": 1.636164
    },
    {
      "acc": 0.65853601,
      "epoch": 0.6035007610350076,
      "grad_norm": 6.09375,
      "learning_rate": 8.365057913978123e-06,
      "loss": 1.65635109,
      "memory(GiB)": 97.17,
      "step": 23790,
      "train_speed(iter/s)": 1.636199
    },
    {
      "acc": 0.6434782,
      "epoch": 0.6036276002029427,
      "grad_norm": 6.125,
      "learning_rate": 8.364282244410394e-06,
      "loss": 1.69672985,
      "memory(GiB)": 97.17,
      "step": 23795,
      "train_speed(iter/s)": 1.636237
    },
    {
      "acc": 0.65837231,
      "epoch": 0.6037544393708777,
      "grad_norm": 4.625,
      "learning_rate": 8.363506426866513e-06,
      "loss": 1.61673279,
      "memory(GiB)": 97.17,
      "step": 23800,
      "train_speed(iter/s)": 1.636275
    },
    {
      "acc": 0.65322962,
      "epoch": 0.6038812785388128,
      "grad_norm": 5.125,
      "learning_rate": 8.362730461380602e-06,
      "loss": 1.61582108,
      "memory(GiB)": 97.17,
      "step": 23805,
      "train_speed(iter/s)": 1.636314
    },
    {
      "acc": 0.65290103,
      "epoch": 0.6040081177067479,
      "grad_norm": 5.6875,
      "learning_rate": 8.361954347986793e-06,
      "loss": 1.60475922,
      "memory(GiB)": 97.17,
      "step": 23810,
      "train_speed(iter/s)": 1.636354
    },
    {
      "acc": 0.67212706,
      "epoch": 0.6041349568746829,
      "grad_norm": 4.53125,
      "learning_rate": 8.36117808671922e-06,
      "loss": 1.50592575,
      "memory(GiB)": 97.17,
      "step": 23815,
      "train_speed(iter/s)": 1.636391
    },
    {
      "acc": 0.65316696,
      "epoch": 0.604261796042618,
      "grad_norm": 5.28125,
      "learning_rate": 8.36040167761203e-06,
      "loss": 1.5869669,
      "memory(GiB)": 97.17,
      "step": 23820,
      "train_speed(iter/s)": 1.636429
    },
    {
      "acc": 0.64897199,
      "epoch": 0.604388635210553,
      "grad_norm": 6.1875,
      "learning_rate": 8.359625120699368e-06,
      "loss": 1.63080406,
      "memory(GiB)": 97.17,
      "step": 23825,
      "train_speed(iter/s)": 1.636469
    },
    {
      "acc": 0.6570981,
      "epoch": 0.6045154743784881,
      "grad_norm": 5.84375,
      "learning_rate": 8.358848416015397e-06,
      "loss": 1.56049919,
      "memory(GiB)": 97.17,
      "step": 23830,
      "train_speed(iter/s)": 1.636506
    },
    {
      "acc": 0.63806753,
      "epoch": 0.6046423135464232,
      "grad_norm": 5.0625,
      "learning_rate": 8.358071563594274e-06,
      "loss": 1.68667126,
      "memory(GiB)": 97.17,
      "step": 23835,
      "train_speed(iter/s)": 1.636543
    },
    {
      "acc": 0.64888859,
      "epoch": 0.6047691527143582,
      "grad_norm": 5.5,
      "learning_rate": 8.357294563470173e-06,
      "loss": 1.62895489,
      "memory(GiB)": 97.17,
      "step": 23840,
      "train_speed(iter/s)": 1.636583
    },
    {
      "acc": 0.66969862,
      "epoch": 0.6048959918822933,
      "grad_norm": 5.6875,
      "learning_rate": 8.356517415677267e-06,
      "loss": 1.54189816,
      "memory(GiB)": 97.17,
      "step": 23845,
      "train_speed(iter/s)": 1.63662
    },
    {
      "acc": 0.63496723,
      "epoch": 0.6050228310502284,
      "grad_norm": 7.65625,
      "learning_rate": 8.355740120249739e-06,
      "loss": 1.67266369,
      "memory(GiB)": 97.17,
      "step": 23850,
      "train_speed(iter/s)": 1.636658
    },
    {
      "acc": 0.6457356,
      "epoch": 0.6051496702181633,
      "grad_norm": 5.75,
      "learning_rate": 8.354962677221779e-06,
      "loss": 1.61528854,
      "memory(GiB)": 97.17,
      "step": 23855,
      "train_speed(iter/s)": 1.636698
    },
    {
      "acc": 0.64900188,
      "epoch": 0.6052765093860984,
      "grad_norm": 5.90625,
      "learning_rate": 8.35418508662758e-06,
      "loss": 1.59869604,
      "memory(GiB)": 97.17,
      "step": 23860,
      "train_speed(iter/s)": 1.636737
    },
    {
      "acc": 0.64458017,
      "epoch": 0.6054033485540334,
      "grad_norm": 7.09375,
      "learning_rate": 8.353407348501346e-06,
      "loss": 1.65958786,
      "memory(GiB)": 97.17,
      "step": 23865,
      "train_speed(iter/s)": 1.636776
    },
    {
      "acc": 0.65521193,
      "epoch": 0.6055301877219685,
      "grad_norm": 5.53125,
      "learning_rate": 8.352629462877286e-06,
      "loss": 1.61344986,
      "memory(GiB)": 97.17,
      "step": 23870,
      "train_speed(iter/s)": 1.636817
    },
    {
      "acc": 0.64518614,
      "epoch": 0.6056570268899036,
      "grad_norm": 6.1875,
      "learning_rate": 8.351851429789613e-06,
      "loss": 1.67500229,
      "memory(GiB)": 97.17,
      "step": 23875,
      "train_speed(iter/s)": 1.636855
    },
    {
      "acc": 0.63837938,
      "epoch": 0.6057838660578386,
      "grad_norm": 5.3125,
      "learning_rate": 8.35107324927255e-06,
      "loss": 1.62520275,
      "memory(GiB)": 97.17,
      "step": 23880,
      "train_speed(iter/s)": 1.636895
    },
    {
      "acc": 0.64832067,
      "epoch": 0.6059107052257737,
      "grad_norm": 6.40625,
      "learning_rate": 8.350294921360323e-06,
      "loss": 1.66826439,
      "memory(GiB)": 97.17,
      "step": 23885,
      "train_speed(iter/s)": 1.636936
    },
    {
      "acc": 0.64945412,
      "epoch": 0.6060375443937088,
      "grad_norm": 5.375,
      "learning_rate": 8.349516446087168e-06,
      "loss": 1.6516819,
      "memory(GiB)": 97.17,
      "step": 23890,
      "train_speed(iter/s)": 1.636973
    },
    {
      "acc": 0.63361063,
      "epoch": 0.6061643835616438,
      "grad_norm": 5.53125,
      "learning_rate": 8.348737823487325e-06,
      "loss": 1.66778126,
      "memory(GiB)": 97.17,
      "step": 23895,
      "train_speed(iter/s)": 1.637013
    },
    {
      "acc": 0.64627576,
      "epoch": 0.6062912227295789,
      "grad_norm": 5.84375,
      "learning_rate": 8.347959053595042e-06,
      "loss": 1.65390759,
      "memory(GiB)": 97.17,
      "step": 23900,
      "train_speed(iter/s)": 1.637049
    },
    {
      "acc": 0.64018984,
      "epoch": 0.6064180618975139,
      "grad_norm": 6.5,
      "learning_rate": 8.347180136444572e-06,
      "loss": 1.6810379,
      "memory(GiB)": 97.17,
      "step": 23905,
      "train_speed(iter/s)": 1.637089
    },
    {
      "acc": 0.66054482,
      "epoch": 0.606544901065449,
      "grad_norm": 5.59375,
      "learning_rate": 8.346401072070174e-06,
      "loss": 1.60496788,
      "memory(GiB)": 97.17,
      "step": 23910,
      "train_speed(iter/s)": 1.637126
    },
    {
      "acc": 0.63156519,
      "epoch": 0.6066717402333841,
      "grad_norm": 6.6875,
      "learning_rate": 8.345621860506119e-06,
      "loss": 1.68892822,
      "memory(GiB)": 97.17,
      "step": 23915,
      "train_speed(iter/s)": 1.637164
    },
    {
      "acc": 0.66439419,
      "epoch": 0.6067985794013191,
      "grad_norm": 6.125,
      "learning_rate": 8.344842501786675e-06,
      "loss": 1.52353735,
      "memory(GiB)": 97.17,
      "step": 23920,
      "train_speed(iter/s)": 1.637203
    },
    {
      "acc": 0.64234953,
      "epoch": 0.6069254185692542,
      "grad_norm": 6.15625,
      "learning_rate": 8.344062995946125e-06,
      "loss": 1.69694901,
      "memory(GiB)": 97.17,
      "step": 23925,
      "train_speed(iter/s)": 1.637241
    },
    {
      "acc": 0.6489996,
      "epoch": 0.6070522577371893,
      "grad_norm": 5.4375,
      "learning_rate": 8.343283343018755e-06,
      "loss": 1.62597313,
      "memory(GiB)": 97.17,
      "step": 23930,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.65009031,
      "epoch": 0.6071790969051243,
      "grad_norm": 5.71875,
      "learning_rate": 8.342503543038855e-06,
      "loss": 1.64044991,
      "memory(GiB)": 97.17,
      "step": 23935,
      "train_speed(iter/s)": 1.637313
    },
    {
      "acc": 0.64959722,
      "epoch": 0.6073059360730594,
      "grad_norm": 6.625,
      "learning_rate": 8.341723596040728e-06,
      "loss": 1.669664,
      "memory(GiB)": 97.17,
      "step": 23940,
      "train_speed(iter/s)": 1.637351
    },
    {
      "acc": 0.63595128,
      "epoch": 0.6074327752409944,
      "grad_norm": 5.15625,
      "learning_rate": 8.340943502058675e-06,
      "loss": 1.68558273,
      "memory(GiB)": 97.17,
      "step": 23945,
      "train_speed(iter/s)": 1.637389
    },
    {
      "acc": 0.6498745,
      "epoch": 0.6075596144089295,
      "grad_norm": 5.84375,
      "learning_rate": 8.340163261127014e-06,
      "loss": 1.66121311,
      "memory(GiB)": 97.17,
      "step": 23950,
      "train_speed(iter/s)": 1.637429
    },
    {
      "acc": 0.64318385,
      "epoch": 0.6076864535768646,
      "grad_norm": 5.46875,
      "learning_rate": 8.339382873280058e-06,
      "loss": 1.65904541,
      "memory(GiB)": 97.17,
      "step": 23955,
      "train_speed(iter/s)": 1.637465
    },
    {
      "acc": 0.6516696,
      "epoch": 0.6078132927447996,
      "grad_norm": 5.0625,
      "learning_rate": 8.338602338552136e-06,
      "loss": 1.66058064,
      "memory(GiB)": 107.26,
      "step": 23960,
      "train_speed(iter/s)": 1.637499
    },
    {
      "acc": 0.65156097,
      "epoch": 0.6079401319127347,
      "grad_norm": 5.4375,
      "learning_rate": 8.337821656977574e-06,
      "loss": 1.60544434,
      "memory(GiB)": 107.26,
      "step": 23965,
      "train_speed(iter/s)": 1.637536
    },
    {
      "acc": 0.66548567,
      "epoch": 0.6080669710806698,
      "grad_norm": 5.65625,
      "learning_rate": 8.337040828590715e-06,
      "loss": 1.588727,
      "memory(GiB)": 107.26,
      "step": 23970,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.6474226,
      "epoch": 0.6081938102486047,
      "grad_norm": 6.0,
      "learning_rate": 8.336259853425901e-06,
      "loss": 1.61458206,
      "memory(GiB)": 107.26,
      "step": 23975,
      "train_speed(iter/s)": 1.637615
    },
    {
      "acc": 0.64859209,
      "epoch": 0.6083206494165398,
      "grad_norm": 5.46875,
      "learning_rate": 8.335478731517484e-06,
      "loss": 1.60038109,
      "memory(GiB)": 107.26,
      "step": 23980,
      "train_speed(iter/s)": 1.637653
    },
    {
      "acc": 0.64348207,
      "epoch": 0.6084474885844748,
      "grad_norm": 6.15625,
      "learning_rate": 8.33469746289982e-06,
      "loss": 1.66140556,
      "memory(GiB)": 107.26,
      "step": 23985,
      "train_speed(iter/s)": 1.637689
    },
    {
      "acc": 0.63877859,
      "epoch": 0.6085743277524099,
      "grad_norm": 5.75,
      "learning_rate": 8.333916047607274e-06,
      "loss": 1.64812298,
      "memory(GiB)": 107.26,
      "step": 23990,
      "train_speed(iter/s)": 1.637728
    },
    {
      "acc": 0.66716499,
      "epoch": 0.608701166920345,
      "grad_norm": 6.5625,
      "learning_rate": 8.333134485674214e-06,
      "loss": 1.61449852,
      "memory(GiB)": 107.26,
      "step": 23995,
      "train_speed(iter/s)": 1.637765
    },
    {
      "acc": 0.64051781,
      "epoch": 0.60882800608828,
      "grad_norm": 5.53125,
      "learning_rate": 8.33235277713502e-06,
      "loss": 1.62289371,
      "memory(GiB)": 107.26,
      "step": 24000,
      "train_speed(iter/s)": 1.637803
    },
    {
      "epoch": 0.60882800608828,
      "eval_acc": 0.6446935097534336,
      "eval_loss": 1.5826815366744995,
      "eval_runtime": 57.9028,
      "eval_samples_per_second": 110.012,
      "eval_steps_per_second": 27.512,
      "step": 24000
    },
    {
      "acc": 0.65905495,
      "epoch": 0.6089548452562151,
      "grad_norm": 7.4375,
      "learning_rate": 8.33157092202407e-06,
      "loss": 1.70113144,
      "memory(GiB)": 107.26,
      "step": 24005,
      "train_speed(iter/s)": 1.630917
    },
    {
      "acc": 0.65316553,
      "epoch": 0.6090816844241502,
      "grad_norm": 4.625,
      "learning_rate": 8.33078892037576e-06,
      "loss": 1.63966274,
      "memory(GiB)": 107.26,
      "step": 24010,
      "train_speed(iter/s)": 1.630954
    },
    {
      "acc": 0.65524273,
      "epoch": 0.6092085235920852,
      "grad_norm": 6.875,
      "learning_rate": 8.33000677222448e-06,
      "loss": 1.61504612,
      "memory(GiB)": 107.26,
      "step": 24015,
      "train_speed(iter/s)": 1.630996
    },
    {
      "acc": 0.65486999,
      "epoch": 0.6093353627600203,
      "grad_norm": 4.96875,
      "learning_rate": 8.329224477604635e-06,
      "loss": 1.62060165,
      "memory(GiB)": 107.26,
      "step": 24020,
      "train_speed(iter/s)": 1.631037
    },
    {
      "acc": 0.64503832,
      "epoch": 0.6094622019279553,
      "grad_norm": 5.96875,
      "learning_rate": 8.328442036550633e-06,
      "loss": 1.64288311,
      "memory(GiB)": 107.26,
      "step": 24025,
      "train_speed(iter/s)": 1.631077
    },
    {
      "acc": 0.65088234,
      "epoch": 0.6095890410958904,
      "grad_norm": 5.65625,
      "learning_rate": 8.327659449096892e-06,
      "loss": 1.57201939,
      "memory(GiB)": 107.26,
      "step": 24030,
      "train_speed(iter/s)": 1.631116
    },
    {
      "acc": 0.67160325,
      "epoch": 0.6097158802638255,
      "grad_norm": 9.0,
      "learning_rate": 8.32687671527783e-06,
      "loss": 1.49210892,
      "memory(GiB)": 107.26,
      "step": 24035,
      "train_speed(iter/s)": 1.631159
    },
    {
      "acc": 0.6553565,
      "epoch": 0.6098427194317605,
      "grad_norm": 5.65625,
      "learning_rate": 8.326093835127878e-06,
      "loss": 1.64779854,
      "memory(GiB)": 107.26,
      "step": 24040,
      "train_speed(iter/s)": 1.6312
    },
    {
      "acc": 0.65048723,
      "epoch": 0.6099695585996956,
      "grad_norm": 7.1875,
      "learning_rate": 8.325310808681466e-06,
      "loss": 1.63453903,
      "memory(GiB)": 107.26,
      "step": 24045,
      "train_speed(iter/s)": 1.631237
    },
    {
      "acc": 0.65614166,
      "epoch": 0.6100963977676307,
      "grad_norm": 5.8125,
      "learning_rate": 8.32452763597304e-06,
      "loss": 1.53854465,
      "memory(GiB)": 107.26,
      "step": 24050,
      "train_speed(iter/s)": 1.631274
    },
    {
      "acc": 0.64905958,
      "epoch": 0.6102232369355657,
      "grad_norm": 5.71875,
      "learning_rate": 8.323744317037048e-06,
      "loss": 1.56913691,
      "memory(GiB)": 107.26,
      "step": 24055,
      "train_speed(iter/s)": 1.631313
    },
    {
      "acc": 0.64826059,
      "epoch": 0.6103500761035008,
      "grad_norm": 4.8125,
      "learning_rate": 8.322960851907937e-06,
      "loss": 1.6305357,
      "memory(GiB)": 107.26,
      "step": 24060,
      "train_speed(iter/s)": 1.631334
    },
    {
      "acc": 0.63951769,
      "epoch": 0.6104769152714358,
      "grad_norm": 7.0,
      "learning_rate": 8.322177240620175e-06,
      "loss": 1.65924416,
      "memory(GiB)": 107.26,
      "step": 24065,
      "train_speed(iter/s)": 1.631374
    },
    {
      "acc": 0.66433921,
      "epoch": 0.6106037544393709,
      "grad_norm": 5.71875,
      "learning_rate": 8.321393483208224e-06,
      "loss": 1.56772337,
      "memory(GiB)": 107.26,
      "step": 24070,
      "train_speed(iter/s)": 1.631407
    },
    {
      "acc": 0.66684809,
      "epoch": 0.610730593607306,
      "grad_norm": 5.28125,
      "learning_rate": 8.32060957970656e-06,
      "loss": 1.56584702,
      "memory(GiB)": 107.26,
      "step": 24075,
      "train_speed(iter/s)": 1.631446
    },
    {
      "acc": 0.64457984,
      "epoch": 0.610857432775241,
      "grad_norm": 5.09375,
      "learning_rate": 8.319825530149661e-06,
      "loss": 1.67626209,
      "memory(GiB)": 107.26,
      "step": 24080,
      "train_speed(iter/s)": 1.631483
    },
    {
      "acc": 0.65042267,
      "epoch": 0.6109842719431761,
      "grad_norm": 5.5,
      "learning_rate": 8.319041334572012e-06,
      "loss": 1.63317566,
      "memory(GiB)": 107.26,
      "step": 24085,
      "train_speed(iter/s)": 1.631521
    },
    {
      "acc": 0.64856639,
      "epoch": 0.6111111111111112,
      "grad_norm": 5.75,
      "learning_rate": 8.318256993008108e-06,
      "loss": 1.65145302,
      "memory(GiB)": 107.26,
      "step": 24090,
      "train_speed(iter/s)": 1.631561
    },
    {
      "acc": 0.63675413,
      "epoch": 0.6112379502790461,
      "grad_norm": 5.25,
      "learning_rate": 8.317472505492446e-06,
      "loss": 1.70657921,
      "memory(GiB)": 107.26,
      "step": 24095,
      "train_speed(iter/s)": 1.631602
    },
    {
      "acc": 0.66812935,
      "epoch": 0.6113647894469812,
      "grad_norm": 5.125,
      "learning_rate": 8.31668787205953e-06,
      "loss": 1.5039525,
      "memory(GiB)": 107.26,
      "step": 24100,
      "train_speed(iter/s)": 1.631641
    },
    {
      "acc": 0.66860933,
      "epoch": 0.6114916286149162,
      "grad_norm": 4.84375,
      "learning_rate": 8.315903092743876e-06,
      "loss": 1.59197607,
      "memory(GiB)": 107.26,
      "step": 24105,
      "train_speed(iter/s)": 1.631682
    },
    {
      "acc": 0.67009511,
      "epoch": 0.6116184677828513,
      "grad_norm": 5.5,
      "learning_rate": 8.315118167579999e-06,
      "loss": 1.57855749,
      "memory(GiB)": 107.26,
      "step": 24110,
      "train_speed(iter/s)": 1.631723
    },
    {
      "acc": 0.6609683,
      "epoch": 0.6117453069507864,
      "grad_norm": 6.09375,
      "learning_rate": 8.314333096602423e-06,
      "loss": 1.56729641,
      "memory(GiB)": 107.26,
      "step": 24115,
      "train_speed(iter/s)": 1.631763
    },
    {
      "acc": 0.65835438,
      "epoch": 0.6118721461187214,
      "grad_norm": 6.0625,
      "learning_rate": 8.313547879845682e-06,
      "loss": 1.63325996,
      "memory(GiB)": 107.26,
      "step": 24120,
      "train_speed(iter/s)": 1.631805
    },
    {
      "acc": 0.64077373,
      "epoch": 0.6119989852866565,
      "grad_norm": 7.34375,
      "learning_rate": 8.312762517344308e-06,
      "loss": 1.66759548,
      "memory(GiB)": 107.26,
      "step": 24125,
      "train_speed(iter/s)": 1.631844
    },
    {
      "acc": 0.64215155,
      "epoch": 0.6121258244545916,
      "grad_norm": 7.21875,
      "learning_rate": 8.311977009132851e-06,
      "loss": 1.64293251,
      "memory(GiB)": 107.26,
      "step": 24130,
      "train_speed(iter/s)": 1.631887
    },
    {
      "acc": 0.65572925,
      "epoch": 0.6122526636225266,
      "grad_norm": 5.40625,
      "learning_rate": 8.311191355245858e-06,
      "loss": 1.64922485,
      "memory(GiB)": 107.26,
      "step": 24135,
      "train_speed(iter/s)": 1.631928
    },
    {
      "acc": 0.66276288,
      "epoch": 0.6123795027904617,
      "grad_norm": 4.75,
      "learning_rate": 8.310405555717884e-06,
      "loss": 1.62353764,
      "memory(GiB)": 107.26,
      "step": 24140,
      "train_speed(iter/s)": 1.631968
    },
    {
      "acc": 0.6509696,
      "epoch": 0.6125063419583967,
      "grad_norm": 5.1875,
      "learning_rate": 8.309619610583495e-06,
      "loss": 1.66010208,
      "memory(GiB)": 107.26,
      "step": 24145,
      "train_speed(iter/s)": 1.632005
    },
    {
      "acc": 0.68065972,
      "epoch": 0.6126331811263318,
      "grad_norm": 5.5,
      "learning_rate": 8.30883351987726e-06,
      "loss": 1.57454548,
      "memory(GiB)": 107.26,
      "step": 24150,
      "train_speed(iter/s)": 1.632043
    },
    {
      "acc": 0.65138063,
      "epoch": 0.6127600202942669,
      "grad_norm": 8.5625,
      "learning_rate": 8.30804728363375e-06,
      "loss": 1.63922977,
      "memory(GiB)": 107.26,
      "step": 24155,
      "train_speed(iter/s)": 1.632082
    },
    {
      "acc": 0.66233044,
      "epoch": 0.6128868594622019,
      "grad_norm": 7.03125,
      "learning_rate": 8.307260901887556e-06,
      "loss": 1.59522343,
      "memory(GiB)": 107.26,
      "step": 24160,
      "train_speed(iter/s)": 1.632122
    },
    {
      "acc": 0.65686007,
      "epoch": 0.613013698630137,
      "grad_norm": 5.125,
      "learning_rate": 8.306474374673259e-06,
      "loss": 1.54250326,
      "memory(GiB)": 107.26,
      "step": 24165,
      "train_speed(iter/s)": 1.632161
    },
    {
      "acc": 0.66284804,
      "epoch": 0.6131405377980721,
      "grad_norm": 6.75,
      "learning_rate": 8.305687702025457e-06,
      "loss": 1.59297419,
      "memory(GiB)": 107.26,
      "step": 24170,
      "train_speed(iter/s)": 1.6322
    },
    {
      "acc": 0.6563828,
      "epoch": 0.6132673769660071,
      "grad_norm": 5.84375,
      "learning_rate": 8.304900883978753e-06,
      "loss": 1.58604946,
      "memory(GiB)": 107.26,
      "step": 24175,
      "train_speed(iter/s)": 1.63224
    },
    {
      "acc": 0.6420083,
      "epoch": 0.6133942161339422,
      "grad_norm": 5.21875,
      "learning_rate": 8.304113920567751e-06,
      "loss": 1.67757549,
      "memory(GiB)": 107.26,
      "step": 24180,
      "train_speed(iter/s)": 1.632279
    },
    {
      "acc": 0.65166378,
      "epoch": 0.6135210553018772,
      "grad_norm": 5.78125,
      "learning_rate": 8.303326811827066e-06,
      "loss": 1.61237316,
      "memory(GiB)": 107.26,
      "step": 24185,
      "train_speed(iter/s)": 1.632318
    },
    {
      "acc": 0.65409741,
      "epoch": 0.6136478944698123,
      "grad_norm": 5.28125,
      "learning_rate": 8.302539557791322e-06,
      "loss": 1.63222599,
      "memory(GiB)": 107.26,
      "step": 24190,
      "train_speed(iter/s)": 1.632356
    },
    {
      "acc": 0.63583212,
      "epoch": 0.6137747336377474,
      "grad_norm": 4.96875,
      "learning_rate": 8.301752158495141e-06,
      "loss": 1.66746311,
      "memory(GiB)": 107.26,
      "step": 24195,
      "train_speed(iter/s)": 1.632393
    },
    {
      "acc": 0.65150318,
      "epoch": 0.6139015728056824,
      "grad_norm": 5.96875,
      "learning_rate": 8.300964613973159e-06,
      "loss": 1.65224075,
      "memory(GiB)": 107.26,
      "step": 24200,
      "train_speed(iter/s)": 1.632433
    },
    {
      "acc": 0.64884634,
      "epoch": 0.6140284119736175,
      "grad_norm": 5.375,
      "learning_rate": 8.300176924260017e-06,
      "loss": 1.60562305,
      "memory(GiB)": 107.26,
      "step": 24205,
      "train_speed(iter/s)": 1.63247
    },
    {
      "acc": 0.65094285,
      "epoch": 0.6141552511415526,
      "grad_norm": 6.25,
      "learning_rate": 8.299389089390359e-06,
      "loss": 1.59919872,
      "memory(GiB)": 107.26,
      "step": 24210,
      "train_speed(iter/s)": 1.632511
    },
    {
      "acc": 0.66140113,
      "epoch": 0.6142820903094875,
      "grad_norm": 4.875,
      "learning_rate": 8.298601109398838e-06,
      "loss": 1.57670517,
      "memory(GiB)": 107.26,
      "step": 24215,
      "train_speed(iter/s)": 1.63255
    },
    {
      "acc": 0.65009112,
      "epoch": 0.6144089294774226,
      "grad_norm": 4.46875,
      "learning_rate": 8.297812984320113e-06,
      "loss": 1.63598061,
      "memory(GiB)": 107.26,
      "step": 24220,
      "train_speed(iter/s)": 1.632588
    },
    {
      "acc": 0.64870777,
      "epoch": 0.6145357686453576,
      "grad_norm": 5.28125,
      "learning_rate": 8.297024714188851e-06,
      "loss": 1.61085854,
      "memory(GiB)": 107.26,
      "step": 24225,
      "train_speed(iter/s)": 1.632629
    },
    {
      "acc": 0.6540554,
      "epoch": 0.6146626078132927,
      "grad_norm": 6.375,
      "learning_rate": 8.296236299039719e-06,
      "loss": 1.65814362,
      "memory(GiB)": 107.26,
      "step": 24230,
      "train_speed(iter/s)": 1.63267
    },
    {
      "acc": 0.65093565,
      "epoch": 0.6147894469812278,
      "grad_norm": 5.125,
      "learning_rate": 8.295447738907401e-06,
      "loss": 1.54922571,
      "memory(GiB)": 107.26,
      "step": 24235,
      "train_speed(iter/s)": 1.632709
    },
    {
      "acc": 0.65365977,
      "epoch": 0.6149162861491628,
      "grad_norm": 5.125,
      "learning_rate": 8.294659033826576e-06,
      "loss": 1.61292439,
      "memory(GiB)": 107.26,
      "step": 24240,
      "train_speed(iter/s)": 1.632749
    },
    {
      "acc": 0.64184256,
      "epoch": 0.6150431253170979,
      "grad_norm": 6.78125,
      "learning_rate": 8.293870183831937e-06,
      "loss": 1.64473724,
      "memory(GiB)": 107.26,
      "step": 24245,
      "train_speed(iter/s)": 1.632788
    },
    {
      "acc": 0.65789547,
      "epoch": 0.615169964485033,
      "grad_norm": 6.375,
      "learning_rate": 8.293081188958183e-06,
      "loss": 1.57331505,
      "memory(GiB)": 107.26,
      "step": 24250,
      "train_speed(iter/s)": 1.632827
    },
    {
      "acc": 0.66379347,
      "epoch": 0.615296803652968,
      "grad_norm": 5.59375,
      "learning_rate": 8.292292049240014e-06,
      "loss": 1.57434521,
      "memory(GiB)": 107.26,
      "step": 24255,
      "train_speed(iter/s)": 1.632866
    },
    {
      "acc": 0.64103479,
      "epoch": 0.6154236428209031,
      "grad_norm": 5.53125,
      "learning_rate": 8.291502764712143e-06,
      "loss": 1.66669807,
      "memory(GiB)": 107.26,
      "step": 24260,
      "train_speed(iter/s)": 1.632907
    },
    {
      "acc": 0.66064053,
      "epoch": 0.6155504819888381,
      "grad_norm": 5.4375,
      "learning_rate": 8.290713335409284e-06,
      "loss": 1.58635864,
      "memory(GiB)": 107.26,
      "step": 24265,
      "train_speed(iter/s)": 1.632946
    },
    {
      "acc": 0.6556314,
      "epoch": 0.6156773211567732,
      "grad_norm": 5.5,
      "learning_rate": 8.28992376136616e-06,
      "loss": 1.58295794,
      "memory(GiB)": 107.26,
      "step": 24270,
      "train_speed(iter/s)": 1.632984
    },
    {
      "acc": 0.64795151,
      "epoch": 0.6158041603247083,
      "grad_norm": 5.5,
      "learning_rate": 8.289134042617502e-06,
      "loss": 1.67456646,
      "memory(GiB)": 107.26,
      "step": 24275,
      "train_speed(iter/s)": 1.633024
    },
    {
      "acc": 0.65625167,
      "epoch": 0.6159309994926433,
      "grad_norm": 5.34375,
      "learning_rate": 8.288344179198043e-06,
      "loss": 1.58869667,
      "memory(GiB)": 107.26,
      "step": 24280,
      "train_speed(iter/s)": 1.633062
    },
    {
      "acc": 0.6535018,
      "epoch": 0.6160578386605784,
      "grad_norm": 5.59375,
      "learning_rate": 8.287554171142525e-06,
      "loss": 1.65524254,
      "memory(GiB)": 107.26,
      "step": 24285,
      "train_speed(iter/s)": 1.633101
    },
    {
      "acc": 0.6364502,
      "epoch": 0.6161846778285135,
      "grad_norm": 5.15625,
      "learning_rate": 8.2867640184857e-06,
      "loss": 1.68619289,
      "memory(GiB)": 107.26,
      "step": 24290,
      "train_speed(iter/s)": 1.63314
    },
    {
      "acc": 0.63597555,
      "epoch": 0.6163115169964485,
      "grad_norm": 5.8125,
      "learning_rate": 8.285973721262315e-06,
      "loss": 1.65202217,
      "memory(GiB)": 107.26,
      "step": 24295,
      "train_speed(iter/s)": 1.63318
    },
    {
      "acc": 0.6548295,
      "epoch": 0.6164383561643836,
      "grad_norm": 5.15625,
      "learning_rate": 8.285183279507135e-06,
      "loss": 1.59142675,
      "memory(GiB)": 107.26,
      "step": 24300,
      "train_speed(iter/s)": 1.633217
    },
    {
      "acc": 0.64779425,
      "epoch": 0.6165651953323186,
      "grad_norm": 9.375,
      "learning_rate": 8.28439269325493e-06,
      "loss": 1.60534821,
      "memory(GiB)": 107.26,
      "step": 24305,
      "train_speed(iter/s)": 1.633255
    },
    {
      "acc": 0.64788494,
      "epoch": 0.6166920345002537,
      "grad_norm": 6.125,
      "learning_rate": 8.28360196254047e-06,
      "loss": 1.68173389,
      "memory(GiB)": 107.26,
      "step": 24310,
      "train_speed(iter/s)": 1.633292
    },
    {
      "acc": 0.65048113,
      "epoch": 0.6168188736681888,
      "grad_norm": 6.28125,
      "learning_rate": 8.282811087398535e-06,
      "loss": 1.62319794,
      "memory(GiB)": 107.26,
      "step": 24315,
      "train_speed(iter/s)": 1.633334
    },
    {
      "acc": 0.6476831,
      "epoch": 0.6169457128361238,
      "grad_norm": 6.625,
      "learning_rate": 8.282020067863911e-06,
      "loss": 1.62564278,
      "memory(GiB)": 107.26,
      "step": 24320,
      "train_speed(iter/s)": 1.633373
    },
    {
      "acc": 0.67075973,
      "epoch": 0.6170725520040589,
      "grad_norm": 5.84375,
      "learning_rate": 8.281228903971391e-06,
      "loss": 1.56481209,
      "memory(GiB)": 107.26,
      "step": 24325,
      "train_speed(iter/s)": 1.633412
    },
    {
      "acc": 0.66688604,
      "epoch": 0.617199391171994,
      "grad_norm": 7.21875,
      "learning_rate": 8.280437595755774e-06,
      "loss": 1.52157927,
      "memory(GiB)": 107.26,
      "step": 24330,
      "train_speed(iter/s)": 1.633451
    },
    {
      "acc": 0.65221734,
      "epoch": 0.617326230339929,
      "grad_norm": 5.28125,
      "learning_rate": 8.279646143251867e-06,
      "loss": 1.53978062,
      "memory(GiB)": 107.26,
      "step": 24335,
      "train_speed(iter/s)": 1.63349
    },
    {
      "acc": 0.65937424,
      "epoch": 0.617453069507864,
      "grad_norm": 5.6875,
      "learning_rate": 8.278854546494479e-06,
      "loss": 1.61242599,
      "memory(GiB)": 107.26,
      "step": 24340,
      "train_speed(iter/s)": 1.633528
    },
    {
      "acc": 0.65502834,
      "epoch": 0.617579908675799,
      "grad_norm": 5.8125,
      "learning_rate": 8.27806280551843e-06,
      "loss": 1.56261711,
      "memory(GiB)": 107.26,
      "step": 24345,
      "train_speed(iter/s)": 1.633567
    },
    {
      "acc": 0.64649534,
      "epoch": 0.6177067478437341,
      "grad_norm": 5.71875,
      "learning_rate": 8.277270920358542e-06,
      "loss": 1.60185318,
      "memory(GiB)": 107.26,
      "step": 24350,
      "train_speed(iter/s)": 1.633606
    },
    {
      "acc": 0.67093916,
      "epoch": 0.6178335870116692,
      "grad_norm": 5.90625,
      "learning_rate": 8.276478891049649e-06,
      "loss": 1.59160185,
      "memory(GiB)": 107.26,
      "step": 24355,
      "train_speed(iter/s)": 1.633647
    },
    {
      "acc": 0.66452913,
      "epoch": 0.6179604261796042,
      "grad_norm": 5.8125,
      "learning_rate": 8.275686717626584e-06,
      "loss": 1.56573763,
      "memory(GiB)": 107.26,
      "step": 24360,
      "train_speed(iter/s)": 1.633684
    },
    {
      "acc": 0.6515511,
      "epoch": 0.6180872653475393,
      "grad_norm": 4.5625,
      "learning_rate": 8.274894400124191e-06,
      "loss": 1.60488739,
      "memory(GiB)": 107.26,
      "step": 24365,
      "train_speed(iter/s)": 1.633722
    },
    {
      "acc": 0.65088544,
      "epoch": 0.6182141045154744,
      "grad_norm": 5.375,
      "learning_rate": 8.274101938577324e-06,
      "loss": 1.64141998,
      "memory(GiB)": 107.26,
      "step": 24370,
      "train_speed(iter/s)": 1.633761
    },
    {
      "acc": 0.65597076,
      "epoch": 0.6183409436834094,
      "grad_norm": 7.0,
      "learning_rate": 8.273309333020834e-06,
      "loss": 1.56200743,
      "memory(GiB)": 107.26,
      "step": 24375,
      "train_speed(iter/s)": 1.633801
    },
    {
      "acc": 0.65608997,
      "epoch": 0.6184677828513445,
      "grad_norm": 4.96875,
      "learning_rate": 8.272516583489587e-06,
      "loss": 1.64275074,
      "memory(GiB)": 107.26,
      "step": 24380,
      "train_speed(iter/s)": 1.633841
    },
    {
      "acc": 0.6670053,
      "epoch": 0.6185946220192795,
      "grad_norm": 6.0625,
      "learning_rate": 8.271723690018448e-06,
      "loss": 1.55916777,
      "memory(GiB)": 107.26,
      "step": 24385,
      "train_speed(iter/s)": 1.633881
    },
    {
      "acc": 0.63828945,
      "epoch": 0.6187214611872146,
      "grad_norm": 5.3125,
      "learning_rate": 8.270930652642295e-06,
      "loss": 1.61743279,
      "memory(GiB)": 107.26,
      "step": 24390,
      "train_speed(iter/s)": 1.633919
    },
    {
      "acc": 0.64620543,
      "epoch": 0.6188483003551497,
      "grad_norm": 5.53125,
      "learning_rate": 8.270137471396007e-06,
      "loss": 1.6379776,
      "memory(GiB)": 107.26,
      "step": 24395,
      "train_speed(iter/s)": 1.633957
    },
    {
      "acc": 0.6568716,
      "epoch": 0.6189751395230847,
      "grad_norm": 4.71875,
      "learning_rate": 8.269344146314475e-06,
      "loss": 1.60519867,
      "memory(GiB)": 107.26,
      "step": 24400,
      "train_speed(iter/s)": 1.633994
    },
    {
      "acc": 0.65031061,
      "epoch": 0.6191019786910198,
      "grad_norm": 5.90625,
      "learning_rate": 8.26855067743259e-06,
      "loss": 1.68257332,
      "memory(GiB)": 107.26,
      "step": 24405,
      "train_speed(iter/s)": 1.634033
    },
    {
      "acc": 0.65250025,
      "epoch": 0.6192288178589549,
      "grad_norm": 5.625,
      "learning_rate": 8.267757064785254e-06,
      "loss": 1.59980431,
      "memory(GiB)": 107.26,
      "step": 24410,
      "train_speed(iter/s)": 1.63407
    },
    {
      "acc": 0.65542336,
      "epoch": 0.6193556570268899,
      "grad_norm": 6.0,
      "learning_rate": 8.26696330840737e-06,
      "loss": 1.61762924,
      "memory(GiB)": 107.26,
      "step": 24415,
      "train_speed(iter/s)": 1.634108
    },
    {
      "acc": 0.6726089,
      "epoch": 0.619482496194825,
      "grad_norm": 5.5625,
      "learning_rate": 8.266169408333856e-06,
      "loss": 1.57673302,
      "memory(GiB)": 107.26,
      "step": 24420,
      "train_speed(iter/s)": 1.634144
    },
    {
      "acc": 0.66403751,
      "epoch": 0.61960933536276,
      "grad_norm": 5.625,
      "learning_rate": 8.265375364599629e-06,
      "loss": 1.60204811,
      "memory(GiB)": 107.26,
      "step": 24425,
      "train_speed(iter/s)": 1.634186
    },
    {
      "acc": 0.66187143,
      "epoch": 0.6197361745306951,
      "grad_norm": 6.625,
      "learning_rate": 8.264581177239615e-06,
      "loss": 1.60062084,
      "memory(GiB)": 107.26,
      "step": 24430,
      "train_speed(iter/s)": 1.634223
    },
    {
      "acc": 0.66657109,
      "epoch": 0.6198630136986302,
      "grad_norm": 5.78125,
      "learning_rate": 8.263786846288745e-06,
      "loss": 1.5310668,
      "memory(GiB)": 107.26,
      "step": 24435,
      "train_speed(iter/s)": 1.634263
    },
    {
      "acc": 0.66046629,
      "epoch": 0.6199898528665652,
      "grad_norm": 6.71875,
      "learning_rate": 8.262992371781956e-06,
      "loss": 1.6019474,
      "memory(GiB)": 107.26,
      "step": 24440,
      "train_speed(iter/s)": 1.6343
    },
    {
      "acc": 0.63187566,
      "epoch": 0.6201166920345003,
      "grad_norm": 5.6875,
      "learning_rate": 8.262197753754195e-06,
      "loss": 1.68738117,
      "memory(GiB)": 107.26,
      "step": 24445,
      "train_speed(iter/s)": 1.634339
    },
    {
      "acc": 0.65081878,
      "epoch": 0.6202435312024354,
      "grad_norm": 5.90625,
      "learning_rate": 8.261402992240414e-06,
      "loss": 1.67191315,
      "memory(GiB)": 107.26,
      "step": 24450,
      "train_speed(iter/s)": 1.634376
    },
    {
      "acc": 0.64992442,
      "epoch": 0.6203703703703703,
      "grad_norm": 6.375,
      "learning_rate": 8.260608087275566e-06,
      "loss": 1.65810432,
      "memory(GiB)": 107.26,
      "step": 24455,
      "train_speed(iter/s)": 1.634415
    },
    {
      "acc": 0.65987325,
      "epoch": 0.6204972095383054,
      "grad_norm": 4.875,
      "learning_rate": 8.259813038894617e-06,
      "loss": 1.60712872,
      "memory(GiB)": 107.26,
      "step": 24460,
      "train_speed(iter/s)": 1.634452
    },
    {
      "acc": 0.65861382,
      "epoch": 0.6206240487062404,
      "grad_norm": 5.59375,
      "learning_rate": 8.259017847132538e-06,
      "loss": 1.57515965,
      "memory(GiB)": 107.26,
      "step": 24465,
      "train_speed(iter/s)": 1.634488
    },
    {
      "acc": 0.6521121,
      "epoch": 0.6207508878741755,
      "grad_norm": 5.9375,
      "learning_rate": 8.258222512024303e-06,
      "loss": 1.64870644,
      "memory(GiB)": 107.26,
      "step": 24470,
      "train_speed(iter/s)": 1.634527
    },
    {
      "acc": 0.65789003,
      "epoch": 0.6208777270421106,
      "grad_norm": 5.25,
      "learning_rate": 8.257427033604894e-06,
      "loss": 1.54153767,
      "memory(GiB)": 107.26,
      "step": 24475,
      "train_speed(iter/s)": 1.634567
    },
    {
      "acc": 0.63943577,
      "epoch": 0.6210045662100456,
      "grad_norm": 6.0,
      "learning_rate": 8.256631411909305e-06,
      "loss": 1.68126507,
      "memory(GiB)": 107.26,
      "step": 24480,
      "train_speed(iter/s)": 1.634602
    },
    {
      "acc": 0.65381846,
      "epoch": 0.6211314053779807,
      "grad_norm": 5.625,
      "learning_rate": 8.25583564697252e-06,
      "loss": 1.54730148,
      "memory(GiB)": 107.26,
      "step": 24485,
      "train_speed(iter/s)": 1.634641
    },
    {
      "acc": 0.65746574,
      "epoch": 0.6212582445459158,
      "grad_norm": 5.53125,
      "learning_rate": 8.255039738829552e-06,
      "loss": 1.62740364,
      "memory(GiB)": 107.26,
      "step": 24490,
      "train_speed(iter/s)": 1.63468
    },
    {
      "acc": 0.65461531,
      "epoch": 0.6213850837138508,
      "grad_norm": 5.78125,
      "learning_rate": 8.254243687515402e-06,
      "loss": 1.61341038,
      "memory(GiB)": 107.26,
      "step": 24495,
      "train_speed(iter/s)": 1.634716
    },
    {
      "acc": 0.65132909,
      "epoch": 0.6215119228817859,
      "grad_norm": 5.65625,
      "learning_rate": 8.253447493065085e-06,
      "loss": 1.60907631,
      "memory(GiB)": 107.26,
      "step": 24500,
      "train_speed(iter/s)": 1.634755
    },
    {
      "acc": 0.64845624,
      "epoch": 0.6216387620497209,
      "grad_norm": 5.65625,
      "learning_rate": 8.252651155513622e-06,
      "loss": 1.64631405,
      "memory(GiB)": 107.26,
      "step": 24505,
      "train_speed(iter/s)": 1.634794
    },
    {
      "acc": 0.63941102,
      "epoch": 0.621765601217656,
      "grad_norm": 5.8125,
      "learning_rate": 8.251854674896039e-06,
      "loss": 1.70469093,
      "memory(GiB)": 107.26,
      "step": 24510,
      "train_speed(iter/s)": 1.634829
    },
    {
      "acc": 0.64546967,
      "epoch": 0.6218924403855911,
      "grad_norm": 5.59375,
      "learning_rate": 8.251058051247368e-06,
      "loss": 1.63095169,
      "memory(GiB)": 107.26,
      "step": 24515,
      "train_speed(iter/s)": 1.634867
    },
    {
      "acc": 0.6682179,
      "epoch": 0.6220192795535261,
      "grad_norm": 5.75,
      "learning_rate": 8.250261284602651e-06,
      "loss": 1.58506727,
      "memory(GiB)": 107.26,
      "step": 24520,
      "train_speed(iter/s)": 1.634907
    },
    {
      "acc": 0.66320562,
      "epoch": 0.6221461187214612,
      "grad_norm": 6.1875,
      "learning_rate": 8.249464374996932e-06,
      "loss": 1.60664139,
      "memory(GiB)": 107.26,
      "step": 24525,
      "train_speed(iter/s)": 1.634945
    },
    {
      "acc": 0.65261164,
      "epoch": 0.6222729578893963,
      "grad_norm": 5.53125,
      "learning_rate": 8.24866732246526e-06,
      "loss": 1.66367607,
      "memory(GiB)": 107.26,
      "step": 24530,
      "train_speed(iter/s)": 1.634984
    },
    {
      "acc": 0.64880095,
      "epoch": 0.6223997970573313,
      "grad_norm": 6.65625,
      "learning_rate": 8.247870127042695e-06,
      "loss": 1.61308403,
      "memory(GiB)": 107.26,
      "step": 24535,
      "train_speed(iter/s)": 1.635021
    },
    {
      "acc": 0.64007292,
      "epoch": 0.6225266362252664,
      "grad_norm": 6.5,
      "learning_rate": 8.247072788764302e-06,
      "loss": 1.63403111,
      "memory(GiB)": 107.26,
      "step": 24540,
      "train_speed(iter/s)": 1.635059
    },
    {
      "acc": 0.64348264,
      "epoch": 0.6226534753932014,
      "grad_norm": 5.71875,
      "learning_rate": 8.246275307665147e-06,
      "loss": 1.66267395,
      "memory(GiB)": 107.26,
      "step": 24545,
      "train_speed(iter/s)": 1.635096
    },
    {
      "acc": 0.66274967,
      "epoch": 0.6227803145611365,
      "grad_norm": 6.28125,
      "learning_rate": 8.245477683780316e-06,
      "loss": 1.54251146,
      "memory(GiB)": 107.26,
      "step": 24550,
      "train_speed(iter/s)": 1.635132
    },
    {
      "acc": 0.6593421,
      "epoch": 0.6229071537290716,
      "grad_norm": 6.0,
      "learning_rate": 8.244679917144883e-06,
      "loss": 1.61581612,
      "memory(GiB)": 107.26,
      "step": 24555,
      "train_speed(iter/s)": 1.635168
    },
    {
      "acc": 0.63921938,
      "epoch": 0.6230339928970066,
      "grad_norm": 6.4375,
      "learning_rate": 8.243882007793941e-06,
      "loss": 1.6940382,
      "memory(GiB)": 107.26,
      "step": 24560,
      "train_speed(iter/s)": 1.635205
    },
    {
      "acc": 0.65796113,
      "epoch": 0.6231608320649417,
      "grad_norm": 6.21875,
      "learning_rate": 8.243083955762588e-06,
      "loss": 1.60132904,
      "memory(GiB)": 107.26,
      "step": 24565,
      "train_speed(iter/s)": 1.635239
    },
    {
      "acc": 0.66274385,
      "epoch": 0.6232876712328768,
      "grad_norm": 6.0625,
      "learning_rate": 8.24228576108592e-06,
      "loss": 1.60198097,
      "memory(GiB)": 107.26,
      "step": 24570,
      "train_speed(iter/s)": 1.635278
    },
    {
      "acc": 0.65700483,
      "epoch": 0.6234145104008117,
      "grad_norm": 6.875,
      "learning_rate": 8.24148742379905e-06,
      "loss": 1.54407921,
      "memory(GiB)": 107.26,
      "step": 24575,
      "train_speed(iter/s)": 1.635316
    },
    {
      "acc": 0.65609951,
      "epoch": 0.6235413495687468,
      "grad_norm": 6.0,
      "learning_rate": 8.240688943937092e-06,
      "loss": 1.60369186,
      "memory(GiB)": 107.26,
      "step": 24580,
      "train_speed(iter/s)": 1.635353
    },
    {
      "acc": 0.65588841,
      "epoch": 0.6236681887366818,
      "grad_norm": 6.125,
      "learning_rate": 8.239890321535163e-06,
      "loss": 1.61838551,
      "memory(GiB)": 107.26,
      "step": 24585,
      "train_speed(iter/s)": 1.63539
    },
    {
      "acc": 0.65276937,
      "epoch": 0.6237950279046169,
      "grad_norm": 5.78125,
      "learning_rate": 8.239091556628395e-06,
      "loss": 1.66119251,
      "memory(GiB)": 107.26,
      "step": 24590,
      "train_speed(iter/s)": 1.635429
    },
    {
      "acc": 0.65116787,
      "epoch": 0.623921867072552,
      "grad_norm": 5.6875,
      "learning_rate": 8.238292649251918e-06,
      "loss": 1.64313469,
      "memory(GiB)": 107.26,
      "step": 24595,
      "train_speed(iter/s)": 1.635466
    },
    {
      "acc": 0.67093649,
      "epoch": 0.624048706240487,
      "grad_norm": 5.375,
      "learning_rate": 8.237493599440871e-06,
      "loss": 1.60220451,
      "memory(GiB)": 107.26,
      "step": 24600,
      "train_speed(iter/s)": 1.635503
    },
    {
      "acc": 0.63234806,
      "epoch": 0.6241755454084221,
      "grad_norm": 5.6875,
      "learning_rate": 8.236694407230402e-06,
      "loss": 1.74058857,
      "memory(GiB)": 107.26,
      "step": 24605,
      "train_speed(iter/s)": 1.635539
    },
    {
      "acc": 0.66838779,
      "epoch": 0.6243023845763572,
      "grad_norm": 5.84375,
      "learning_rate": 8.235895072655664e-06,
      "loss": 1.4857338,
      "memory(GiB)": 107.26,
      "step": 24610,
      "train_speed(iter/s)": 1.635572
    },
    {
      "acc": 0.6534894,
      "epoch": 0.6244292237442922,
      "grad_norm": 6.0,
      "learning_rate": 8.235095595751809e-06,
      "loss": 1.62717571,
      "memory(GiB)": 107.26,
      "step": 24615,
      "train_speed(iter/s)": 1.635609
    },
    {
      "acc": 0.66915188,
      "epoch": 0.6245560629122273,
      "grad_norm": 5.78125,
      "learning_rate": 8.23429597655401e-06,
      "loss": 1.58127661,
      "memory(GiB)": 107.26,
      "step": 24620,
      "train_speed(iter/s)": 1.635646
    },
    {
      "acc": 0.63129644,
      "epoch": 0.6246829020801623,
      "grad_norm": 5.875,
      "learning_rate": 8.233496215097433e-06,
      "loss": 1.67797775,
      "memory(GiB)": 107.26,
      "step": 24625,
      "train_speed(iter/s)": 1.635685
    },
    {
      "acc": 0.65792122,
      "epoch": 0.6248097412480974,
      "grad_norm": 5.46875,
      "learning_rate": 8.232696311417256e-06,
      "loss": 1.54957199,
      "memory(GiB)": 107.26,
      "step": 24630,
      "train_speed(iter/s)": 1.635723
    },
    {
      "acc": 0.64836216,
      "epoch": 0.6249365804160325,
      "grad_norm": 7.1875,
      "learning_rate": 8.231896265548662e-06,
      "loss": 1.66277733,
      "memory(GiB)": 107.26,
      "step": 24635,
      "train_speed(iter/s)": 1.635762
    },
    {
      "acc": 0.6531055,
      "epoch": 0.6250634195839675,
      "grad_norm": 5.3125,
      "learning_rate": 8.231096077526841e-06,
      "loss": 1.64894123,
      "memory(GiB)": 107.26,
      "step": 24640,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.6701932,
      "epoch": 0.6251902587519026,
      "grad_norm": 5.59375,
      "learning_rate": 8.230295747386988e-06,
      "loss": 1.56419325,
      "memory(GiB)": 107.26,
      "step": 24645,
      "train_speed(iter/s)": 1.635836
    },
    {
      "acc": 0.66264434,
      "epoch": 0.6253170979198377,
      "grad_norm": 6.0625,
      "learning_rate": 8.229495275164307e-06,
      "loss": 1.52827473,
      "memory(GiB)": 107.26,
      "step": 24650,
      "train_speed(iter/s)": 1.635875
    },
    {
      "acc": 0.66026759,
      "epoch": 0.6254439370877727,
      "grad_norm": 6.1875,
      "learning_rate": 8.228694660894003e-06,
      "loss": 1.65783253,
      "memory(GiB)": 107.26,
      "step": 24655,
      "train_speed(iter/s)": 1.635912
    },
    {
      "acc": 0.66281977,
      "epoch": 0.6255707762557078,
      "grad_norm": 7.9375,
      "learning_rate": 8.227893904611295e-06,
      "loss": 1.58205214,
      "memory(GiB)": 107.26,
      "step": 24660,
      "train_speed(iter/s)": 1.63595
    },
    {
      "acc": 0.66056638,
      "epoch": 0.6256976154236428,
      "grad_norm": 5.0,
      "learning_rate": 8.2270930063514e-06,
      "loss": 1.58064556,
      "memory(GiB)": 107.26,
      "step": 24665,
      "train_speed(iter/s)": 1.635988
    },
    {
      "acc": 0.65215459,
      "epoch": 0.6258244545915779,
      "grad_norm": 5.5,
      "learning_rate": 8.226291966149549e-06,
      "loss": 1.66009369,
      "memory(GiB)": 107.26,
      "step": 24670,
      "train_speed(iter/s)": 1.636026
    },
    {
      "acc": 0.65944614,
      "epoch": 0.625951293759513,
      "grad_norm": 5.40625,
      "learning_rate": 8.225490784040971e-06,
      "loss": 1.57912216,
      "memory(GiB)": 107.26,
      "step": 24675,
      "train_speed(iter/s)": 1.636064
    },
    {
      "acc": 0.65118546,
      "epoch": 0.626078132927448,
      "grad_norm": 6.34375,
      "learning_rate": 8.224689460060908e-06,
      "loss": 1.63942261,
      "memory(GiB)": 107.26,
      "step": 24680,
      "train_speed(iter/s)": 1.6361
    },
    {
      "acc": 0.66725736,
      "epoch": 0.6262049720953831,
      "grad_norm": 5.625,
      "learning_rate": 8.223887994244604e-06,
      "loss": 1.58191662,
      "memory(GiB)": 107.26,
      "step": 24685,
      "train_speed(iter/s)": 1.636136
    },
    {
      "acc": 0.66614647,
      "epoch": 0.6263318112633182,
      "grad_norm": 5.5,
      "learning_rate": 8.223086386627314e-06,
      "loss": 1.57282143,
      "memory(GiB)": 107.26,
      "step": 24690,
      "train_speed(iter/s)": 1.636175
    },
    {
      "acc": 0.6728343,
      "epoch": 0.6264586504312532,
      "grad_norm": 6.1875,
      "learning_rate": 8.222284637244296e-06,
      "loss": 1.53513422,
      "memory(GiB)": 107.26,
      "step": 24695,
      "train_speed(iter/s)": 1.636214
    },
    {
      "acc": 0.66431308,
      "epoch": 0.6265854895991883,
      "grad_norm": 4.71875,
      "learning_rate": 8.221482746130811e-06,
      "loss": 1.55937099,
      "memory(GiB)": 107.26,
      "step": 24700,
      "train_speed(iter/s)": 1.636253
    },
    {
      "acc": 0.65554366,
      "epoch": 0.6267123287671232,
      "grad_norm": 5.6875,
      "learning_rate": 8.220680713322131e-06,
      "loss": 1.61941605,
      "memory(GiB)": 107.26,
      "step": 24705,
      "train_speed(iter/s)": 1.636289
    },
    {
      "acc": 0.65283508,
      "epoch": 0.6268391679350583,
      "grad_norm": 5.5,
      "learning_rate": 8.219878538853537e-06,
      "loss": 1.60055885,
      "memory(GiB)": 107.26,
      "step": 24710,
      "train_speed(iter/s)": 1.636323
    },
    {
      "acc": 0.6452179,
      "epoch": 0.6269660071029934,
      "grad_norm": 4.96875,
      "learning_rate": 8.219076222760307e-06,
      "loss": 1.67959328,
      "memory(GiB)": 107.26,
      "step": 24715,
      "train_speed(iter/s)": 1.636361
    },
    {
      "acc": 0.63807435,
      "epoch": 0.6270928462709284,
      "grad_norm": 4.9375,
      "learning_rate": 8.218273765077734e-06,
      "loss": 1.62349262,
      "memory(GiB)": 107.26,
      "step": 24720,
      "train_speed(iter/s)": 1.636399
    },
    {
      "acc": 0.6541852,
      "epoch": 0.6272196854388635,
      "grad_norm": 5.375,
      "learning_rate": 8.21747116584111e-06,
      "loss": 1.6081852,
      "memory(GiB)": 107.26,
      "step": 24725,
      "train_speed(iter/s)": 1.636436
    },
    {
      "acc": 0.64146233,
      "epoch": 0.6273465246067986,
      "grad_norm": 5.0625,
      "learning_rate": 8.21666842508574e-06,
      "loss": 1.56283884,
      "memory(GiB)": 107.26,
      "step": 24730,
      "train_speed(iter/s)": 1.636473
    },
    {
      "acc": 0.64809847,
      "epoch": 0.6274733637747336,
      "grad_norm": 6.46875,
      "learning_rate": 8.215865542846932e-06,
      "loss": 1.68420887,
      "memory(GiB)": 107.26,
      "step": 24735,
      "train_speed(iter/s)": 1.636511
    },
    {
      "acc": 0.64047184,
      "epoch": 0.6276002029426687,
      "grad_norm": 7.46875,
      "learning_rate": 8.215062519160002e-06,
      "loss": 1.63542347,
      "memory(GiB)": 107.26,
      "step": 24740,
      "train_speed(iter/s)": 1.636548
    },
    {
      "acc": 0.65750084,
      "epoch": 0.6277270421106037,
      "grad_norm": 5.9375,
      "learning_rate": 8.214259354060263e-06,
      "loss": 1.62848167,
      "memory(GiB)": 107.26,
      "step": 24745,
      "train_speed(iter/s)": 1.636587
    },
    {
      "acc": 0.65126572,
      "epoch": 0.6278538812785388,
      "grad_norm": 6.34375,
      "learning_rate": 8.21345604758305e-06,
      "loss": 1.58686419,
      "memory(GiB)": 107.26,
      "step": 24750,
      "train_speed(iter/s)": 1.636626
    },
    {
      "acc": 0.65963607,
      "epoch": 0.6279807204464739,
      "grad_norm": 4.96875,
      "learning_rate": 8.212652599763693e-06,
      "loss": 1.5857255,
      "memory(GiB)": 107.26,
      "step": 24755,
      "train_speed(iter/s)": 1.636663
    },
    {
      "acc": 0.64530449,
      "epoch": 0.6281075596144089,
      "grad_norm": 7.46875,
      "learning_rate": 8.211849010637532e-06,
      "loss": 1.6340374,
      "memory(GiB)": 107.26,
      "step": 24760,
      "train_speed(iter/s)": 1.636703
    },
    {
      "acc": 0.6404911,
      "epoch": 0.628234398782344,
      "grad_norm": 5.8125,
      "learning_rate": 8.211045280239908e-06,
      "loss": 1.5899806,
      "memory(GiB)": 107.26,
      "step": 24765,
      "train_speed(iter/s)": 1.636739
    },
    {
      "acc": 0.65176077,
      "epoch": 0.6283612379502791,
      "grad_norm": 6.0625,
      "learning_rate": 8.210241408606182e-06,
      "loss": 1.5757411,
      "memory(GiB)": 107.26,
      "step": 24770,
      "train_speed(iter/s)": 1.636775
    },
    {
      "acc": 0.66138029,
      "epoch": 0.6284880771182141,
      "grad_norm": 6.59375,
      "learning_rate": 8.2094373957717e-06,
      "loss": 1.62178822,
      "memory(GiB)": 107.26,
      "step": 24775,
      "train_speed(iter/s)": 1.636815
    },
    {
      "acc": 0.64928465,
      "epoch": 0.6286149162861492,
      "grad_norm": 5.6875,
      "learning_rate": 8.208633241771836e-06,
      "loss": 1.62128124,
      "memory(GiB)": 107.26,
      "step": 24780,
      "train_speed(iter/s)": 1.636852
    },
    {
      "acc": 0.65925827,
      "epoch": 0.6287417554540842,
      "grad_norm": 5.53125,
      "learning_rate": 8.207828946641956e-06,
      "loss": 1.61395226,
      "memory(GiB)": 107.26,
      "step": 24785,
      "train_speed(iter/s)": 1.636888
    },
    {
      "acc": 0.66922345,
      "epoch": 0.6288685946220193,
      "grad_norm": 4.625,
      "learning_rate": 8.207024510417436e-06,
      "loss": 1.59678698,
      "memory(GiB)": 107.26,
      "step": 24790,
      "train_speed(iter/s)": 1.636927
    },
    {
      "acc": 0.65218124,
      "epoch": 0.6289954337899544,
      "grad_norm": 7.21875,
      "learning_rate": 8.20621993313366e-06,
      "loss": 1.65590324,
      "memory(GiB)": 107.26,
      "step": 24795,
      "train_speed(iter/s)": 1.636966
    },
    {
      "acc": 0.66004591,
      "epoch": 0.6291222729578894,
      "grad_norm": 5.46875,
      "learning_rate": 8.205415214826018e-06,
      "loss": 1.74226646,
      "memory(GiB)": 107.26,
      "step": 24800,
      "train_speed(iter/s)": 1.637001
    },
    {
      "acc": 0.6490797,
      "epoch": 0.6292491121258245,
      "grad_norm": 5.15625,
      "learning_rate": 8.204610355529901e-06,
      "loss": 1.62131805,
      "memory(GiB)": 107.26,
      "step": 24805,
      "train_speed(iter/s)": 1.637039
    },
    {
      "acc": 0.64782681,
      "epoch": 0.6293759512937596,
      "grad_norm": 5.375,
      "learning_rate": 8.203805355280715e-06,
      "loss": 1.6390831,
      "memory(GiB)": 107.26,
      "step": 24810,
      "train_speed(iter/s)": 1.637075
    },
    {
      "acc": 0.65443325,
      "epoch": 0.6295027904616946,
      "grad_norm": 5.15625,
      "learning_rate": 8.203000214113865e-06,
      "loss": 1.64088039,
      "memory(GiB)": 107.26,
      "step": 24815,
      "train_speed(iter/s)": 1.637113
    },
    {
      "acc": 0.66053181,
      "epoch": 0.6296296296296297,
      "grad_norm": 5.3125,
      "learning_rate": 8.202194932064767e-06,
      "loss": 1.58313808,
      "memory(GiB)": 107.26,
      "step": 24820,
      "train_speed(iter/s)": 1.637148
    },
    {
      "acc": 0.65155458,
      "epoch": 0.6297564687975646,
      "grad_norm": 6.59375,
      "learning_rate": 8.201389509168836e-06,
      "loss": 1.63907413,
      "memory(GiB)": 107.26,
      "step": 24825,
      "train_speed(iter/s)": 1.637184
    },
    {
      "acc": 0.6377501,
      "epoch": 0.6298833079654997,
      "grad_norm": 6.40625,
      "learning_rate": 8.200583945461502e-06,
      "loss": 1.64891891,
      "memory(GiB)": 107.26,
      "step": 24830,
      "train_speed(iter/s)": 1.637221
    },
    {
      "acc": 0.65754781,
      "epoch": 0.6300101471334348,
      "grad_norm": 5.96875,
      "learning_rate": 8.199778240978197e-06,
      "loss": 1.57515907,
      "memory(GiB)": 107.26,
      "step": 24835,
      "train_speed(iter/s)": 1.637257
    },
    {
      "acc": 0.63422813,
      "epoch": 0.6301369863013698,
      "grad_norm": 7.03125,
      "learning_rate": 8.19897239575436e-06,
      "loss": 1.69267731,
      "memory(GiB)": 107.26,
      "step": 24840,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.63644643,
      "epoch": 0.6302638254693049,
      "grad_norm": 4.71875,
      "learning_rate": 8.198166409825434e-06,
      "loss": 1.68144569,
      "memory(GiB)": 107.26,
      "step": 24845,
      "train_speed(iter/s)": 1.637328
    },
    {
      "acc": 0.65338507,
      "epoch": 0.63039066463724,
      "grad_norm": 6.84375,
      "learning_rate": 8.19736028322687e-06,
      "loss": 1.60599079,
      "memory(GiB)": 107.26,
      "step": 24850,
      "train_speed(iter/s)": 1.637365
    },
    {
      "acc": 0.64865341,
      "epoch": 0.630517503805175,
      "grad_norm": 5.15625,
      "learning_rate": 8.196554015994126e-06,
      "loss": 1.64108543,
      "memory(GiB)": 107.26,
      "step": 24855,
      "train_speed(iter/s)": 1.637401
    },
    {
      "acc": 0.66276264,
      "epoch": 0.6306443429731101,
      "grad_norm": 6.15625,
      "learning_rate": 8.195747608162665e-06,
      "loss": 1.6303894,
      "memory(GiB)": 107.26,
      "step": 24860,
      "train_speed(iter/s)": 1.637439
    },
    {
      "acc": 0.65164747,
      "epoch": 0.6307711821410451,
      "grad_norm": 5.09375,
      "learning_rate": 8.194941059767957e-06,
      "loss": 1.61740971,
      "memory(GiB)": 107.26,
      "step": 24865,
      "train_speed(iter/s)": 1.637477
    },
    {
      "acc": 0.64635897,
      "epoch": 0.6308980213089802,
      "grad_norm": 5.78125,
      "learning_rate": 8.194134370845474e-06,
      "loss": 1.60363312,
      "memory(GiB)": 107.26,
      "step": 24870,
      "train_speed(iter/s)": 1.637514
    },
    {
      "acc": 0.6533236,
      "epoch": 0.6310248604769153,
      "grad_norm": 6.5625,
      "learning_rate": 8.193327541430703e-06,
      "loss": 1.64744797,
      "memory(GiB)": 107.26,
      "step": 24875,
      "train_speed(iter/s)": 1.637552
    },
    {
      "acc": 0.65546293,
      "epoch": 0.6311516996448503,
      "grad_norm": 5.84375,
      "learning_rate": 8.192520571559128e-06,
      "loss": 1.64636726,
      "memory(GiB)": 107.26,
      "step": 24880,
      "train_speed(iter/s)": 1.637589
    },
    {
      "acc": 0.65899286,
      "epoch": 0.6312785388127854,
      "grad_norm": 5.90625,
      "learning_rate": 8.191713461266246e-06,
      "loss": 1.60760269,
      "memory(GiB)": 107.26,
      "step": 24885,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.65114822,
      "epoch": 0.6314053779807205,
      "grad_norm": 5.65625,
      "learning_rate": 8.190906210587555e-06,
      "loss": 1.63634949,
      "memory(GiB)": 107.26,
      "step": 24890,
      "train_speed(iter/s)": 1.637663
    },
    {
      "acc": 0.65625172,
      "epoch": 0.6315322171486555,
      "grad_norm": 5.40625,
      "learning_rate": 8.190098819558562e-06,
      "loss": 1.58358479,
      "memory(GiB)": 107.26,
      "step": 24895,
      "train_speed(iter/s)": 1.637701
    },
    {
      "acc": 0.65328989,
      "epoch": 0.6316590563165906,
      "grad_norm": 4.53125,
      "learning_rate": 8.189291288214782e-06,
      "loss": 1.60500851,
      "memory(GiB)": 107.26,
      "step": 24900,
      "train_speed(iter/s)": 1.637738
    },
    {
      "acc": 0.64872632,
      "epoch": 0.6317858954845256,
      "grad_norm": 4.46875,
      "learning_rate": 8.18848361659173e-06,
      "loss": 1.62037506,
      "memory(GiB)": 107.26,
      "step": 24905,
      "train_speed(iter/s)": 1.637774
    },
    {
      "acc": 0.67482018,
      "epoch": 0.6319127346524607,
      "grad_norm": 5.46875,
      "learning_rate": 8.187675804724935e-06,
      "loss": 1.56209764,
      "memory(GiB)": 107.26,
      "step": 24910,
      "train_speed(iter/s)": 1.63781
    },
    {
      "acc": 0.64697728,
      "epoch": 0.6320395738203958,
      "grad_norm": 5.28125,
      "learning_rate": 8.186867852649925e-06,
      "loss": 1.62317314,
      "memory(GiB)": 107.26,
      "step": 24915,
      "train_speed(iter/s)": 1.637848
    },
    {
      "acc": 0.65313296,
      "epoch": 0.6321664129883308,
      "grad_norm": 5.59375,
      "learning_rate": 8.186059760402238e-06,
      "loss": 1.62176342,
      "memory(GiB)": 107.26,
      "step": 24920,
      "train_speed(iter/s)": 1.637884
    },
    {
      "acc": 0.66063852,
      "epoch": 0.6322932521562659,
      "grad_norm": 5.40625,
      "learning_rate": 8.185251528017419e-06,
      "loss": 1.60959129,
      "memory(GiB)": 107.26,
      "step": 24925,
      "train_speed(iter/s)": 1.637919
    },
    {
      "acc": 0.65080404,
      "epoch": 0.632420091324201,
      "grad_norm": 6.0625,
      "learning_rate": 8.184443155531016e-06,
      "loss": 1.62780094,
      "memory(GiB)": 107.26,
      "step": 24930,
      "train_speed(iter/s)": 1.637956
    },
    {
      "acc": 0.65380893,
      "epoch": 0.632546930492136,
      "grad_norm": 5.25,
      "learning_rate": 8.183634642978586e-06,
      "loss": 1.61431084,
      "memory(GiB)": 107.26,
      "step": 24935,
      "train_speed(iter/s)": 1.637992
    },
    {
      "acc": 0.65905294,
      "epoch": 0.632673769660071,
      "grad_norm": 7.25,
      "learning_rate": 8.18282599039569e-06,
      "loss": 1.62732964,
      "memory(GiB)": 107.26,
      "step": 24940,
      "train_speed(iter/s)": 1.638029
    },
    {
      "acc": 0.65347195,
      "epoch": 0.632800608828006,
      "grad_norm": 6.03125,
      "learning_rate": 8.182017197817898e-06,
      "loss": 1.6477684,
      "memory(GiB)": 107.26,
      "step": 24945,
      "train_speed(iter/s)": 1.638064
    },
    {
      "acc": 0.66291871,
      "epoch": 0.6329274479959411,
      "grad_norm": 5.5,
      "learning_rate": 8.181208265280782e-06,
      "loss": 1.59911957,
      "memory(GiB)": 107.26,
      "step": 24950,
      "train_speed(iter/s)": 1.638098
    },
    {
      "acc": 0.64377847,
      "epoch": 0.6330542871638762,
      "grad_norm": 5.71875,
      "learning_rate": 8.180399192819923e-06,
      "loss": 1.70650063,
      "memory(GiB)": 107.26,
      "step": 24955,
      "train_speed(iter/s)": 1.638134
    },
    {
      "acc": 0.65112553,
      "epoch": 0.6331811263318112,
      "grad_norm": 6.78125,
      "learning_rate": 8.17958998047091e-06,
      "loss": 1.63111649,
      "memory(GiB)": 107.26,
      "step": 24960,
      "train_speed(iter/s)": 1.638169
    },
    {
      "acc": 0.6614501,
      "epoch": 0.6333079654997463,
      "grad_norm": 6.5625,
      "learning_rate": 8.178780628269332e-06,
      "loss": 1.56051483,
      "memory(GiB)": 107.26,
      "step": 24965,
      "train_speed(iter/s)": 1.638207
    },
    {
      "acc": 0.65216112,
      "epoch": 0.6334348046676814,
      "grad_norm": 5.125,
      "learning_rate": 8.177971136250788e-06,
      "loss": 1.67489262,
      "memory(GiB)": 107.26,
      "step": 24970,
      "train_speed(iter/s)": 1.638239
    },
    {
      "acc": 0.65246515,
      "epoch": 0.6335616438356164,
      "grad_norm": 5.28125,
      "learning_rate": 8.177161504450887e-06,
      "loss": 1.59421234,
      "memory(GiB)": 107.26,
      "step": 24975,
      "train_speed(iter/s)": 1.638276
    },
    {
      "acc": 0.64446716,
      "epoch": 0.6336884830035515,
      "grad_norm": 5.59375,
      "learning_rate": 8.176351732905239e-06,
      "loss": 1.68438053,
      "memory(GiB)": 107.26,
      "step": 24980,
      "train_speed(iter/s)": 1.638313
    },
    {
      "acc": 0.65932193,
      "epoch": 0.6338153221714865,
      "grad_norm": 6.40625,
      "learning_rate": 8.175541821649459e-06,
      "loss": 1.60514469,
      "memory(GiB)": 107.26,
      "step": 24985,
      "train_speed(iter/s)": 1.638348
    },
    {
      "acc": 0.6764039,
      "epoch": 0.6339421613394216,
      "grad_norm": 6.09375,
      "learning_rate": 8.174731770719173e-06,
      "loss": 1.53645611,
      "memory(GiB)": 107.26,
      "step": 24990,
      "train_speed(iter/s)": 1.638382
    },
    {
      "acc": 0.65205746,
      "epoch": 0.6340690005073567,
      "grad_norm": 5.15625,
      "learning_rate": 8.173921580150008e-06,
      "loss": 1.62166176,
      "memory(GiB)": 107.26,
      "step": 24995,
      "train_speed(iter/s)": 1.638417
    },
    {
      "acc": 0.645013,
      "epoch": 0.6341958396752917,
      "grad_norm": 6.40625,
      "learning_rate": 8.173111249977602e-06,
      "loss": 1.68191547,
      "memory(GiB)": 107.26,
      "step": 25000,
      "train_speed(iter/s)": 1.638454
    },
    {
      "epoch": 0.6341958396752917,
      "eval_acc": 0.644863497870972,
      "eval_loss": 1.581565022468567,
      "eval_runtime": 58.0014,
      "eval_samples_per_second": 109.825,
      "eval_steps_per_second": 27.465,
      "step": 25000
    },
    {
      "acc": 0.66443939,
      "epoch": 0.6343226788432268,
      "grad_norm": 5.71875,
      "learning_rate": 8.172300780237596e-06,
      "loss": 1.55179424,
      "memory(GiB)": 107.26,
      "step": 25005,
      "train_speed(iter/s)": 1.631833
    },
    {
      "acc": 0.65388923,
      "epoch": 0.6344495180111619,
      "grad_norm": 6.125,
      "learning_rate": 8.171490170965639e-06,
      "loss": 1.59664783,
      "memory(GiB)": 107.26,
      "step": 25010,
      "train_speed(iter/s)": 1.63187
    },
    {
      "acc": 0.66988387,
      "epoch": 0.6345763571790969,
      "grad_norm": 5.9375,
      "learning_rate": 8.170679422197385e-06,
      "loss": 1.52830715,
      "memory(GiB)": 107.26,
      "step": 25015,
      "train_speed(iter/s)": 1.631905
    },
    {
      "acc": 0.65053511,
      "epoch": 0.634703196347032,
      "grad_norm": 5.875,
      "learning_rate": 8.169868533968493e-06,
      "loss": 1.61230907,
      "memory(GiB)": 107.26,
      "step": 25020,
      "train_speed(iter/s)": 1.631943
    },
    {
      "acc": 0.64172406,
      "epoch": 0.634830035514967,
      "grad_norm": 5.0625,
      "learning_rate": 8.16905750631463e-06,
      "loss": 1.63221207,
      "memory(GiB)": 107.26,
      "step": 25025,
      "train_speed(iter/s)": 1.631979
    },
    {
      "acc": 0.66764474,
      "epoch": 0.6349568746829021,
      "grad_norm": 7.71875,
      "learning_rate": 8.168246339271471e-06,
      "loss": 1.60535107,
      "memory(GiB)": 107.26,
      "step": 25030,
      "train_speed(iter/s)": 1.632015
    },
    {
      "acc": 0.63745842,
      "epoch": 0.6350837138508372,
      "grad_norm": 5.71875,
      "learning_rate": 8.16743503287469e-06,
      "loss": 1.63422623,
      "memory(GiB)": 107.26,
      "step": 25035,
      "train_speed(iter/s)": 1.632053
    },
    {
      "acc": 0.65079484,
      "epoch": 0.6352105530187722,
      "grad_norm": 5.8125,
      "learning_rate": 8.166623587159978e-06,
      "loss": 1.57841473,
      "memory(GiB)": 107.26,
      "step": 25040,
      "train_speed(iter/s)": 1.632087
    },
    {
      "acc": 0.65513973,
      "epoch": 0.6353373921867073,
      "grad_norm": 5.84375,
      "learning_rate": 8.16581200216302e-06,
      "loss": 1.57435713,
      "memory(GiB)": 107.26,
      "step": 25045,
      "train_speed(iter/s)": 1.632126
    },
    {
      "acc": 0.64942665,
      "epoch": 0.6354642313546424,
      "grad_norm": 4.875,
      "learning_rate": 8.165000277919517e-06,
      "loss": 1.64163494,
      "memory(GiB)": 107.26,
      "step": 25050,
      "train_speed(iter/s)": 1.632162
    },
    {
      "acc": 0.65720091,
      "epoch": 0.6355910705225774,
      "grad_norm": 6.09375,
      "learning_rate": 8.16418841446517e-06,
      "loss": 1.59496078,
      "memory(GiB)": 107.26,
      "step": 25055,
      "train_speed(iter/s)": 1.632198
    },
    {
      "acc": 0.65523882,
      "epoch": 0.6357179096905125,
      "grad_norm": 5.28125,
      "learning_rate": 8.163376411835691e-06,
      "loss": 1.57789049,
      "memory(GiB)": 107.26,
      "step": 25060,
      "train_speed(iter/s)": 1.632237
    },
    {
      "acc": 0.67631674,
      "epoch": 0.6358447488584474,
      "grad_norm": 4.53125,
      "learning_rate": 8.162564270066793e-06,
      "loss": 1.504039,
      "memory(GiB)": 107.26,
      "step": 25065,
      "train_speed(iter/s)": 1.632275
    },
    {
      "acc": 0.65791121,
      "epoch": 0.6359715880263825,
      "grad_norm": 5.8125,
      "learning_rate": 8.1617519891942e-06,
      "loss": 1.61463451,
      "memory(GiB)": 107.26,
      "step": 25070,
      "train_speed(iter/s)": 1.632312
    },
    {
      "acc": 0.65800676,
      "epoch": 0.6360984271943176,
      "grad_norm": 5.5,
      "learning_rate": 8.160939569253637e-06,
      "loss": 1.55168028,
      "memory(GiB)": 107.26,
      "step": 25075,
      "train_speed(iter/s)": 1.632348
    },
    {
      "acc": 0.65362477,
      "epoch": 0.6362252663622526,
      "grad_norm": 5.375,
      "learning_rate": 8.160127010280838e-06,
      "loss": 1.60622139,
      "memory(GiB)": 107.26,
      "step": 25080,
      "train_speed(iter/s)": 1.632386
    },
    {
      "acc": 0.64838934,
      "epoch": 0.6363521055301877,
      "grad_norm": 7.09375,
      "learning_rate": 8.159314312311546e-06,
      "loss": 1.66545372,
      "memory(GiB)": 107.26,
      "step": 25085,
      "train_speed(iter/s)": 1.632422
    },
    {
      "acc": 0.64707794,
      "epoch": 0.6364789446981228,
      "grad_norm": 5.375,
      "learning_rate": 8.158501475381505e-06,
      "loss": 1.66130714,
      "memory(GiB)": 107.26,
      "step": 25090,
      "train_speed(iter/s)": 1.632458
    },
    {
      "acc": 0.65416427,
      "epoch": 0.6366057838660578,
      "grad_norm": 5.375,
      "learning_rate": 8.157688499526466e-06,
      "loss": 1.55415554,
      "memory(GiB)": 107.26,
      "step": 25095,
      "train_speed(iter/s)": 1.632494
    },
    {
      "acc": 0.65607867,
      "epoch": 0.6367326230339929,
      "grad_norm": 6.0,
      "learning_rate": 8.15687538478219e-06,
      "loss": 1.61145306,
      "memory(GiB)": 107.26,
      "step": 25100,
      "train_speed(iter/s)": 1.632533
    },
    {
      "acc": 0.64754839,
      "epoch": 0.6368594622019279,
      "grad_norm": 5.9375,
      "learning_rate": 8.156062131184439e-06,
      "loss": 1.65939617,
      "memory(GiB)": 107.26,
      "step": 25105,
      "train_speed(iter/s)": 1.632571
    },
    {
      "acc": 0.65528903,
      "epoch": 0.636986301369863,
      "grad_norm": 6.1875,
      "learning_rate": 8.155248738768986e-06,
      "loss": 1.61583443,
      "memory(GiB)": 107.26,
      "step": 25110,
      "train_speed(iter/s)": 1.632608
    },
    {
      "acc": 0.66784687,
      "epoch": 0.6371131405377981,
      "grad_norm": 5.8125,
      "learning_rate": 8.154435207571606e-06,
      "loss": 1.52366409,
      "memory(GiB)": 107.26,
      "step": 25115,
      "train_speed(iter/s)": 1.632646
    },
    {
      "acc": 0.66205354,
      "epoch": 0.6372399797057331,
      "grad_norm": 8.3125,
      "learning_rate": 8.153621537628083e-06,
      "loss": 1.58188858,
      "memory(GiB)": 107.26,
      "step": 25120,
      "train_speed(iter/s)": 1.632686
    },
    {
      "acc": 0.65227323,
      "epoch": 0.6373668188736682,
      "grad_norm": 5.96875,
      "learning_rate": 8.152807728974203e-06,
      "loss": 1.59889174,
      "memory(GiB)": 107.26,
      "step": 25125,
      "train_speed(iter/s)": 1.632723
    },
    {
      "acc": 0.6689105,
      "epoch": 0.6374936580416033,
      "grad_norm": 6.875,
      "learning_rate": 8.151993781645765e-06,
      "loss": 1.58072891,
      "memory(GiB)": 107.26,
      "step": 25130,
      "train_speed(iter/s)": 1.632761
    },
    {
      "acc": 0.63539681,
      "epoch": 0.6376204972095383,
      "grad_norm": 8.0,
      "learning_rate": 8.151179695678565e-06,
      "loss": 1.65160522,
      "memory(GiB)": 107.26,
      "step": 25135,
      "train_speed(iter/s)": 1.632801
    },
    {
      "acc": 0.67017813,
      "epoch": 0.6377473363774734,
      "grad_norm": 5.4375,
      "learning_rate": 8.150365471108414e-06,
      "loss": 1.53730745,
      "memory(GiB)": 107.26,
      "step": 25140,
      "train_speed(iter/s)": 1.63284
    },
    {
      "acc": 0.64716587,
      "epoch": 0.6378741755454084,
      "grad_norm": 4.96875,
      "learning_rate": 8.149551107971125e-06,
      "loss": 1.67645149,
      "memory(GiB)": 107.26,
      "step": 25145,
      "train_speed(iter/s)": 1.632879
    },
    {
      "acc": 0.64729729,
      "epoch": 0.6380010147133435,
      "grad_norm": 5.8125,
      "learning_rate": 8.148736606302517e-06,
      "loss": 1.63931675,
      "memory(GiB)": 107.26,
      "step": 25150,
      "train_speed(iter/s)": 1.632919
    },
    {
      "acc": 0.6407692,
      "epoch": 0.6381278538812786,
      "grad_norm": 5.65625,
      "learning_rate": 8.147921966138412e-06,
      "loss": 1.65959816,
      "memory(GiB)": 107.26,
      "step": 25155,
      "train_speed(iter/s)": 1.632958
    },
    {
      "acc": 0.67805433,
      "epoch": 0.6382546930492136,
      "grad_norm": 4.78125,
      "learning_rate": 8.147107187514647e-06,
      "loss": 1.52108879,
      "memory(GiB)": 107.26,
      "step": 25160,
      "train_speed(iter/s)": 1.632995
    },
    {
      "acc": 0.66085944,
      "epoch": 0.6383815322171487,
      "grad_norm": 4.625,
      "learning_rate": 8.146292270467056e-06,
      "loss": 1.50633984,
      "memory(GiB)": 107.26,
      "step": 25165,
      "train_speed(iter/s)": 1.633031
    },
    {
      "acc": 0.66203742,
      "epoch": 0.6385083713850838,
      "grad_norm": 5.65625,
      "learning_rate": 8.145477215031486e-06,
      "loss": 1.57740507,
      "memory(GiB)": 107.26,
      "step": 25170,
      "train_speed(iter/s)": 1.633069
    },
    {
      "acc": 0.65342994,
      "epoch": 0.6386352105530188,
      "grad_norm": 6.21875,
      "learning_rate": 8.144662021243782e-06,
      "loss": 1.53255253,
      "memory(GiB)": 107.26,
      "step": 25175,
      "train_speed(iter/s)": 1.633109
    },
    {
      "acc": 0.64925594,
      "epoch": 0.6387620497209539,
      "grad_norm": 5.1875,
      "learning_rate": 8.143846689139805e-06,
      "loss": 1.65110722,
      "memory(GiB)": 107.26,
      "step": 25180,
      "train_speed(iter/s)": 1.633149
    },
    {
      "acc": 0.64615192,
      "epoch": 0.6388888888888888,
      "grad_norm": 4.8125,
      "learning_rate": 8.143031218755411e-06,
      "loss": 1.64296265,
      "memory(GiB)": 107.26,
      "step": 25185,
      "train_speed(iter/s)": 1.633188
    },
    {
      "acc": 0.63147774,
      "epoch": 0.6390157280568239,
      "grad_norm": 6.21875,
      "learning_rate": 8.142215610126474e-06,
      "loss": 1.68469028,
      "memory(GiB)": 107.26,
      "step": 25190,
      "train_speed(iter/s)": 1.633225
    },
    {
      "acc": 0.66989598,
      "epoch": 0.639142567224759,
      "grad_norm": 5.59375,
      "learning_rate": 8.141399863288863e-06,
      "loss": 1.56986675,
      "memory(GiB)": 107.26,
      "step": 25195,
      "train_speed(iter/s)": 1.633261
    },
    {
      "acc": 0.64840164,
      "epoch": 0.639269406392694,
      "grad_norm": 5.1875,
      "learning_rate": 8.140583978278463e-06,
      "loss": 1.60602264,
      "memory(GiB)": 107.26,
      "step": 25200,
      "train_speed(iter/s)": 1.633299
    },
    {
      "acc": 0.67312622,
      "epoch": 0.6393962455606291,
      "grad_norm": 5.0625,
      "learning_rate": 8.139767955131157e-06,
      "loss": 1.41391335,
      "memory(GiB)": 107.26,
      "step": 25205,
      "train_speed(iter/s)": 1.633336
    },
    {
      "acc": 0.67137156,
      "epoch": 0.6395230847285642,
      "grad_norm": 5.28125,
      "learning_rate": 8.138951793882838e-06,
      "loss": 1.54170761,
      "memory(GiB)": 107.26,
      "step": 25210,
      "train_speed(iter/s)": 1.633374
    },
    {
      "acc": 0.67052155,
      "epoch": 0.6396499238964992,
      "grad_norm": 5.78125,
      "learning_rate": 8.138135494569405e-06,
      "loss": 1.55081882,
      "memory(GiB)": 107.26,
      "step": 25215,
      "train_speed(iter/s)": 1.633412
    },
    {
      "acc": 0.65794039,
      "epoch": 0.6397767630644343,
      "grad_norm": 5.9375,
      "learning_rate": 8.137319057226763e-06,
      "loss": 1.54577208,
      "memory(GiB)": 107.26,
      "step": 25220,
      "train_speed(iter/s)": 1.633449
    },
    {
      "acc": 0.65920897,
      "epoch": 0.6399036022323693,
      "grad_norm": 5.09375,
      "learning_rate": 8.136502481890821e-06,
      "loss": 1.62497635,
      "memory(GiB)": 107.26,
      "step": 25225,
      "train_speed(iter/s)": 1.633486
    },
    {
      "acc": 0.66269212,
      "epoch": 0.6400304414003044,
      "grad_norm": 7.21875,
      "learning_rate": 8.135685768597496e-06,
      "loss": 1.59908581,
      "memory(GiB)": 107.26,
      "step": 25230,
      "train_speed(iter/s)": 1.633514
    },
    {
      "acc": 0.65427208,
      "epoch": 0.6401572805682395,
      "grad_norm": 5.125,
      "learning_rate": 8.134868917382713e-06,
      "loss": 1.56183453,
      "memory(GiB)": 107.26,
      "step": 25235,
      "train_speed(iter/s)": 1.633549
    },
    {
      "acc": 0.64797058,
      "epoch": 0.6402841197361745,
      "grad_norm": 5.15625,
      "learning_rate": 8.134051928282396e-06,
      "loss": 1.61085243,
      "memory(GiB)": 107.26,
      "step": 25240,
      "train_speed(iter/s)": 1.633585
    },
    {
      "acc": 0.66116238,
      "epoch": 0.6404109589041096,
      "grad_norm": 5.21875,
      "learning_rate": 8.133234801332484e-06,
      "loss": 1.63974171,
      "memory(GiB)": 107.26,
      "step": 25245,
      "train_speed(iter/s)": 1.633622
    },
    {
      "acc": 0.66599889,
      "epoch": 0.6405377980720447,
      "grad_norm": 5.875,
      "learning_rate": 8.132417536568918e-06,
      "loss": 1.55043354,
      "memory(GiB)": 107.26,
      "step": 25250,
      "train_speed(iter/s)": 1.63366
    },
    {
      "acc": 0.6334969,
      "epoch": 0.6406646372399797,
      "grad_norm": 5.6875,
      "learning_rate": 8.131600134027641e-06,
      "loss": 1.67072468,
      "memory(GiB)": 107.26,
      "step": 25255,
      "train_speed(iter/s)": 1.633697
    },
    {
      "acc": 0.65395894,
      "epoch": 0.6407914764079148,
      "grad_norm": 5.96875,
      "learning_rate": 8.13078259374461e-06,
      "loss": 1.62380371,
      "memory(GiB)": 107.26,
      "step": 25260,
      "train_speed(iter/s)": 1.633734
    },
    {
      "acc": 0.64973154,
      "epoch": 0.6409183155758498,
      "grad_norm": 5.9375,
      "learning_rate": 8.129964915755781e-06,
      "loss": 1.62462616,
      "memory(GiB)": 107.26,
      "step": 25265,
      "train_speed(iter/s)": 1.633772
    },
    {
      "acc": 0.66672859,
      "epoch": 0.6410451547437849,
      "grad_norm": 4.65625,
      "learning_rate": 8.129147100097122e-06,
      "loss": 1.52427235,
      "memory(GiB)": 107.26,
      "step": 25270,
      "train_speed(iter/s)": 1.633809
    },
    {
      "acc": 0.65974579,
      "epoch": 0.64117199391172,
      "grad_norm": 6.53125,
      "learning_rate": 8.128329146804604e-06,
      "loss": 1.62272339,
      "memory(GiB)": 107.26,
      "step": 25275,
      "train_speed(iter/s)": 1.633845
    },
    {
      "acc": 0.65485783,
      "epoch": 0.641298833079655,
      "grad_norm": 6.84375,
      "learning_rate": 8.127511055914201e-06,
      "loss": 1.58440361,
      "memory(GiB)": 107.26,
      "step": 25280,
      "train_speed(iter/s)": 1.633883
    },
    {
      "acc": 0.67908206,
      "epoch": 0.6414256722475901,
      "grad_norm": 6.5,
      "learning_rate": 8.1266928274619e-06,
      "loss": 1.58362217,
      "memory(GiB)": 107.26,
      "step": 25285,
      "train_speed(iter/s)": 1.633919
    },
    {
      "acc": 0.65474596,
      "epoch": 0.6415525114155252,
      "grad_norm": 5.03125,
      "learning_rate": 8.125874461483687e-06,
      "loss": 1.58742056,
      "memory(GiB)": 107.26,
      "step": 25290,
      "train_speed(iter/s)": 1.633956
    },
    {
      "acc": 0.62787018,
      "epoch": 0.6416793505834602,
      "grad_norm": 5.6875,
      "learning_rate": 8.12505595801556e-06,
      "loss": 1.69383907,
      "memory(GiB)": 107.26,
      "step": 25295,
      "train_speed(iter/s)": 1.633992
    },
    {
      "acc": 0.64466057,
      "epoch": 0.6418061897513953,
      "grad_norm": 9.125,
      "learning_rate": 8.12423731709352e-06,
      "loss": 1.66952839,
      "memory(GiB)": 107.26,
      "step": 25300,
      "train_speed(iter/s)": 1.634029
    },
    {
      "acc": 0.6457571,
      "epoch": 0.6419330289193302,
      "grad_norm": 5.21875,
      "learning_rate": 8.123418538753573e-06,
      "loss": 1.56150846,
      "memory(GiB)": 107.26,
      "step": 25305,
      "train_speed(iter/s)": 1.63407
    },
    {
      "acc": 0.65962567,
      "epoch": 0.6420598680872653,
      "grad_norm": 6.40625,
      "learning_rate": 8.122599623031735e-06,
      "loss": 1.57684612,
      "memory(GiB)": 107.26,
      "step": 25310,
      "train_speed(iter/s)": 1.634105
    },
    {
      "acc": 0.64726105,
      "epoch": 0.6421867072552004,
      "grad_norm": 5.09375,
      "learning_rate": 8.121780569964024e-06,
      "loss": 1.6117382,
      "memory(GiB)": 107.26,
      "step": 25315,
      "train_speed(iter/s)": 1.63414
    },
    {
      "acc": 0.66389976,
      "epoch": 0.6423135464231354,
      "grad_norm": 5.34375,
      "learning_rate": 8.120961379586466e-06,
      "loss": 1.57469578,
      "memory(GiB)": 107.26,
      "step": 25320,
      "train_speed(iter/s)": 1.634177
    },
    {
      "acc": 0.64762573,
      "epoch": 0.6424403855910705,
      "grad_norm": 6.0,
      "learning_rate": 8.120142051935092e-06,
      "loss": 1.66447239,
      "memory(GiB)": 107.26,
      "step": 25325,
      "train_speed(iter/s)": 1.634214
    },
    {
      "acc": 0.62941637,
      "epoch": 0.6425672247590056,
      "grad_norm": 5.9375,
      "learning_rate": 8.11932258704594e-06,
      "loss": 1.73394642,
      "memory(GiB)": 107.26,
      "step": 25330,
      "train_speed(iter/s)": 1.634251
    },
    {
      "acc": 0.63256741,
      "epoch": 0.6426940639269406,
      "grad_norm": 8.6875,
      "learning_rate": 8.118502984955053e-06,
      "loss": 1.7257637,
      "memory(GiB)": 107.26,
      "step": 25335,
      "train_speed(iter/s)": 1.634288
    },
    {
      "acc": 0.65086489,
      "epoch": 0.6428209030948757,
      "grad_norm": 5.71875,
      "learning_rate": 8.117683245698483e-06,
      "loss": 1.63473816,
      "memory(GiB)": 107.26,
      "step": 25340,
      "train_speed(iter/s)": 1.634325
    },
    {
      "acc": 0.65323277,
      "epoch": 0.6429477422628107,
      "grad_norm": 5.75,
      "learning_rate": 8.116863369312283e-06,
      "loss": 1.64795647,
      "memory(GiB)": 107.26,
      "step": 25345,
      "train_speed(iter/s)": 1.634363
    },
    {
      "acc": 0.66284151,
      "epoch": 0.6430745814307458,
      "grad_norm": 5.53125,
      "learning_rate": 8.116043355832518e-06,
      "loss": 1.62753296,
      "memory(GiB)": 107.26,
      "step": 25350,
      "train_speed(iter/s)": 1.6344
    },
    {
      "acc": 0.6454772,
      "epoch": 0.6432014205986809,
      "grad_norm": 4.5625,
      "learning_rate": 8.115223205295253e-06,
      "loss": 1.6379509,
      "memory(GiB)": 107.26,
      "step": 25355,
      "train_speed(iter/s)": 1.634437
    },
    {
      "acc": 0.65888605,
      "epoch": 0.6433282597666159,
      "grad_norm": 6.875,
      "learning_rate": 8.114402917736563e-06,
      "loss": 1.60328274,
      "memory(GiB)": 107.26,
      "step": 25360,
      "train_speed(iter/s)": 1.634474
    },
    {
      "acc": 0.65666022,
      "epoch": 0.643455098934551,
      "grad_norm": 5.90625,
      "learning_rate": 8.113582493192529e-06,
      "loss": 1.55814066,
      "memory(GiB)": 107.26,
      "step": 25365,
      "train_speed(iter/s)": 1.634511
    },
    {
      "acc": 0.66418591,
      "epoch": 0.6435819381024861,
      "grad_norm": 5.84375,
      "learning_rate": 8.112761931699235e-06,
      "loss": 1.61468201,
      "memory(GiB)": 107.26,
      "step": 25370,
      "train_speed(iter/s)": 1.634545
    },
    {
      "acc": 0.6513351,
      "epoch": 0.6437087772704211,
      "grad_norm": 5.875,
      "learning_rate": 8.111941233292772e-06,
      "loss": 1.64838982,
      "memory(GiB)": 107.26,
      "step": 25375,
      "train_speed(iter/s)": 1.634581
    },
    {
      "acc": 0.66825876,
      "epoch": 0.6438356164383562,
      "grad_norm": 6.3125,
      "learning_rate": 8.111120398009243e-06,
      "loss": 1.56160393,
      "memory(GiB)": 107.26,
      "step": 25380,
      "train_speed(iter/s)": 1.634617
    },
    {
      "acc": 0.64440546,
      "epoch": 0.6439624556062912,
      "grad_norm": 5.40625,
      "learning_rate": 8.110299425884745e-06,
      "loss": 1.6469408,
      "memory(GiB)": 107.26,
      "step": 25385,
      "train_speed(iter/s)": 1.634654
    },
    {
      "acc": 0.64203129,
      "epoch": 0.6440892947742263,
      "grad_norm": 5.46875,
      "learning_rate": 8.109478316955394e-06,
      "loss": 1.67655334,
      "memory(GiB)": 107.26,
      "step": 25390,
      "train_speed(iter/s)": 1.634689
    },
    {
      "acc": 0.64533777,
      "epoch": 0.6442161339421614,
      "grad_norm": 5.3125,
      "learning_rate": 8.108657071257304e-06,
      "loss": 1.63391838,
      "memory(GiB)": 107.26,
      "step": 25395,
      "train_speed(iter/s)": 1.634727
    },
    {
      "acc": 0.64848204,
      "epoch": 0.6443429731100964,
      "grad_norm": 5.875,
      "learning_rate": 8.107835688826598e-06,
      "loss": 1.65008049,
      "memory(GiB)": 107.26,
      "step": 25400,
      "train_speed(iter/s)": 1.634763
    },
    {
      "acc": 0.66970668,
      "epoch": 0.6444698122780315,
      "grad_norm": 4.9375,
      "learning_rate": 8.1070141696994e-06,
      "loss": 1.56662645,
      "memory(GiB)": 107.26,
      "step": 25405,
      "train_speed(iter/s)": 1.634799
    },
    {
      "acc": 0.65604649,
      "epoch": 0.6445966514459666,
      "grad_norm": 6.0,
      "learning_rate": 8.106192513911849e-06,
      "loss": 1.59266396,
      "memory(GiB)": 107.26,
      "step": 25410,
      "train_speed(iter/s)": 1.634835
    },
    {
      "acc": 0.65174699,
      "epoch": 0.6447234906139016,
      "grad_norm": 5.28125,
      "learning_rate": 8.105370721500083e-06,
      "loss": 1.68827057,
      "memory(GiB)": 107.26,
      "step": 25415,
      "train_speed(iter/s)": 1.634869
    },
    {
      "acc": 0.65298123,
      "epoch": 0.6448503297818367,
      "grad_norm": 5.125,
      "learning_rate": 8.104548792500246e-06,
      "loss": 1.57413826,
      "memory(GiB)": 107.26,
      "step": 25420,
      "train_speed(iter/s)": 1.634904
    },
    {
      "acc": 0.6495698,
      "epoch": 0.6449771689497716,
      "grad_norm": 5.65625,
      "learning_rate": 8.103726726948495e-06,
      "loss": 1.64965038,
      "memory(GiB)": 107.26,
      "step": 25425,
      "train_speed(iter/s)": 1.63494
    },
    {
      "acc": 0.65652938,
      "epoch": 0.6451040081177067,
      "grad_norm": 4.8125,
      "learning_rate": 8.102904524880985e-06,
      "loss": 1.62134857,
      "memory(GiB)": 107.26,
      "step": 25430,
      "train_speed(iter/s)": 1.634976
    },
    {
      "acc": 0.64343343,
      "epoch": 0.6452308472856418,
      "grad_norm": 5.09375,
      "learning_rate": 8.10208218633388e-06,
      "loss": 1.64101143,
      "memory(GiB)": 107.26,
      "step": 25435,
      "train_speed(iter/s)": 1.635012
    },
    {
      "acc": 0.64966459,
      "epoch": 0.6453576864535768,
      "grad_norm": 11.75,
      "learning_rate": 8.10125971134335e-06,
      "loss": 1.64917374,
      "memory(GiB)": 107.26,
      "step": 25440,
      "train_speed(iter/s)": 1.635051
    },
    {
      "acc": 0.65247822,
      "epoch": 0.6454845256215119,
      "grad_norm": 5.5,
      "learning_rate": 8.100437099945572e-06,
      "loss": 1.58667545,
      "memory(GiB)": 107.26,
      "step": 25445,
      "train_speed(iter/s)": 1.635089
    },
    {
      "acc": 0.66119399,
      "epoch": 0.645611364789447,
      "grad_norm": 4.96875,
      "learning_rate": 8.099614352176727e-06,
      "loss": 1.56738052,
      "memory(GiB)": 107.26,
      "step": 25450,
      "train_speed(iter/s)": 1.635127
    },
    {
      "acc": 0.65651379,
      "epoch": 0.645738203957382,
      "grad_norm": 5.5625,
      "learning_rate": 8.098791468073007e-06,
      "loss": 1.61565437,
      "memory(GiB)": 107.26,
      "step": 25455,
      "train_speed(iter/s)": 1.635163
    },
    {
      "acc": 0.66619954,
      "epoch": 0.6458650431253171,
      "grad_norm": 6.625,
      "learning_rate": 8.097968447670601e-06,
      "loss": 1.59684935,
      "memory(GiB)": 107.26,
      "step": 25460,
      "train_speed(iter/s)": 1.6352
    },
    {
      "acc": 0.65974846,
      "epoch": 0.6459918822932521,
      "grad_norm": 6.125,
      "learning_rate": 8.09714529100571e-06,
      "loss": 1.60825329,
      "memory(GiB)": 107.26,
      "step": 25465,
      "train_speed(iter/s)": 1.635234
    },
    {
      "acc": 0.65640121,
      "epoch": 0.6461187214611872,
      "grad_norm": 6.0,
      "learning_rate": 8.096321998114545e-06,
      "loss": 1.61770458,
      "memory(GiB)": 107.26,
      "step": 25470,
      "train_speed(iter/s)": 1.63527
    },
    {
      "acc": 0.66895566,
      "epoch": 0.6462455606291223,
      "grad_norm": 6.5625,
      "learning_rate": 8.09549856903331e-06,
      "loss": 1.53334856,
      "memory(GiB)": 107.26,
      "step": 25475,
      "train_speed(iter/s)": 1.635305
    },
    {
      "acc": 0.64003773,
      "epoch": 0.6463723997970573,
      "grad_norm": 5.125,
      "learning_rate": 8.094675003798232e-06,
      "loss": 1.65117798,
      "memory(GiB)": 107.26,
      "step": 25480,
      "train_speed(iter/s)": 1.63534
    },
    {
      "acc": 0.67290869,
      "epoch": 0.6464992389649924,
      "grad_norm": 6.8125,
      "learning_rate": 8.093851302445528e-06,
      "loss": 1.54114876,
      "memory(GiB)": 107.26,
      "step": 25485,
      "train_speed(iter/s)": 1.635376
    },
    {
      "acc": 0.65590672,
      "epoch": 0.6466260781329275,
      "grad_norm": 5.40625,
      "learning_rate": 8.093027465011431e-06,
      "loss": 1.59310751,
      "memory(GiB)": 107.26,
      "step": 25490,
      "train_speed(iter/s)": 1.635413
    },
    {
      "acc": 0.65389652,
      "epoch": 0.6467529173008625,
      "grad_norm": 5.34375,
      "learning_rate": 8.092203491532178e-06,
      "loss": 1.57344627,
      "memory(GiB)": 107.26,
      "step": 25495,
      "train_speed(iter/s)": 1.635449
    },
    {
      "acc": 0.65665984,
      "epoch": 0.6468797564687976,
      "grad_norm": 5.40625,
      "learning_rate": 8.091379382044009e-06,
      "loss": 1.63595314,
      "memory(GiB)": 107.26,
      "step": 25500,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.66399288,
      "epoch": 0.6470065956367326,
      "grad_norm": 6.28125,
      "learning_rate": 8.090555136583172e-06,
      "loss": 1.56749935,
      "memory(GiB)": 107.26,
      "step": 25505,
      "train_speed(iter/s)": 1.635521
    },
    {
      "acc": 0.65745196,
      "epoch": 0.6471334348046677,
      "grad_norm": 5.53125,
      "learning_rate": 8.089730755185921e-06,
      "loss": 1.58977823,
      "memory(GiB)": 107.26,
      "step": 25510,
      "train_speed(iter/s)": 1.635557
    },
    {
      "acc": 0.6484602,
      "epoch": 0.6472602739726028,
      "grad_norm": 5.6875,
      "learning_rate": 8.088906237888517e-06,
      "loss": 1.64411335,
      "memory(GiB)": 107.26,
      "step": 25515,
      "train_speed(iter/s)": 1.635591
    },
    {
      "acc": 0.64755487,
      "epoch": 0.6473871131405378,
      "grad_norm": 4.96875,
      "learning_rate": 8.088081584727228e-06,
      "loss": 1.59233646,
      "memory(GiB)": 107.26,
      "step": 25520,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.64910173,
      "epoch": 0.6475139523084729,
      "grad_norm": 6.90625,
      "learning_rate": 8.08725679573832e-06,
      "loss": 1.60463009,
      "memory(GiB)": 107.26,
      "step": 25525,
      "train_speed(iter/s)": 1.635666
    },
    {
      "acc": 0.66866031,
      "epoch": 0.647640791476408,
      "grad_norm": 5.28125,
      "learning_rate": 8.086431870958078e-06,
      "loss": 1.58421097,
      "memory(GiB)": 107.26,
      "step": 25530,
      "train_speed(iter/s)": 1.635701
    },
    {
      "acc": 0.65110388,
      "epoch": 0.647767630644343,
      "grad_norm": 4.59375,
      "learning_rate": 8.085606810422781e-06,
      "loss": 1.62232876,
      "memory(GiB)": 107.26,
      "step": 25535,
      "train_speed(iter/s)": 1.635739
    },
    {
      "acc": 0.64505992,
      "epoch": 0.647894469812278,
      "grad_norm": 5.78125,
      "learning_rate": 8.08478161416872e-06,
      "loss": 1.68454304,
      "memory(GiB)": 107.26,
      "step": 25540,
      "train_speed(iter/s)": 1.635777
    },
    {
      "acc": 0.65496173,
      "epoch": 0.648021308980213,
      "grad_norm": 5.125,
      "learning_rate": 8.083956282232192e-06,
      "loss": 1.65139961,
      "memory(GiB)": 107.26,
      "step": 25545,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.64949837,
      "epoch": 0.6481481481481481,
      "grad_norm": 6.21875,
      "learning_rate": 8.083130814649498e-06,
      "loss": 1.56567659,
      "memory(GiB)": 107.26,
      "step": 25550,
      "train_speed(iter/s)": 1.635852
    },
    {
      "acc": 0.68030043,
      "epoch": 0.6482749873160832,
      "grad_norm": 5.96875,
      "learning_rate": 8.082305211456943e-06,
      "loss": 1.5074996,
      "memory(GiB)": 107.26,
      "step": 25555,
      "train_speed(iter/s)": 1.635889
    },
    {
      "acc": 0.65378113,
      "epoch": 0.6484018264840182,
      "grad_norm": 5.8125,
      "learning_rate": 8.081479472690846e-06,
      "loss": 1.66485443,
      "memory(GiB)": 107.26,
      "step": 25560,
      "train_speed(iter/s)": 1.635924
    },
    {
      "acc": 0.65303106,
      "epoch": 0.6485286656519533,
      "grad_norm": 4.625,
      "learning_rate": 8.080653598387522e-06,
      "loss": 1.54809875,
      "memory(GiB)": 107.26,
      "step": 25565,
      "train_speed(iter/s)": 1.63596
    },
    {
      "acc": 0.67481833,
      "epoch": 0.6486555048198884,
      "grad_norm": 6.96875,
      "learning_rate": 8.0798275885833e-06,
      "loss": 1.55298853,
      "memory(GiB)": 107.26,
      "step": 25570,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.65893068,
      "epoch": 0.6487823439878234,
      "grad_norm": 5.5625,
      "learning_rate": 8.07900144331451e-06,
      "loss": 1.54101124,
      "memory(GiB)": 107.26,
      "step": 25575,
      "train_speed(iter/s)": 1.636034
    },
    {
      "acc": 0.65369158,
      "epoch": 0.6489091831557585,
      "grad_norm": 6.15625,
      "learning_rate": 8.07817516261749e-06,
      "loss": 1.6125845,
      "memory(GiB)": 107.26,
      "step": 25580,
      "train_speed(iter/s)": 1.63607
    },
    {
      "acc": 0.65587826,
      "epoch": 0.6490360223236935,
      "grad_norm": 5.15625,
      "learning_rate": 8.077348746528583e-06,
      "loss": 1.57133379,
      "memory(GiB)": 107.26,
      "step": 25585,
      "train_speed(iter/s)": 1.636105
    },
    {
      "acc": 0.66055965,
      "epoch": 0.6491628614916286,
      "grad_norm": 5.0625,
      "learning_rate": 8.076522195084139e-06,
      "loss": 1.54116821,
      "memory(GiB)": 107.26,
      "step": 25590,
      "train_speed(iter/s)": 1.636139
    },
    {
      "acc": 0.66508985,
      "epoch": 0.6492897006595637,
      "grad_norm": 4.875,
      "learning_rate": 8.075695508320512e-06,
      "loss": 1.62400665,
      "memory(GiB)": 107.26,
      "step": 25595,
      "train_speed(iter/s)": 1.636173
    },
    {
      "acc": 0.64792643,
      "epoch": 0.6494165398274987,
      "grad_norm": 6.5625,
      "learning_rate": 8.074868686274065e-06,
      "loss": 1.71825066,
      "memory(GiB)": 107.26,
      "step": 25600,
      "train_speed(iter/s)": 1.636209
    },
    {
      "acc": 0.66676998,
      "epoch": 0.6495433789954338,
      "grad_norm": 7.1875,
      "learning_rate": 8.074041728981166e-06,
      "loss": 1.55171442,
      "memory(GiB)": 107.26,
      "step": 25605,
      "train_speed(iter/s)": 1.636243
    },
    {
      "acc": 0.6556746,
      "epoch": 0.6496702181633689,
      "grad_norm": 5.1875,
      "learning_rate": 8.073214636478186e-06,
      "loss": 1.59134932,
      "memory(GiB)": 107.26,
      "step": 25610,
      "train_speed(iter/s)": 1.636279
    },
    {
      "acc": 0.659267,
      "epoch": 0.6497970573313039,
      "grad_norm": 8.1875,
      "learning_rate": 8.072387408801506e-06,
      "loss": 1.63330154,
      "memory(GiB)": 107.26,
      "step": 25615,
      "train_speed(iter/s)": 1.636315
    },
    {
      "acc": 0.64754276,
      "epoch": 0.649923896499239,
      "grad_norm": 5.125,
      "learning_rate": 8.07156004598751e-06,
      "loss": 1.61135445,
      "memory(GiB)": 107.26,
      "step": 25620,
      "train_speed(iter/s)": 1.636351
    },
    {
      "acc": 0.65480671,
      "epoch": 0.650050735667174,
      "grad_norm": 5.34375,
      "learning_rate": 8.07073254807259e-06,
      "loss": 1.59963512,
      "memory(GiB)": 107.26,
      "step": 25625,
      "train_speed(iter/s)": 1.636387
    },
    {
      "acc": 0.65050282,
      "epoch": 0.6501775748351091,
      "grad_norm": 4.96875,
      "learning_rate": 8.069904915093144e-06,
      "loss": 1.62776031,
      "memory(GiB)": 107.26,
      "step": 25630,
      "train_speed(iter/s)": 1.636425
    },
    {
      "acc": 0.6454741,
      "epoch": 0.6503044140030442,
      "grad_norm": 4.75,
      "learning_rate": 8.069077147085571e-06,
      "loss": 1.6130579,
      "memory(GiB)": 107.26,
      "step": 25635,
      "train_speed(iter/s)": 1.636459
    },
    {
      "acc": 0.66796341,
      "epoch": 0.6504312531709792,
      "grad_norm": 5.75,
      "learning_rate": 8.068249244086283e-06,
      "loss": 1.59014292,
      "memory(GiB)": 107.26,
      "step": 25640,
      "train_speed(iter/s)": 1.636495
    },
    {
      "acc": 0.64008842,
      "epoch": 0.6505580923389143,
      "grad_norm": 5.75,
      "learning_rate": 8.067421206131696e-06,
      "loss": 1.6800972,
      "memory(GiB)": 107.26,
      "step": 25645,
      "train_speed(iter/s)": 1.636533
    },
    {
      "acc": 0.6439374,
      "epoch": 0.6506849315068494,
      "grad_norm": 6.0,
      "learning_rate": 8.06659303325823e-06,
      "loss": 1.61153755,
      "memory(GiB)": 107.26,
      "step": 25650,
      "train_speed(iter/s)": 1.636568
    },
    {
      "acc": 0.65181694,
      "epoch": 0.6508117706747844,
      "grad_norm": 5.375,
      "learning_rate": 8.06576472550231e-06,
      "loss": 1.65965424,
      "memory(GiB)": 107.26,
      "step": 25655,
      "train_speed(iter/s)": 1.636605
    },
    {
      "acc": 0.65107126,
      "epoch": 0.6509386098427195,
      "grad_norm": 5.59375,
      "learning_rate": 8.064936282900368e-06,
      "loss": 1.64304657,
      "memory(GiB)": 107.26,
      "step": 25660,
      "train_speed(iter/s)": 1.636639
    },
    {
      "acc": 0.6511466,
      "epoch": 0.6510654490106544,
      "grad_norm": 5.96875,
      "learning_rate": 8.064107705488846e-06,
      "loss": 1.56734219,
      "memory(GiB)": 107.26,
      "step": 25665,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.65623379,
      "epoch": 0.6511922881785895,
      "grad_norm": 5.9375,
      "learning_rate": 8.063278993304188e-06,
      "loss": 1.55760822,
      "memory(GiB)": 107.26,
      "step": 25670,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.64906135,
      "epoch": 0.6513191273465246,
      "grad_norm": 7.34375,
      "learning_rate": 8.06245014638284e-06,
      "loss": 1.62841072,
      "memory(GiB)": 107.26,
      "step": 25675,
      "train_speed(iter/s)": 1.636743
    },
    {
      "acc": 0.66630692,
      "epoch": 0.6514459665144596,
      "grad_norm": 6.53125,
      "learning_rate": 8.061621164761266e-06,
      "loss": 1.53510332,
      "memory(GiB)": 107.26,
      "step": 25680,
      "train_speed(iter/s)": 1.636779
    },
    {
      "acc": 0.65046349,
      "epoch": 0.6515728056823947,
      "grad_norm": 5.4375,
      "learning_rate": 8.06079204847592e-06,
      "loss": 1.57576828,
      "memory(GiB)": 107.26,
      "step": 25685,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.65991106,
      "epoch": 0.6516996448503298,
      "grad_norm": 5.0625,
      "learning_rate": 8.059962797563277e-06,
      "loss": 1.58262768,
      "memory(GiB)": 107.26,
      "step": 25690,
      "train_speed(iter/s)": 1.636849
    },
    {
      "acc": 0.65574341,
      "epoch": 0.6518264840182648,
      "grad_norm": 4.59375,
      "learning_rate": 8.059133412059808e-06,
      "loss": 1.56862288,
      "memory(GiB)": 107.26,
      "step": 25695,
      "train_speed(iter/s)": 1.636886
    },
    {
      "acc": 0.65753093,
      "epoch": 0.6519533231861999,
      "grad_norm": 6.0,
      "learning_rate": 8.058303892001993e-06,
      "loss": 1.59334984,
      "memory(GiB)": 107.26,
      "step": 25700,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.65194035,
      "epoch": 0.6520801623541349,
      "grad_norm": 7.125,
      "learning_rate": 8.057474237426318e-06,
      "loss": 1.59372311,
      "memory(GiB)": 107.26,
      "step": 25705,
      "train_speed(iter/s)": 1.636957
    },
    {
      "acc": 0.6390759,
      "epoch": 0.65220700152207,
      "grad_norm": 5.46875,
      "learning_rate": 8.056644448369275e-06,
      "loss": 1.68950539,
      "memory(GiB)": 107.26,
      "step": 25710,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.64411707,
      "epoch": 0.6523338406900051,
      "grad_norm": 4.875,
      "learning_rate": 8.055814524867364e-06,
      "loss": 1.66304226,
      "memory(GiB)": 107.26,
      "step": 25715,
      "train_speed(iter/s)": 1.637028
    },
    {
      "acc": 0.64946504,
      "epoch": 0.6524606798579401,
      "grad_norm": 6.28125,
      "learning_rate": 8.054984466957085e-06,
      "loss": 1.67126904,
      "memory(GiB)": 107.26,
      "step": 25720,
      "train_speed(iter/s)": 1.637063
    },
    {
      "acc": 0.66729188,
      "epoch": 0.6525875190258752,
      "grad_norm": 5.375,
      "learning_rate": 8.05415427467495e-06,
      "loss": 1.47327223,
      "memory(GiB)": 107.26,
      "step": 25725,
      "train_speed(iter/s)": 1.637101
    },
    {
      "acc": 0.65841341,
      "epoch": 0.6527143581938103,
      "grad_norm": 5.1875,
      "learning_rate": 8.053323948057477e-06,
      "loss": 1.60801105,
      "memory(GiB)": 107.26,
      "step": 25730,
      "train_speed(iter/s)": 1.637136
    },
    {
      "acc": 0.64222345,
      "epoch": 0.6528411973617453,
      "grad_norm": 5.5625,
      "learning_rate": 8.052493487141183e-06,
      "loss": 1.64245872,
      "memory(GiB)": 107.26,
      "step": 25735,
      "train_speed(iter/s)": 1.637169
    },
    {
      "acc": 0.64860888,
      "epoch": 0.6529680365296804,
      "grad_norm": 5.09375,
      "learning_rate": 8.051662891962594e-06,
      "loss": 1.62688332,
      "memory(GiB)": 107.26,
      "step": 25740,
      "train_speed(iter/s)": 1.637206
    },
    {
      "acc": 0.65567026,
      "epoch": 0.6530948756976154,
      "grad_norm": 5.21875,
      "learning_rate": 8.05083216255825e-06,
      "loss": 1.61858711,
      "memory(GiB)": 107.26,
      "step": 25745,
      "train_speed(iter/s)": 1.63724
    },
    {
      "acc": 0.65512767,
      "epoch": 0.6532217148655505,
      "grad_norm": 5.0625,
      "learning_rate": 8.050001298964685e-06,
      "loss": 1.63614407,
      "memory(GiB)": 107.26,
      "step": 25750,
      "train_speed(iter/s)": 1.637274
    },
    {
      "acc": 0.66472425,
      "epoch": 0.6533485540334856,
      "grad_norm": 5.8125,
      "learning_rate": 8.049170301218445e-06,
      "loss": 1.62809258,
      "memory(GiB)": 107.26,
      "step": 25755,
      "train_speed(iter/s)": 1.637309
    },
    {
      "acc": 0.66269274,
      "epoch": 0.6534753932014206,
      "grad_norm": 6.84375,
      "learning_rate": 8.048339169356085e-06,
      "loss": 1.59124031,
      "memory(GiB)": 107.26,
      "step": 25760,
      "train_speed(iter/s)": 1.637346
    },
    {
      "acc": 0.67250175,
      "epoch": 0.6536022323693557,
      "grad_norm": 5.78125,
      "learning_rate": 8.047507903414155e-06,
      "loss": 1.52040024,
      "memory(GiB)": 107.26,
      "step": 25765,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.67127681,
      "epoch": 0.6537290715372908,
      "grad_norm": 5.25,
      "learning_rate": 8.046676503429222e-06,
      "loss": 1.54010944,
      "memory(GiB)": 107.26,
      "step": 25770,
      "train_speed(iter/s)": 1.637417
    },
    {
      "acc": 0.65577064,
      "epoch": 0.6538559107052258,
      "grad_norm": 5.9375,
      "learning_rate": 8.045844969437855e-06,
      "loss": 1.5808691,
      "memory(GiB)": 107.26,
      "step": 25775,
      "train_speed(iter/s)": 1.637452
    },
    {
      "acc": 0.64445257,
      "epoch": 0.6539827498731609,
      "grad_norm": 7.03125,
      "learning_rate": 8.045013301476625e-06,
      "loss": 1.61710358,
      "memory(GiB)": 107.26,
      "step": 25780,
      "train_speed(iter/s)": 1.637488
    },
    {
      "acc": 0.6572897,
      "epoch": 0.6541095890410958,
      "grad_norm": 4.59375,
      "learning_rate": 8.044181499582117e-06,
      "loss": 1.61221962,
      "memory(GiB)": 107.26,
      "step": 25785,
      "train_speed(iter/s)": 1.637526
    },
    {
      "acc": 0.66239758,
      "epoch": 0.6542364282090309,
      "grad_norm": 5.59375,
      "learning_rate": 8.043349563790917e-06,
      "loss": 1.61270027,
      "memory(GiB)": 107.26,
      "step": 25790,
      "train_speed(iter/s)": 1.63756
    },
    {
      "acc": 0.65623088,
      "epoch": 0.654363267376966,
      "grad_norm": 5.28125,
      "learning_rate": 8.042517494139612e-06,
      "loss": 1.64167175,
      "memory(GiB)": 107.26,
      "step": 25795,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.65754857,
      "epoch": 0.654490106544901,
      "grad_norm": 4.90625,
      "learning_rate": 8.041685290664806e-06,
      "loss": 1.52577028,
      "memory(GiB)": 107.26,
      "step": 25800,
      "train_speed(iter/s)": 1.63763
    },
    {
      "acc": 0.64661393,
      "epoch": 0.6546169457128361,
      "grad_norm": 5.71875,
      "learning_rate": 8.0408529534031e-06,
      "loss": 1.60060444,
      "memory(GiB)": 107.26,
      "step": 25805,
      "train_speed(iter/s)": 1.637665
    },
    {
      "acc": 0.64958339,
      "epoch": 0.6547437848807712,
      "grad_norm": 4.90625,
      "learning_rate": 8.040020482391105e-06,
      "loss": 1.58185863,
      "memory(GiB)": 107.26,
      "step": 25810,
      "train_speed(iter/s)": 1.637702
    },
    {
      "acc": 0.68604307,
      "epoch": 0.6548706240487062,
      "grad_norm": 5.625,
      "learning_rate": 8.039187877665435e-06,
      "loss": 1.49172001,
      "memory(GiB)": 107.26,
      "step": 25815,
      "train_speed(iter/s)": 1.637739
    },
    {
      "acc": 0.66448898,
      "epoch": 0.6549974632166413,
      "grad_norm": 9.0,
      "learning_rate": 8.038355139262716e-06,
      "loss": 1.56033497,
      "memory(GiB)": 107.26,
      "step": 25820,
      "train_speed(iter/s)": 1.637773
    },
    {
      "acc": 0.66828508,
      "epoch": 0.6551243023845763,
      "grad_norm": 6.0625,
      "learning_rate": 8.037522267219571e-06,
      "loss": 1.57431402,
      "memory(GiB)": 107.26,
      "step": 25825,
      "train_speed(iter/s)": 1.637806
    },
    {
      "acc": 0.64303837,
      "epoch": 0.6552511415525114,
      "grad_norm": 7.1875,
      "learning_rate": 8.036689261572636e-06,
      "loss": 1.61256237,
      "memory(GiB)": 107.26,
      "step": 25830,
      "train_speed(iter/s)": 1.637841
    },
    {
      "acc": 0.64496369,
      "epoch": 0.6553779807204465,
      "grad_norm": 6.71875,
      "learning_rate": 8.035856122358548e-06,
      "loss": 1.66227551,
      "memory(GiB)": 107.26,
      "step": 25835,
      "train_speed(iter/s)": 1.637877
    },
    {
      "acc": 0.65122848,
      "epoch": 0.6555048198883815,
      "grad_norm": 5.21875,
      "learning_rate": 8.035022849613954e-06,
      "loss": 1.63981819,
      "memory(GiB)": 107.26,
      "step": 25840,
      "train_speed(iter/s)": 1.637914
    },
    {
      "acc": 0.65888414,
      "epoch": 0.6556316590563166,
      "grad_norm": 6.40625,
      "learning_rate": 8.034189443375505e-06,
      "loss": 1.64588718,
      "memory(GiB)": 107.26,
      "step": 25845,
      "train_speed(iter/s)": 1.63795
    },
    {
      "acc": 0.65115261,
      "epoch": 0.6557584982242517,
      "grad_norm": 5.34375,
      "learning_rate": 8.033355903679858e-06,
      "loss": 1.5615099,
      "memory(GiB)": 107.26,
      "step": 25850,
      "train_speed(iter/s)": 1.637988
    },
    {
      "acc": 0.66259556,
      "epoch": 0.6558853373921867,
      "grad_norm": 5.90625,
      "learning_rate": 8.032522230563676e-06,
      "loss": 1.61838951,
      "memory(GiB)": 107.26,
      "step": 25855,
      "train_speed(iter/s)": 1.638023
    },
    {
      "acc": 0.65130081,
      "epoch": 0.6560121765601218,
      "grad_norm": 6.65625,
      "learning_rate": 8.031688424063625e-06,
      "loss": 1.57806177,
      "memory(GiB)": 107.26,
      "step": 25860,
      "train_speed(iter/s)": 1.638058
    },
    {
      "acc": 0.65015593,
      "epoch": 0.6561390157280568,
      "grad_norm": 5.8125,
      "learning_rate": 8.030854484216381e-06,
      "loss": 1.62416039,
      "memory(GiB)": 107.26,
      "step": 25865,
      "train_speed(iter/s)": 1.638095
    },
    {
      "acc": 0.65527468,
      "epoch": 0.6562658548959919,
      "grad_norm": 4.78125,
      "learning_rate": 8.030020411058627e-06,
      "loss": 1.66755371,
      "memory(GiB)": 107.26,
      "step": 25870,
      "train_speed(iter/s)": 1.63813
    },
    {
      "acc": 0.66531949,
      "epoch": 0.656392694063927,
      "grad_norm": 5.875,
      "learning_rate": 8.029186204627049e-06,
      "loss": 1.57561426,
      "memory(GiB)": 107.26,
      "step": 25875,
      "train_speed(iter/s)": 1.638166
    },
    {
      "acc": 0.65105863,
      "epoch": 0.656519533231862,
      "grad_norm": 7.3125,
      "learning_rate": 8.028351864958335e-06,
      "loss": 1.59731588,
      "memory(GiB)": 107.26,
      "step": 25880,
      "train_speed(iter/s)": 1.638203
    },
    {
      "acc": 0.67144046,
      "epoch": 0.6566463723997971,
      "grad_norm": 6.40625,
      "learning_rate": 8.027517392089185e-06,
      "loss": 1.55755978,
      "memory(GiB)": 107.26,
      "step": 25885,
      "train_speed(iter/s)": 1.63824
    },
    {
      "acc": 0.65607567,
      "epoch": 0.6567732115677322,
      "grad_norm": 6.0,
      "learning_rate": 8.026682786056304e-06,
      "loss": 1.62314148,
      "memory(GiB)": 107.26,
      "step": 25890,
      "train_speed(iter/s)": 1.638278
    },
    {
      "acc": 0.63641014,
      "epoch": 0.6569000507356672,
      "grad_norm": 4.5625,
      "learning_rate": 8.025848046896401e-06,
      "loss": 1.69427681,
      "memory(GiB)": 107.26,
      "step": 25895,
      "train_speed(iter/s)": 1.638316
    },
    {
      "acc": 0.65331173,
      "epoch": 0.6570268899036023,
      "grad_norm": 5.75,
      "learning_rate": 8.02501317464619e-06,
      "loss": 1.61411819,
      "memory(GiB)": 107.26,
      "step": 25900,
      "train_speed(iter/s)": 1.638353
    },
    {
      "acc": 0.65533915,
      "epoch": 0.6571537290715372,
      "grad_norm": 7.125,
      "learning_rate": 8.024178169342396e-06,
      "loss": 1.6569809,
      "memory(GiB)": 107.26,
      "step": 25905,
      "train_speed(iter/s)": 1.638388
    },
    {
      "acc": 0.65819635,
      "epoch": 0.6572805682394723,
      "grad_norm": 5.8125,
      "learning_rate": 8.023343031021744e-06,
      "loss": 1.6106245,
      "memory(GiB)": 107.26,
      "step": 25910,
      "train_speed(iter/s)": 1.638423
    },
    {
      "acc": 0.64963164,
      "epoch": 0.6574074074074074,
      "grad_norm": 5.0,
      "learning_rate": 8.022507759720966e-06,
      "loss": 1.5897171,
      "memory(GiB)": 107.26,
      "step": 25915,
      "train_speed(iter/s)": 1.638461
    },
    {
      "acc": 0.65574932,
      "epoch": 0.6575342465753424,
      "grad_norm": 5.6875,
      "learning_rate": 8.021672355476802e-06,
      "loss": 1.63314705,
      "memory(GiB)": 107.26,
      "step": 25920,
      "train_speed(iter/s)": 1.638497
    },
    {
      "acc": 0.6544919,
      "epoch": 0.6576610857432775,
      "grad_norm": 5.3125,
      "learning_rate": 8.020836818325997e-06,
      "loss": 1.58537827,
      "memory(GiB)": 107.26,
      "step": 25925,
      "train_speed(iter/s)": 1.638533
    },
    {
      "acc": 0.65422497,
      "epoch": 0.6577879249112126,
      "grad_norm": 7.71875,
      "learning_rate": 8.020001148305304e-06,
      "loss": 1.59328852,
      "memory(GiB)": 107.26,
      "step": 25930,
      "train_speed(iter/s)": 1.638568
    },
    {
      "acc": 0.67387271,
      "epoch": 0.6579147640791476,
      "grad_norm": 5.9375,
      "learning_rate": 8.019165345451475e-06,
      "loss": 1.56165943,
      "memory(GiB)": 107.26,
      "step": 25935,
      "train_speed(iter/s)": 1.638602
    },
    {
      "acc": 0.66296759,
      "epoch": 0.6580416032470827,
      "grad_norm": 6.84375,
      "learning_rate": 8.018329409801276e-06,
      "loss": 1.59441996,
      "memory(GiB)": 107.26,
      "step": 25940,
      "train_speed(iter/s)": 1.63864
    },
    {
      "acc": 0.65683632,
      "epoch": 0.6581684424150177,
      "grad_norm": 8.5625,
      "learning_rate": 8.017493341391471e-06,
      "loss": 1.58436546,
      "memory(GiB)": 107.26,
      "step": 25945,
      "train_speed(iter/s)": 1.638674
    },
    {
      "acc": 0.65543962,
      "epoch": 0.6582952815829528,
      "grad_norm": 5.28125,
      "learning_rate": 8.016657140258839e-06,
      "loss": 1.5694334,
      "memory(GiB)": 107.26,
      "step": 25950,
      "train_speed(iter/s)": 1.638709
    },
    {
      "acc": 0.65649805,
      "epoch": 0.6584221207508879,
      "grad_norm": 5.84375,
      "learning_rate": 8.015820806440157e-06,
      "loss": 1.62765369,
      "memory(GiB)": 107.26,
      "step": 25955,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.63261914,
      "epoch": 0.6585489599188229,
      "grad_norm": 5.875,
      "learning_rate": 8.014984339972211e-06,
      "loss": 1.62459164,
      "memory(GiB)": 107.26,
      "step": 25960,
      "train_speed(iter/s)": 1.638782
    },
    {
      "acc": 0.65641041,
      "epoch": 0.658675799086758,
      "grad_norm": 5.21875,
      "learning_rate": 8.014147740891793e-06,
      "loss": 1.61278076,
      "memory(GiB)": 107.26,
      "step": 25965,
      "train_speed(iter/s)": 1.638816
    },
    {
      "acc": 0.65521011,
      "epoch": 0.6588026382546931,
      "grad_norm": 7.0,
      "learning_rate": 8.0133110092357e-06,
      "loss": 1.56442394,
      "memory(GiB)": 107.26,
      "step": 25970,
      "train_speed(iter/s)": 1.638852
    },
    {
      "acc": 0.66287298,
      "epoch": 0.6589294774226281,
      "grad_norm": 5.5625,
      "learning_rate": 8.012474145040737e-06,
      "loss": 1.66549492,
      "memory(GiB)": 107.26,
      "step": 25975,
      "train_speed(iter/s)": 1.638888
    },
    {
      "acc": 0.65622835,
      "epoch": 0.6590563165905632,
      "grad_norm": 5.71875,
      "learning_rate": 8.01163714834371e-06,
      "loss": 1.62561264,
      "memory(GiB)": 107.26,
      "step": 25980,
      "train_speed(iter/s)": 1.638921
    },
    {
      "acc": 0.65145273,
      "epoch": 0.6591831557584982,
      "grad_norm": 6.34375,
      "learning_rate": 8.010800019181433e-06,
      "loss": 1.68249474,
      "memory(GiB)": 107.26,
      "step": 25985,
      "train_speed(iter/s)": 1.638959
    },
    {
      "acc": 0.64840899,
      "epoch": 0.6593099949264333,
      "grad_norm": 5.84375,
      "learning_rate": 8.009962757590732e-06,
      "loss": 1.6307621,
      "memory(GiB)": 107.26,
      "step": 25990,
      "train_speed(iter/s)": 1.638993
    },
    {
      "acc": 0.65632353,
      "epoch": 0.6594368340943684,
      "grad_norm": 5.09375,
      "learning_rate": 8.00912536360843e-06,
      "loss": 1.6047554,
      "memory(GiB)": 107.26,
      "step": 25995,
      "train_speed(iter/s)": 1.63903
    },
    {
      "acc": 0.65990715,
      "epoch": 0.6595636732623034,
      "grad_norm": 6.21875,
      "learning_rate": 8.008287837271359e-06,
      "loss": 1.56774426,
      "memory(GiB)": 107.26,
      "step": 26000,
      "train_speed(iter/s)": 1.639064
    },
    {
      "epoch": 0.6595636732623034,
      "eval_acc": 0.6449340826175665,
      "eval_loss": 1.5801098346710205,
      "eval_runtime": 58.8201,
      "eval_samples_per_second": 108.296,
      "eval_steps_per_second": 27.083,
      "step": 26000
    },
    {
      "acc": 0.66742239,
      "epoch": 0.6596905124302385,
      "grad_norm": 5.8125,
      "learning_rate": 8.007450178616356e-06,
      "loss": 1.5487195,
      "memory(GiB)": 107.26,
      "step": 26005,
      "train_speed(iter/s)": 1.632609
    },
    {
      "acc": 0.64972801,
      "epoch": 0.6598173515981736,
      "grad_norm": 5.46875,
      "learning_rate": 8.00661238768027e-06,
      "loss": 1.59287224,
      "memory(GiB)": 107.26,
      "step": 26010,
      "train_speed(iter/s)": 1.632642
    },
    {
      "acc": 0.65350962,
      "epoch": 0.6599441907661086,
      "grad_norm": 6.25,
      "learning_rate": 8.005774464499947e-06,
      "loss": 1.6239933,
      "memory(GiB)": 107.26,
      "step": 26015,
      "train_speed(iter/s)": 1.632676
    },
    {
      "acc": 0.63990555,
      "epoch": 0.6600710299340437,
      "grad_norm": 5.15625,
      "learning_rate": 8.004936409112243e-06,
      "loss": 1.63208122,
      "memory(GiB)": 107.26,
      "step": 26020,
      "train_speed(iter/s)": 1.632711
    },
    {
      "acc": 0.65411482,
      "epoch": 0.6601978691019786,
      "grad_norm": 5.09375,
      "learning_rate": 8.004098221554018e-06,
      "loss": 1.61860123,
      "memory(GiB)": 107.26,
      "step": 26025,
      "train_speed(iter/s)": 1.632743
    },
    {
      "acc": 0.6635293,
      "epoch": 0.6603247082699137,
      "grad_norm": 4.5625,
      "learning_rate": 8.003259901862143e-06,
      "loss": 1.59302425,
      "memory(GiB)": 107.26,
      "step": 26030,
      "train_speed(iter/s)": 1.632776
    },
    {
      "acc": 0.64447856,
      "epoch": 0.6604515474378488,
      "grad_norm": 5.46875,
      "learning_rate": 8.002421450073488e-06,
      "loss": 1.57079191,
      "memory(GiB)": 107.26,
      "step": 26035,
      "train_speed(iter/s)": 1.63281
    },
    {
      "acc": 0.65432482,
      "epoch": 0.6605783866057838,
      "grad_norm": 6.15625,
      "learning_rate": 8.001582866224932e-06,
      "loss": 1.56841364,
      "memory(GiB)": 107.26,
      "step": 26040,
      "train_speed(iter/s)": 1.632844
    },
    {
      "acc": 0.66357298,
      "epoch": 0.6607052257737189,
      "grad_norm": 6.21875,
      "learning_rate": 8.000744150353362e-06,
      "loss": 1.63216171,
      "memory(GiB)": 107.26,
      "step": 26045,
      "train_speed(iter/s)": 1.632876
    },
    {
      "acc": 0.65270128,
      "epoch": 0.660832064941654,
      "grad_norm": 5.46875,
      "learning_rate": 7.999905302495667e-06,
      "loss": 1.64064026,
      "memory(GiB)": 107.26,
      "step": 26050,
      "train_speed(iter/s)": 1.632909
    },
    {
      "acc": 0.66189537,
      "epoch": 0.660958904109589,
      "grad_norm": 6.03125,
      "learning_rate": 7.999066322688743e-06,
      "loss": 1.59782085,
      "memory(GiB)": 107.26,
      "step": 26055,
      "train_speed(iter/s)": 1.632944
    },
    {
      "acc": 0.6538331,
      "epoch": 0.6610857432775241,
      "grad_norm": 6.15625,
      "learning_rate": 7.998227210969491e-06,
      "loss": 1.67116566,
      "memory(GiB)": 107.26,
      "step": 26060,
      "train_speed(iter/s)": 1.632976
    },
    {
      "acc": 0.66555872,
      "epoch": 0.6612125824454591,
      "grad_norm": 5.5,
      "learning_rate": 7.997387967374821e-06,
      "loss": 1.56799259,
      "memory(GiB)": 107.26,
      "step": 26065,
      "train_speed(iter/s)": 1.63301
    },
    {
      "acc": 0.6442678,
      "epoch": 0.6613394216133942,
      "grad_norm": 6.71875,
      "learning_rate": 7.996548591941647e-06,
      "loss": 1.68439751,
      "memory(GiB)": 107.26,
      "step": 26070,
      "train_speed(iter/s)": 1.633044
    },
    {
      "acc": 0.65063286,
      "epoch": 0.6614662607813293,
      "grad_norm": 6.75,
      "learning_rate": 7.995709084706884e-06,
      "loss": 1.59219618,
      "memory(GiB)": 107.26,
      "step": 26075,
      "train_speed(iter/s)": 1.633077
    },
    {
      "acc": 0.63879976,
      "epoch": 0.6615930999492643,
      "grad_norm": 5.5,
      "learning_rate": 7.994869445707463e-06,
      "loss": 1.64366302,
      "memory(GiB)": 107.26,
      "step": 26080,
      "train_speed(iter/s)": 1.63311
    },
    {
      "acc": 0.65073891,
      "epoch": 0.6617199391171994,
      "grad_norm": 5.4375,
      "learning_rate": 7.994029674980313e-06,
      "loss": 1.59470654,
      "memory(GiB)": 107.26,
      "step": 26085,
      "train_speed(iter/s)": 1.633143
    },
    {
      "acc": 0.66801548,
      "epoch": 0.6618467782851345,
      "grad_norm": 5.0,
      "learning_rate": 7.99318977256237e-06,
      "loss": 1.56970501,
      "memory(GiB)": 107.26,
      "step": 26090,
      "train_speed(iter/s)": 1.633177
    },
    {
      "acc": 0.64988208,
      "epoch": 0.6619736174530695,
      "grad_norm": 4.3125,
      "learning_rate": 7.992349738490576e-06,
      "loss": 1.63021088,
      "memory(GiB)": 107.26,
      "step": 26095,
      "train_speed(iter/s)": 1.633204
    },
    {
      "acc": 0.66828499,
      "epoch": 0.6621004566210046,
      "grad_norm": 6.40625,
      "learning_rate": 7.991509572801883e-06,
      "loss": 1.58921242,
      "memory(GiB)": 107.26,
      "step": 26100,
      "train_speed(iter/s)": 1.633234
    },
    {
      "acc": 0.65110073,
      "epoch": 0.6622272957889396,
      "grad_norm": 4.90625,
      "learning_rate": 7.990669275533241e-06,
      "loss": 1.6571209,
      "memory(GiB)": 107.26,
      "step": 26105,
      "train_speed(iter/s)": 1.633266
    },
    {
      "acc": 0.64173956,
      "epoch": 0.6623541349568747,
      "grad_norm": 5.8125,
      "learning_rate": 7.989828846721613e-06,
      "loss": 1.65434036,
      "memory(GiB)": 107.26,
      "step": 26110,
      "train_speed(iter/s)": 1.633301
    },
    {
      "acc": 0.64650497,
      "epoch": 0.6624809741248098,
      "grad_norm": 5.1875,
      "learning_rate": 7.98898828640396e-06,
      "loss": 1.60591927,
      "memory(GiB)": 107.26,
      "step": 26115,
      "train_speed(iter/s)": 1.633327
    },
    {
      "acc": 0.65101547,
      "epoch": 0.6626078132927448,
      "grad_norm": 6.03125,
      "learning_rate": 7.988147594617262e-06,
      "loss": 1.61112423,
      "memory(GiB)": 107.26,
      "step": 26120,
      "train_speed(iter/s)": 1.633361
    },
    {
      "acc": 0.65958891,
      "epoch": 0.6627346524606799,
      "grad_norm": 4.96875,
      "learning_rate": 7.987306771398489e-06,
      "loss": 1.55805855,
      "memory(GiB)": 107.26,
      "step": 26125,
      "train_speed(iter/s)": 1.633393
    },
    {
      "acc": 0.67060671,
      "epoch": 0.662861491628615,
      "grad_norm": 5.9375,
      "learning_rate": 7.986465816784628e-06,
      "loss": 1.60093422,
      "memory(GiB)": 107.26,
      "step": 26130,
      "train_speed(iter/s)": 1.633426
    },
    {
      "acc": 0.65147324,
      "epoch": 0.66298833079655,
      "grad_norm": 5.90625,
      "learning_rate": 7.985624730812667e-06,
      "loss": 1.61662483,
      "memory(GiB)": 107.26,
      "step": 26135,
      "train_speed(iter/s)": 1.633459
    },
    {
      "acc": 0.64554853,
      "epoch": 0.663115169964485,
      "grad_norm": 6.21875,
      "learning_rate": 7.984783513519601e-06,
      "loss": 1.64859066,
      "memory(GiB)": 107.26,
      "step": 26140,
      "train_speed(iter/s)": 1.633494
    },
    {
      "acc": 0.64900913,
      "epoch": 0.66324200913242,
      "grad_norm": 6.0625,
      "learning_rate": 7.98394216494243e-06,
      "loss": 1.68912888,
      "memory(GiB)": 107.26,
      "step": 26145,
      "train_speed(iter/s)": 1.633526
    },
    {
      "acc": 0.65496206,
      "epoch": 0.6633688483003551,
      "grad_norm": 5.875,
      "learning_rate": 7.983100685118157e-06,
      "loss": 1.65027428,
      "memory(GiB)": 107.26,
      "step": 26150,
      "train_speed(iter/s)": 1.633558
    },
    {
      "acc": 0.6627738,
      "epoch": 0.6634956874682902,
      "grad_norm": 5.21875,
      "learning_rate": 7.9822590740838e-06,
      "loss": 1.53033924,
      "memory(GiB)": 107.26,
      "step": 26155,
      "train_speed(iter/s)": 1.633593
    },
    {
      "acc": 0.65780592,
      "epoch": 0.6636225266362252,
      "grad_norm": 5.21875,
      "learning_rate": 7.981417331876373e-06,
      "loss": 1.55905523,
      "memory(GiB)": 107.26,
      "step": 26160,
      "train_speed(iter/s)": 1.633624
    },
    {
      "acc": 0.65119009,
      "epoch": 0.6637493658041603,
      "grad_norm": 4.96875,
      "learning_rate": 7.980575458532901e-06,
      "loss": 1.65087109,
      "memory(GiB)": 107.26,
      "step": 26165,
      "train_speed(iter/s)": 1.633658
    },
    {
      "acc": 0.65455079,
      "epoch": 0.6638762049720954,
      "grad_norm": 5.34375,
      "learning_rate": 7.979733454090415e-06,
      "loss": 1.60217419,
      "memory(GiB)": 107.26,
      "step": 26170,
      "train_speed(iter/s)": 1.633693
    },
    {
      "acc": 0.64061966,
      "epoch": 0.6640030441400304,
      "grad_norm": 6.375,
      "learning_rate": 7.978891318585947e-06,
      "loss": 1.59168797,
      "memory(GiB)": 107.26,
      "step": 26175,
      "train_speed(iter/s)": 1.633726
    },
    {
      "acc": 0.65740409,
      "epoch": 0.6641298833079655,
      "grad_norm": 6.3125,
      "learning_rate": 7.978049052056537e-06,
      "loss": 1.62550316,
      "memory(GiB)": 107.26,
      "step": 26180,
      "train_speed(iter/s)": 1.633761
    },
    {
      "acc": 0.6493063,
      "epoch": 0.6642567224759005,
      "grad_norm": 6.0625,
      "learning_rate": 7.977206654539235e-06,
      "loss": 1.60285969,
      "memory(GiB)": 107.26,
      "step": 26185,
      "train_speed(iter/s)": 1.633797
    },
    {
      "acc": 0.64898381,
      "epoch": 0.6643835616438356,
      "grad_norm": 5.21875,
      "learning_rate": 7.976364126071092e-06,
      "loss": 1.65029259,
      "memory(GiB)": 107.26,
      "step": 26190,
      "train_speed(iter/s)": 1.633833
    },
    {
      "acc": 0.62990637,
      "epoch": 0.6645104008117707,
      "grad_norm": 6.03125,
      "learning_rate": 7.975521466689166e-06,
      "loss": 1.7216938,
      "memory(GiB)": 107.26,
      "step": 26195,
      "train_speed(iter/s)": 1.633869
    },
    {
      "acc": 0.64958706,
      "epoch": 0.6646372399797057,
      "grad_norm": 7.3125,
      "learning_rate": 7.974678676430523e-06,
      "loss": 1.59830379,
      "memory(GiB)": 107.26,
      "step": 26200,
      "train_speed(iter/s)": 1.633902
    },
    {
      "acc": 0.66740875,
      "epoch": 0.6647640791476408,
      "grad_norm": 5.46875,
      "learning_rate": 7.97383575533223e-06,
      "loss": 1.53848362,
      "memory(GiB)": 107.26,
      "step": 26205,
      "train_speed(iter/s)": 1.633936
    },
    {
      "acc": 0.64491701,
      "epoch": 0.6648909183155759,
      "grad_norm": 5.4375,
      "learning_rate": 7.972992703431362e-06,
      "loss": 1.61117287,
      "memory(GiB)": 107.26,
      "step": 26210,
      "train_speed(iter/s)": 1.63397
    },
    {
      "acc": 0.65159121,
      "epoch": 0.6650177574835109,
      "grad_norm": 5.8125,
      "learning_rate": 7.972149520765e-06,
      "loss": 1.62900295,
      "memory(GiB)": 107.26,
      "step": 26215,
      "train_speed(iter/s)": 1.634006
    },
    {
      "acc": 0.64747009,
      "epoch": 0.665144596651446,
      "grad_norm": 5.84375,
      "learning_rate": 7.971306207370236e-06,
      "loss": 1.62082176,
      "memory(GiB)": 107.26,
      "step": 26220,
      "train_speed(iter/s)": 1.63404
    },
    {
      "acc": 0.65735497,
      "epoch": 0.665271435819381,
      "grad_norm": 5.4375,
      "learning_rate": 7.970462763284157e-06,
      "loss": 1.59599295,
      "memory(GiB)": 107.26,
      "step": 26225,
      "train_speed(iter/s)": 1.634074
    },
    {
      "acc": 0.65048742,
      "epoch": 0.6653982749873161,
      "grad_norm": 5.6875,
      "learning_rate": 7.969619188543865e-06,
      "loss": 1.66062508,
      "memory(GiB)": 107.26,
      "step": 26230,
      "train_speed(iter/s)": 1.63411
    },
    {
      "acc": 0.65688505,
      "epoch": 0.6655251141552512,
      "grad_norm": 6.25,
      "learning_rate": 7.968775483186462e-06,
      "loss": 1.5941186,
      "memory(GiB)": 107.26,
      "step": 26235,
      "train_speed(iter/s)": 1.634142
    },
    {
      "acc": 0.67182746,
      "epoch": 0.6656519533231862,
      "grad_norm": 5.46875,
      "learning_rate": 7.967931647249058e-06,
      "loss": 1.54654465,
      "memory(GiB)": 107.26,
      "step": 26240,
      "train_speed(iter/s)": 1.634176
    },
    {
      "acc": 0.6422019,
      "epoch": 0.6657787924911213,
      "grad_norm": 6.40625,
      "learning_rate": 7.967087680768768e-06,
      "loss": 1.609688,
      "memory(GiB)": 107.26,
      "step": 26245,
      "train_speed(iter/s)": 1.63421
    },
    {
      "acc": 0.65624475,
      "epoch": 0.6659056316590564,
      "grad_norm": 5.8125,
      "learning_rate": 7.966243583782718e-06,
      "loss": 1.6517086,
      "memory(GiB)": 107.26,
      "step": 26250,
      "train_speed(iter/s)": 1.634245
    },
    {
      "acc": 0.64523468,
      "epoch": 0.6660324708269914,
      "grad_norm": 5.25,
      "learning_rate": 7.96539935632803e-06,
      "loss": 1.62207298,
      "memory(GiB)": 107.26,
      "step": 26255,
      "train_speed(iter/s)": 1.63428
    },
    {
      "acc": 0.65360136,
      "epoch": 0.6661593099949265,
      "grad_norm": 5.75,
      "learning_rate": 7.964554998441839e-06,
      "loss": 1.59817991,
      "memory(GiB)": 107.26,
      "step": 26260,
      "train_speed(iter/s)": 1.634315
    },
    {
      "acc": 0.66992517,
      "epoch": 0.6662861491628614,
      "grad_norm": 5.4375,
      "learning_rate": 7.963710510161282e-06,
      "loss": 1.57187328,
      "memory(GiB)": 107.26,
      "step": 26265,
      "train_speed(iter/s)": 1.634349
    },
    {
      "acc": 0.67026062,
      "epoch": 0.6664129883307965,
      "grad_norm": 5.21875,
      "learning_rate": 7.962865891523508e-06,
      "loss": 1.55525513,
      "memory(GiB)": 107.26,
      "step": 26270,
      "train_speed(iter/s)": 1.634382
    },
    {
      "acc": 0.65087891,
      "epoch": 0.6665398274987316,
      "grad_norm": 6.0625,
      "learning_rate": 7.96202114256566e-06,
      "loss": 1.70106716,
      "memory(GiB)": 107.26,
      "step": 26275,
      "train_speed(iter/s)": 1.634417
    },
    {
      "acc": 0.65026321,
      "epoch": 0.6666666666666666,
      "grad_norm": 5.40625,
      "learning_rate": 7.961176263324902e-06,
      "loss": 1.60736542,
      "memory(GiB)": 107.26,
      "step": 26280,
      "train_speed(iter/s)": 1.63445
    },
    {
      "acc": 0.64772739,
      "epoch": 0.6667935058346017,
      "grad_norm": 5.21875,
      "learning_rate": 7.960331253838387e-06,
      "loss": 1.59110584,
      "memory(GiB)": 107.26,
      "step": 26285,
      "train_speed(iter/s)": 1.634483
    },
    {
      "acc": 0.64581356,
      "epoch": 0.6669203450025368,
      "grad_norm": 4.75,
      "learning_rate": 7.95948611414329e-06,
      "loss": 1.58659763,
      "memory(GiB)": 107.26,
      "step": 26290,
      "train_speed(iter/s)": 1.634518
    },
    {
      "acc": 0.65254569,
      "epoch": 0.6670471841704718,
      "grad_norm": 6.21875,
      "learning_rate": 7.958640844276776e-06,
      "loss": 1.65601578,
      "memory(GiB)": 107.26,
      "step": 26295,
      "train_speed(iter/s)": 1.634553
    },
    {
      "acc": 0.64018664,
      "epoch": 0.6671740233384069,
      "grad_norm": 5.25,
      "learning_rate": 7.957795444276033e-06,
      "loss": 1.65617447,
      "memory(GiB)": 107.26,
      "step": 26300,
      "train_speed(iter/s)": 1.634587
    },
    {
      "acc": 0.65503931,
      "epoch": 0.6673008625063419,
      "grad_norm": 6.46875,
      "learning_rate": 7.956949914178239e-06,
      "loss": 1.56485348,
      "memory(GiB)": 107.26,
      "step": 26305,
      "train_speed(iter/s)": 1.634622
    },
    {
      "acc": 0.65427999,
      "epoch": 0.667427701674277,
      "grad_norm": 6.5625,
      "learning_rate": 7.956104254020587e-06,
      "loss": 1.56067181,
      "memory(GiB)": 107.26,
      "step": 26310,
      "train_speed(iter/s)": 1.634658
    },
    {
      "acc": 0.65704389,
      "epoch": 0.6675545408422121,
      "grad_norm": 5.15625,
      "learning_rate": 7.95525846384027e-06,
      "loss": 1.59198112,
      "memory(GiB)": 107.26,
      "step": 26315,
      "train_speed(iter/s)": 1.634692
    },
    {
      "acc": 0.66022663,
      "epoch": 0.6676813800101471,
      "grad_norm": 4.5625,
      "learning_rate": 7.954412543674493e-06,
      "loss": 1.55944347,
      "memory(GiB)": 107.26,
      "step": 26320,
      "train_speed(iter/s)": 1.634727
    },
    {
      "acc": 0.65044961,
      "epoch": 0.6678082191780822,
      "grad_norm": 6.3125,
      "learning_rate": 7.95356649356046e-06,
      "loss": 1.6676157,
      "memory(GiB)": 107.26,
      "step": 26325,
      "train_speed(iter/s)": 1.634763
    },
    {
      "acc": 0.63911214,
      "epoch": 0.6679350583460173,
      "grad_norm": 6.53125,
      "learning_rate": 7.952720313535387e-06,
      "loss": 1.67508221,
      "memory(GiB)": 107.26,
      "step": 26330,
      "train_speed(iter/s)": 1.634799
    },
    {
      "acc": 0.6503706,
      "epoch": 0.6680618975139523,
      "grad_norm": 5.59375,
      "learning_rate": 7.951874003636492e-06,
      "loss": 1.67283287,
      "memory(GiB)": 107.26,
      "step": 26335,
      "train_speed(iter/s)": 1.634834
    },
    {
      "acc": 0.66250377,
      "epoch": 0.6681887366818874,
      "grad_norm": 4.65625,
      "learning_rate": 7.951027563901e-06,
      "loss": 1.51327457,
      "memory(GiB)": 107.26,
      "step": 26340,
      "train_speed(iter/s)": 1.634868
    },
    {
      "acc": 0.64156828,
      "epoch": 0.6683155758498224,
      "grad_norm": 7.125,
      "learning_rate": 7.950180994366138e-06,
      "loss": 1.65698223,
      "memory(GiB)": 107.26,
      "step": 26345,
      "train_speed(iter/s)": 1.634904
    },
    {
      "acc": 0.67646832,
      "epoch": 0.6684424150177575,
      "grad_norm": 5.75,
      "learning_rate": 7.949334295069147e-06,
      "loss": 1.50760193,
      "memory(GiB)": 107.26,
      "step": 26350,
      "train_speed(iter/s)": 1.634939
    },
    {
      "acc": 0.64608178,
      "epoch": 0.6685692541856926,
      "grad_norm": 6.8125,
      "learning_rate": 7.948487466047263e-06,
      "loss": 1.66705494,
      "memory(GiB)": 107.26,
      "step": 26355,
      "train_speed(iter/s)": 1.634974
    },
    {
      "acc": 0.6460598,
      "epoch": 0.6686960933536276,
      "grad_norm": 7.15625,
      "learning_rate": 7.947640507337737e-06,
      "loss": 1.59428129,
      "memory(GiB)": 107.26,
      "step": 26360,
      "train_speed(iter/s)": 1.635008
    },
    {
      "acc": 0.64024282,
      "epoch": 0.6688229325215627,
      "grad_norm": 5.5625,
      "learning_rate": 7.946793418977821e-06,
      "loss": 1.61949196,
      "memory(GiB)": 107.26,
      "step": 26365,
      "train_speed(iter/s)": 1.635041
    },
    {
      "acc": 0.63931704,
      "epoch": 0.6689497716894978,
      "grad_norm": 5.375,
      "learning_rate": 7.945946201004775e-06,
      "loss": 1.65089874,
      "memory(GiB)": 107.26,
      "step": 26370,
      "train_speed(iter/s)": 1.635073
    },
    {
      "acc": 0.64310131,
      "epoch": 0.6690766108574328,
      "grad_norm": 5.46875,
      "learning_rate": 7.945098853455862e-06,
      "loss": 1.64254761,
      "memory(GiB)": 107.26,
      "step": 26375,
      "train_speed(iter/s)": 1.635109
    },
    {
      "acc": 0.66963959,
      "epoch": 0.6692034500253679,
      "grad_norm": 4.40625,
      "learning_rate": 7.944251376368352e-06,
      "loss": 1.54499846,
      "memory(GiB)": 107.26,
      "step": 26380,
      "train_speed(iter/s)": 1.635142
    },
    {
      "acc": 0.64572573,
      "epoch": 0.6693302891933028,
      "grad_norm": 5.59375,
      "learning_rate": 7.943403769779523e-06,
      "loss": 1.67245369,
      "memory(GiB)": 107.26,
      "step": 26385,
      "train_speed(iter/s)": 1.635175
    },
    {
      "acc": 0.65155425,
      "epoch": 0.6694571283612379,
      "grad_norm": 6.28125,
      "learning_rate": 7.942556033726654e-06,
      "loss": 1.6684967,
      "memory(GiB)": 107.26,
      "step": 26390,
      "train_speed(iter/s)": 1.635211
    },
    {
      "acc": 0.66894259,
      "epoch": 0.669583967529173,
      "grad_norm": 5.25,
      "learning_rate": 7.941708168247033e-06,
      "loss": 1.61360855,
      "memory(GiB)": 107.26,
      "step": 26395,
      "train_speed(iter/s)": 1.635246
    },
    {
      "acc": 0.66247182,
      "epoch": 0.669710806697108,
      "grad_norm": 5.53125,
      "learning_rate": 7.940860173377952e-06,
      "loss": 1.6089695,
      "memory(GiB)": 107.26,
      "step": 26400,
      "train_speed(iter/s)": 1.63528
    },
    {
      "acc": 0.65205727,
      "epoch": 0.6698376458650431,
      "grad_norm": 5.8125,
      "learning_rate": 7.940012049156711e-06,
      "loss": 1.60652428,
      "memory(GiB)": 107.26,
      "step": 26405,
      "train_speed(iter/s)": 1.635314
    },
    {
      "acc": 0.66205568,
      "epoch": 0.6699644850329782,
      "grad_norm": 5.0625,
      "learning_rate": 7.939163795620614e-06,
      "loss": 1.60997677,
      "memory(GiB)": 107.26,
      "step": 26410,
      "train_speed(iter/s)": 1.635348
    },
    {
      "acc": 0.64853101,
      "epoch": 0.6700913242009132,
      "grad_norm": 5.71875,
      "learning_rate": 7.938315412806971e-06,
      "loss": 1.64800568,
      "memory(GiB)": 107.26,
      "step": 26415,
      "train_speed(iter/s)": 1.635381
    },
    {
      "acc": 0.67020588,
      "epoch": 0.6702181633688483,
      "grad_norm": 6.5,
      "learning_rate": 7.937466900753098e-06,
      "loss": 1.54874592,
      "memory(GiB)": 107.26,
      "step": 26420,
      "train_speed(iter/s)": 1.635416
    },
    {
      "acc": 0.64904571,
      "epoch": 0.6703450025367833,
      "grad_norm": 7.03125,
      "learning_rate": 7.936618259496316e-06,
      "loss": 1.64735374,
      "memory(GiB)": 107.26,
      "step": 26425,
      "train_speed(iter/s)": 1.635452
    },
    {
      "acc": 0.6575202,
      "epoch": 0.6704718417047184,
      "grad_norm": 5.5625,
      "learning_rate": 7.935769489073952e-06,
      "loss": 1.58822384,
      "memory(GiB)": 107.26,
      "step": 26430,
      "train_speed(iter/s)": 1.635488
    },
    {
      "acc": 0.66376667,
      "epoch": 0.6705986808726535,
      "grad_norm": 5.3125,
      "learning_rate": 7.934920589523336e-06,
      "loss": 1.52993689,
      "memory(GiB)": 107.26,
      "step": 26435,
      "train_speed(iter/s)": 1.635521
    },
    {
      "acc": 0.64739075,
      "epoch": 0.6707255200405885,
      "grad_norm": 7.71875,
      "learning_rate": 7.934071560881812e-06,
      "loss": 1.64702415,
      "memory(GiB)": 107.26,
      "step": 26440,
      "train_speed(iter/s)": 1.635554
    },
    {
      "acc": 0.66285353,
      "epoch": 0.6708523592085236,
      "grad_norm": 5.28125,
      "learning_rate": 7.93322240318672e-06,
      "loss": 1.57003212,
      "memory(GiB)": 107.26,
      "step": 26445,
      "train_speed(iter/s)": 1.635584
    },
    {
      "acc": 0.65634179,
      "epoch": 0.6709791983764587,
      "grad_norm": 5.0,
      "learning_rate": 7.93237311647541e-06,
      "loss": 1.65199566,
      "memory(GiB)": 107.26,
      "step": 26450,
      "train_speed(iter/s)": 1.635617
    },
    {
      "acc": 0.66775026,
      "epoch": 0.6711060375443937,
      "grad_norm": 5.5625,
      "learning_rate": 7.93152370078524e-06,
      "loss": 1.5171627,
      "memory(GiB)": 107.26,
      "step": 26455,
      "train_speed(iter/s)": 1.635652
    },
    {
      "acc": 0.65419035,
      "epoch": 0.6712328767123288,
      "grad_norm": 4.75,
      "learning_rate": 7.930674156153569e-06,
      "loss": 1.57321243,
      "memory(GiB)": 107.26,
      "step": 26460,
      "train_speed(iter/s)": 1.635684
    },
    {
      "acc": 0.65150967,
      "epoch": 0.6713597158802638,
      "grad_norm": 5.1875,
      "learning_rate": 7.929824482617763e-06,
      "loss": 1.62609501,
      "memory(GiB)": 107.26,
      "step": 26465,
      "train_speed(iter/s)": 1.635716
    },
    {
      "acc": 0.64687262,
      "epoch": 0.6714865550481989,
      "grad_norm": 6.5625,
      "learning_rate": 7.928974680215196e-06,
      "loss": 1.69332085,
      "memory(GiB)": 107.26,
      "step": 26470,
      "train_speed(iter/s)": 1.635749
    },
    {
      "acc": 0.64114003,
      "epoch": 0.671613394216134,
      "grad_norm": 5.21875,
      "learning_rate": 7.928124748983244e-06,
      "loss": 1.65573158,
      "memory(GiB)": 107.26,
      "step": 26475,
      "train_speed(iter/s)": 1.635782
    },
    {
      "acc": 0.64556017,
      "epoch": 0.671740233384069,
      "grad_norm": 6.59375,
      "learning_rate": 7.927274688959294e-06,
      "loss": 1.67613335,
      "memory(GiB)": 107.26,
      "step": 26480,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.64693079,
      "epoch": 0.6718670725520041,
      "grad_norm": 6.5625,
      "learning_rate": 7.926424500180734e-06,
      "loss": 1.66498871,
      "memory(GiB)": 107.26,
      "step": 26485,
      "train_speed(iter/s)": 1.635848
    },
    {
      "acc": 0.68129091,
      "epoch": 0.6719939117199392,
      "grad_norm": 5.8125,
      "learning_rate": 7.92557418268496e-06,
      "loss": 1.5222827,
      "memory(GiB)": 107.26,
      "step": 26490,
      "train_speed(iter/s)": 1.63588
    },
    {
      "acc": 0.65798445,
      "epoch": 0.6721207508878742,
      "grad_norm": 5.4375,
      "learning_rate": 7.92472373650937e-06,
      "loss": 1.58806095,
      "memory(GiB)": 107.26,
      "step": 26495,
      "train_speed(iter/s)": 1.635915
    },
    {
      "acc": 0.63552518,
      "epoch": 0.6722475900558093,
      "grad_norm": 5.1875,
      "learning_rate": 7.923873161691373e-06,
      "loss": 1.70570908,
      "memory(GiB)": 107.26,
      "step": 26500,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.66993885,
      "epoch": 0.6723744292237442,
      "grad_norm": 9.0,
      "learning_rate": 7.923022458268379e-06,
      "loss": 1.60077896,
      "memory(GiB)": 107.26,
      "step": 26505,
      "train_speed(iter/s)": 1.635979
    },
    {
      "acc": 0.66220317,
      "epoch": 0.6725012683916793,
      "grad_norm": 5.53125,
      "learning_rate": 7.922171626277809e-06,
      "loss": 1.60844975,
      "memory(GiB)": 107.26,
      "step": 26510,
      "train_speed(iter/s)": 1.636015
    },
    {
      "acc": 0.65502162,
      "epoch": 0.6726281075596144,
      "grad_norm": 4.90625,
      "learning_rate": 7.921320665757081e-06,
      "loss": 1.61531067,
      "memory(GiB)": 107.26,
      "step": 26515,
      "train_speed(iter/s)": 1.636048
    },
    {
      "acc": 0.65748725,
      "epoch": 0.6727549467275494,
      "grad_norm": 5.71875,
      "learning_rate": 7.920469576743631e-06,
      "loss": 1.575597,
      "memory(GiB)": 107.26,
      "step": 26520,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.65715456,
      "epoch": 0.6728817858954845,
      "grad_norm": 6.03125,
      "learning_rate": 7.919618359274888e-06,
      "loss": 1.62790604,
      "memory(GiB)": 107.26,
      "step": 26525,
      "train_speed(iter/s)": 1.636118
    },
    {
      "acc": 0.65933003,
      "epoch": 0.6730086250634196,
      "grad_norm": 6.125,
      "learning_rate": 7.918767013388295e-06,
      "loss": 1.62225952,
      "memory(GiB)": 107.26,
      "step": 26530,
      "train_speed(iter/s)": 1.636149
    },
    {
      "acc": 0.64483819,
      "epoch": 0.6731354642313546,
      "grad_norm": 5.875,
      "learning_rate": 7.917915539121297e-06,
      "loss": 1.64615746,
      "memory(GiB)": 107.26,
      "step": 26535,
      "train_speed(iter/s)": 1.636185
    },
    {
      "acc": 0.6288146,
      "epoch": 0.6732623033992897,
      "grad_norm": 5.25,
      "learning_rate": 7.917063936511347e-06,
      "loss": 1.7355484,
      "memory(GiB)": 107.26,
      "step": 26540,
      "train_speed(iter/s)": 1.636215
    },
    {
      "acc": 0.64846973,
      "epoch": 0.6733891425672247,
      "grad_norm": 6.5625,
      "learning_rate": 7.9162122055959e-06,
      "loss": 1.65554886,
      "memory(GiB)": 107.26,
      "step": 26545,
      "train_speed(iter/s)": 1.636249
    },
    {
      "acc": 0.66079869,
      "epoch": 0.6735159817351598,
      "grad_norm": 6.03125,
      "learning_rate": 7.91536034641242e-06,
      "loss": 1.60821095,
      "memory(GiB)": 107.26,
      "step": 26550,
      "train_speed(iter/s)": 1.636283
    },
    {
      "acc": 0.6459981,
      "epoch": 0.6736428209030949,
      "grad_norm": 9.5,
      "learning_rate": 7.914508358998376e-06,
      "loss": 1.67073231,
      "memory(GiB)": 107.26,
      "step": 26555,
      "train_speed(iter/s)": 1.636318
    },
    {
      "acc": 0.67174993,
      "epoch": 0.6737696600710299,
      "grad_norm": 5.65625,
      "learning_rate": 7.913656243391243e-06,
      "loss": 1.54314861,
      "memory(GiB)": 107.26,
      "step": 26560,
      "train_speed(iter/s)": 1.636354
    },
    {
      "acc": 0.64914966,
      "epoch": 0.673896499238965,
      "grad_norm": 6.125,
      "learning_rate": 7.9128039996285e-06,
      "loss": 1.5862114,
      "memory(GiB)": 107.26,
      "step": 26565,
      "train_speed(iter/s)": 1.636388
    },
    {
      "acc": 0.64251785,
      "epoch": 0.6740233384069001,
      "grad_norm": 6.21875,
      "learning_rate": 7.911951627747633e-06,
      "loss": 1.64313965,
      "memory(GiB)": 107.26,
      "step": 26570,
      "train_speed(iter/s)": 1.63642
    },
    {
      "acc": 0.64136143,
      "epoch": 0.6741501775748351,
      "grad_norm": 5.75,
      "learning_rate": 7.91109912778613e-06,
      "loss": 1.65305634,
      "memory(GiB)": 107.26,
      "step": 26575,
      "train_speed(iter/s)": 1.636455
    },
    {
      "acc": 0.65396233,
      "epoch": 0.6742770167427702,
      "grad_norm": 5.0625,
      "learning_rate": 7.910246499781492e-06,
      "loss": 1.60790348,
      "memory(GiB)": 107.26,
      "step": 26580,
      "train_speed(iter/s)": 1.636487
    },
    {
      "acc": 0.64579554,
      "epoch": 0.6744038559107052,
      "grad_norm": 5.53125,
      "learning_rate": 7.90939374377122e-06,
      "loss": 1.68032074,
      "memory(GiB)": 107.26,
      "step": 26585,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.65164728,
      "epoch": 0.6745306950786403,
      "grad_norm": 4.90625,
      "learning_rate": 7.908540859792821e-06,
      "loss": 1.57334747,
      "memory(GiB)": 107.26,
      "step": 26590,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.65671206,
      "epoch": 0.6746575342465754,
      "grad_norm": 6.0,
      "learning_rate": 7.907687847883809e-06,
      "loss": 1.56122694,
      "memory(GiB)": 107.26,
      "step": 26595,
      "train_speed(iter/s)": 1.636588
    },
    {
      "acc": 0.65064335,
      "epoch": 0.6747843734145104,
      "grad_norm": 6.25,
      "learning_rate": 7.906834708081703e-06,
      "loss": 1.60919304,
      "memory(GiB)": 107.26,
      "step": 26600,
      "train_speed(iter/s)": 1.636624
    },
    {
      "acc": 0.65889888,
      "epoch": 0.6749112125824455,
      "grad_norm": 5.59375,
      "learning_rate": 7.90598144042403e-06,
      "loss": 1.58028049,
      "memory(GiB)": 107.26,
      "step": 26605,
      "train_speed(iter/s)": 1.636657
    },
    {
      "acc": 0.6562705,
      "epoch": 0.6750380517503806,
      "grad_norm": 6.625,
      "learning_rate": 7.905128044948318e-06,
      "loss": 1.6336153,
      "memory(GiB)": 107.26,
      "step": 26610,
      "train_speed(iter/s)": 1.636689
    },
    {
      "acc": 0.64722996,
      "epoch": 0.6751648909183156,
      "grad_norm": 5.46875,
      "learning_rate": 7.904274521692104e-06,
      "loss": 1.65594254,
      "memory(GiB)": 107.26,
      "step": 26615,
      "train_speed(iter/s)": 1.636722
    },
    {
      "acc": 0.65281363,
      "epoch": 0.6752917300862507,
      "grad_norm": 5.21875,
      "learning_rate": 7.90342087069293e-06,
      "loss": 1.60127831,
      "memory(GiB)": 107.26,
      "step": 26620,
      "train_speed(iter/s)": 1.636757
    },
    {
      "acc": 0.65799212,
      "epoch": 0.6754185692541856,
      "grad_norm": 5.46875,
      "learning_rate": 7.902567091988343e-06,
      "loss": 1.56904078,
      "memory(GiB)": 107.26,
      "step": 26625,
      "train_speed(iter/s)": 1.636791
    },
    {
      "acc": 0.65744662,
      "epoch": 0.6755454084221207,
      "grad_norm": 6.6875,
      "learning_rate": 7.901713185615898e-06,
      "loss": 1.60119095,
      "memory(GiB)": 107.26,
      "step": 26630,
      "train_speed(iter/s)": 1.636824
    },
    {
      "acc": 0.64757876,
      "epoch": 0.6756722475900558,
      "grad_norm": 6.125,
      "learning_rate": 7.90085915161315e-06,
      "loss": 1.61928406,
      "memory(GiB)": 107.26,
      "step": 26635,
      "train_speed(iter/s)": 1.636856
    },
    {
      "acc": 0.65160193,
      "epoch": 0.6757990867579908,
      "grad_norm": 5.125,
      "learning_rate": 7.900004990017667e-06,
      "loss": 1.63451233,
      "memory(GiB)": 107.26,
      "step": 26640,
      "train_speed(iter/s)": 1.636891
    },
    {
      "acc": 0.63665543,
      "epoch": 0.6759259259259259,
      "grad_norm": 6.3125,
      "learning_rate": 7.899150700867014e-06,
      "loss": 1.68126888,
      "memory(GiB)": 107.26,
      "step": 26645,
      "train_speed(iter/s)": 1.636925
    },
    {
      "acc": 0.65109377,
      "epoch": 0.676052765093861,
      "grad_norm": 5.9375,
      "learning_rate": 7.898296284198772e-06,
      "loss": 1.58115616,
      "memory(GiB)": 107.26,
      "step": 26650,
      "train_speed(iter/s)": 1.636957
    },
    {
      "acc": 0.66057358,
      "epoch": 0.676179604261796,
      "grad_norm": 6.625,
      "learning_rate": 7.897441740050518e-06,
      "loss": 1.64735184,
      "memory(GiB)": 107.26,
      "step": 26655,
      "train_speed(iter/s)": 1.63699
    },
    {
      "acc": 0.63221507,
      "epoch": 0.6763064434297311,
      "grad_norm": 5.34375,
      "learning_rate": 7.89658706845984e-06,
      "loss": 1.63628731,
      "memory(GiB)": 107.26,
      "step": 26660,
      "train_speed(iter/s)": 1.637023
    },
    {
      "acc": 0.64983578,
      "epoch": 0.6764332825976661,
      "grad_norm": 6.09375,
      "learning_rate": 7.89573226946433e-06,
      "loss": 1.62094765,
      "memory(GiB)": 107.26,
      "step": 26665,
      "train_speed(iter/s)": 1.637056
    },
    {
      "acc": 0.65202594,
      "epoch": 0.6765601217656012,
      "grad_norm": 6.96875,
      "learning_rate": 7.89487734310159e-06,
      "loss": 1.62998619,
      "memory(GiB)": 107.26,
      "step": 26670,
      "train_speed(iter/s)": 1.637088
    },
    {
      "acc": 0.65368414,
      "epoch": 0.6766869609335363,
      "grad_norm": 5.09375,
      "learning_rate": 7.894022289409216e-06,
      "loss": 1.62135048,
      "memory(GiB)": 107.26,
      "step": 26675,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.63958931,
      "epoch": 0.6768138001014713,
      "grad_norm": 5.09375,
      "learning_rate": 7.893167108424822e-06,
      "loss": 1.66207352,
      "memory(GiB)": 107.26,
      "step": 26680,
      "train_speed(iter/s)": 1.637155
    },
    {
      "acc": 0.66196418,
      "epoch": 0.6769406392694064,
      "grad_norm": 5.4375,
      "learning_rate": 7.89231180018602e-06,
      "loss": 1.53666916,
      "memory(GiB)": 107.26,
      "step": 26685,
      "train_speed(iter/s)": 1.637191
    },
    {
      "acc": 0.65888853,
      "epoch": 0.6770674784373415,
      "grad_norm": 5.59375,
      "learning_rate": 7.891456364730434e-06,
      "loss": 1.6374464,
      "memory(GiB)": 107.26,
      "step": 26690,
      "train_speed(iter/s)": 1.637224
    },
    {
      "acc": 0.6450736,
      "epoch": 0.6771943176052765,
      "grad_norm": 5.21875,
      "learning_rate": 7.890600802095686e-06,
      "loss": 1.70170803,
      "memory(GiB)": 107.26,
      "step": 26695,
      "train_speed(iter/s)": 1.637258
    },
    {
      "acc": 0.65078168,
      "epoch": 0.6773211567732116,
      "grad_norm": 6.875,
      "learning_rate": 7.889745112319411e-06,
      "loss": 1.58687868,
      "memory(GiB)": 107.26,
      "step": 26700,
      "train_speed(iter/s)": 1.637291
    },
    {
      "acc": 0.65395117,
      "epoch": 0.6774479959411466,
      "grad_norm": 7.3125,
      "learning_rate": 7.888889295439244e-06,
      "loss": 1.59613609,
      "memory(GiB)": 107.26,
      "step": 26705,
      "train_speed(iter/s)": 1.637325
    },
    {
      "acc": 0.65182972,
      "epoch": 0.6775748351090817,
      "grad_norm": 6.4375,
      "learning_rate": 7.888033351492827e-06,
      "loss": 1.63670216,
      "memory(GiB)": 107.26,
      "step": 26710,
      "train_speed(iter/s)": 1.637357
    },
    {
      "acc": 0.64360676,
      "epoch": 0.6777016742770168,
      "grad_norm": 5.34375,
      "learning_rate": 7.887177280517808e-06,
      "loss": 1.696245,
      "memory(GiB)": 107.26,
      "step": 26715,
      "train_speed(iter/s)": 1.63739
    },
    {
      "acc": 0.65883231,
      "epoch": 0.6778285134449518,
      "grad_norm": 5.8125,
      "learning_rate": 7.886321082551845e-06,
      "loss": 1.5388957,
      "memory(GiB)": 107.26,
      "step": 26720,
      "train_speed(iter/s)": 1.637424
    },
    {
      "acc": 0.66942749,
      "epoch": 0.6779553526128869,
      "grad_norm": 6.03125,
      "learning_rate": 7.88546475763259e-06,
      "loss": 1.55770645,
      "memory(GiB)": 107.26,
      "step": 26725,
      "train_speed(iter/s)": 1.637455
    },
    {
      "acc": 0.66165552,
      "epoch": 0.678082191780822,
      "grad_norm": 5.21875,
      "learning_rate": 7.884608305797716e-06,
      "loss": 1.5738019,
      "memory(GiB)": 107.26,
      "step": 26730,
      "train_speed(iter/s)": 1.637487
    },
    {
      "acc": 0.64709148,
      "epoch": 0.678209030948757,
      "grad_norm": 4.90625,
      "learning_rate": 7.883751727084888e-06,
      "loss": 1.66402664,
      "memory(GiB)": 107.26,
      "step": 26735,
      "train_speed(iter/s)": 1.63752
    },
    {
      "acc": 0.65054336,
      "epoch": 0.678335870116692,
      "grad_norm": 5.125,
      "learning_rate": 7.882895021531784e-06,
      "loss": 1.65783081,
      "memory(GiB)": 107.26,
      "step": 26740,
      "train_speed(iter/s)": 1.637553
    },
    {
      "acc": 0.65563211,
      "epoch": 0.678462709284627,
      "grad_norm": 6.28125,
      "learning_rate": 7.882038189176085e-06,
      "loss": 1.59870319,
      "memory(GiB)": 107.26,
      "step": 26745,
      "train_speed(iter/s)": 1.637587
    },
    {
      "acc": 0.66284204,
      "epoch": 0.6785895484525621,
      "grad_norm": 5.375,
      "learning_rate": 7.881181230055481e-06,
      "loss": 1.59951248,
      "memory(GiB)": 107.26,
      "step": 26750,
      "train_speed(iter/s)": 1.637619
    },
    {
      "acc": 0.66546164,
      "epoch": 0.6787163876204972,
      "grad_norm": 12.25,
      "learning_rate": 7.880324144207663e-06,
      "loss": 1.58989954,
      "memory(GiB)": 107.26,
      "step": 26755,
      "train_speed(iter/s)": 1.637651
    },
    {
      "acc": 0.66123896,
      "epoch": 0.6788432267884322,
      "grad_norm": 6.1875,
      "learning_rate": 7.879466931670328e-06,
      "loss": 1.61991444,
      "memory(GiB)": 107.26,
      "step": 26760,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.64596043,
      "epoch": 0.6789700659563673,
      "grad_norm": 6.59375,
      "learning_rate": 7.878609592481182e-06,
      "loss": 1.70888786,
      "memory(GiB)": 107.26,
      "step": 26765,
      "train_speed(iter/s)": 1.63772
    },
    {
      "acc": 0.66199121,
      "epoch": 0.6790969051243024,
      "grad_norm": 5.46875,
      "learning_rate": 7.877752126677933e-06,
      "loss": 1.61715145,
      "memory(GiB)": 107.26,
      "step": 26770,
      "train_speed(iter/s)": 1.637753
    },
    {
      "acc": 0.67377157,
      "epoch": 0.6792237442922374,
      "grad_norm": 5.40625,
      "learning_rate": 7.876894534298298e-06,
      "loss": 1.5004343,
      "memory(GiB)": 107.26,
      "step": 26775,
      "train_speed(iter/s)": 1.637784
    },
    {
      "acc": 0.66394854,
      "epoch": 0.6793505834601725,
      "grad_norm": 6.03125,
      "learning_rate": 7.87603681538e-06,
      "loss": 1.54943047,
      "memory(GiB)": 107.26,
      "step": 26780,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.63446035,
      "epoch": 0.6794774226281075,
      "grad_norm": 6.125,
      "learning_rate": 7.875178969960757e-06,
      "loss": 1.72093811,
      "memory(GiB)": 107.26,
      "step": 26785,
      "train_speed(iter/s)": 1.637849
    },
    {
      "acc": 0.64778862,
      "epoch": 0.6796042617960426,
      "grad_norm": 4.8125,
      "learning_rate": 7.87432099807831e-06,
      "loss": 1.6499033,
      "memory(GiB)": 107.26,
      "step": 26790,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.65446739,
      "epoch": 0.6797311009639777,
      "grad_norm": 5.9375,
      "learning_rate": 7.87346289977039e-06,
      "loss": 1.62626915,
      "memory(GiB)": 107.26,
      "step": 26795,
      "train_speed(iter/s)": 1.637913
    },
    {
      "acc": 0.66241598,
      "epoch": 0.6798579401319127,
      "grad_norm": 5.84375,
      "learning_rate": 7.872604675074745e-06,
      "loss": 1.59820137,
      "memory(GiB)": 107.26,
      "step": 26800,
      "train_speed(iter/s)": 1.637946
    },
    {
      "acc": 0.65282211,
      "epoch": 0.6799847792998478,
      "grad_norm": 7.34375,
      "learning_rate": 7.871746324029119e-06,
      "loss": 1.5940114,
      "memory(GiB)": 107.26,
      "step": 26805,
      "train_speed(iter/s)": 1.637981
    },
    {
      "acc": 0.65721221,
      "epoch": 0.6801116184677829,
      "grad_norm": 5.8125,
      "learning_rate": 7.87088784667127e-06,
      "loss": 1.58220139,
      "memory(GiB)": 107.26,
      "step": 26810,
      "train_speed(iter/s)": 1.638015
    },
    {
      "acc": 0.63236394,
      "epoch": 0.6802384576357179,
      "grad_norm": 5.5625,
      "learning_rate": 7.870029243038955e-06,
      "loss": 1.70133324,
      "memory(GiB)": 107.26,
      "step": 26815,
      "train_speed(iter/s)": 1.638049
    },
    {
      "acc": 0.6395833,
      "epoch": 0.680365296803653,
      "grad_norm": 5.75,
      "learning_rate": 7.869170513169941e-06,
      "loss": 1.65109482,
      "memory(GiB)": 107.26,
      "step": 26820,
      "train_speed(iter/s)": 1.638081
    },
    {
      "acc": 0.63203564,
      "epoch": 0.680492135971588,
      "grad_norm": 5.25,
      "learning_rate": 7.868311657101996e-06,
      "loss": 1.62600861,
      "memory(GiB)": 107.26,
      "step": 26825,
      "train_speed(iter/s)": 1.638115
    },
    {
      "acc": 0.64707937,
      "epoch": 0.6806189751395231,
      "grad_norm": 7.40625,
      "learning_rate": 7.8674526748729e-06,
      "loss": 1.65282154,
      "memory(GiB)": 107.26,
      "step": 26830,
      "train_speed(iter/s)": 1.638148
    },
    {
      "acc": 0.67791576,
      "epoch": 0.6807458143074582,
      "grad_norm": 7.5,
      "learning_rate": 7.866593566520432e-06,
      "loss": 1.48686743,
      "memory(GiB)": 107.26,
      "step": 26835,
      "train_speed(iter/s)": 1.638185
    },
    {
      "acc": 0.65274925,
      "epoch": 0.6808726534753932,
      "grad_norm": 5.21875,
      "learning_rate": 7.865734332082382e-06,
      "loss": 1.68021507,
      "memory(GiB)": 107.26,
      "step": 26840,
      "train_speed(iter/s)": 1.638217
    },
    {
      "acc": 0.66506386,
      "epoch": 0.6809994926433283,
      "grad_norm": 7.03125,
      "learning_rate": 7.86487497159654e-06,
      "loss": 1.54281139,
      "memory(GiB)": 107.26,
      "step": 26845,
      "train_speed(iter/s)": 1.63825
    },
    {
      "acc": 0.65327525,
      "epoch": 0.6811263318112634,
      "grad_norm": 5.25,
      "learning_rate": 7.864015485100706e-06,
      "loss": 1.59985142,
      "memory(GiB)": 107.26,
      "step": 26850,
      "train_speed(iter/s)": 1.638284
    },
    {
      "acc": 0.65137606,
      "epoch": 0.6812531709791984,
      "grad_norm": 5.53125,
      "learning_rate": 7.863155872632685e-06,
      "loss": 1.54490366,
      "memory(GiB)": 107.26,
      "step": 26855,
      "train_speed(iter/s)": 1.638317
    },
    {
      "acc": 0.65559554,
      "epoch": 0.6813800101471335,
      "grad_norm": 6.0625,
      "learning_rate": 7.862296134230287e-06,
      "loss": 1.57524509,
      "memory(GiB)": 107.26,
      "step": 26860,
      "train_speed(iter/s)": 1.638351
    },
    {
      "acc": 0.65322442,
      "epoch": 0.6815068493150684,
      "grad_norm": 5.96875,
      "learning_rate": 7.861436269931322e-06,
      "loss": 1.59274559,
      "memory(GiB)": 107.26,
      "step": 26865,
      "train_speed(iter/s)": 1.638383
    },
    {
      "acc": 0.6607255,
      "epoch": 0.6816336884830035,
      "grad_norm": 5.21875,
      "learning_rate": 7.860576279773617e-06,
      "loss": 1.53485146,
      "memory(GiB)": 107.26,
      "step": 26870,
      "train_speed(iter/s)": 1.638416
    },
    {
      "acc": 0.65396767,
      "epoch": 0.6817605276509386,
      "grad_norm": 6.0,
      "learning_rate": 7.859716163794995e-06,
      "loss": 1.58389664,
      "memory(GiB)": 107.26,
      "step": 26875,
      "train_speed(iter/s)": 1.638449
    },
    {
      "acc": 0.65315032,
      "epoch": 0.6818873668188736,
      "grad_norm": 4.21875,
      "learning_rate": 7.858855922033289e-06,
      "loss": 1.62126102,
      "memory(GiB)": 107.26,
      "step": 26880,
      "train_speed(iter/s)": 1.638482
    },
    {
      "acc": 0.64716854,
      "epoch": 0.6820142059868087,
      "grad_norm": 6.28125,
      "learning_rate": 7.857995554526334e-06,
      "loss": 1.67432861,
      "memory(GiB)": 107.26,
      "step": 26885,
      "train_speed(iter/s)": 1.638515
    },
    {
      "acc": 0.64961729,
      "epoch": 0.6821410451547438,
      "grad_norm": 5.28125,
      "learning_rate": 7.857135061311977e-06,
      "loss": 1.62102013,
      "memory(GiB)": 107.26,
      "step": 26890,
      "train_speed(iter/s)": 1.638547
    },
    {
      "acc": 0.64940009,
      "epoch": 0.6822678843226788,
      "grad_norm": 5.4375,
      "learning_rate": 7.856274442428062e-06,
      "loss": 1.65618038,
      "memory(GiB)": 107.26,
      "step": 26895,
      "train_speed(iter/s)": 1.63858
    },
    {
      "acc": 0.6480588,
      "epoch": 0.6823947234906139,
      "grad_norm": 5.8125,
      "learning_rate": 7.855413697912446e-06,
      "loss": 1.61122971,
      "memory(GiB)": 107.26,
      "step": 26900,
      "train_speed(iter/s)": 1.638615
    },
    {
      "acc": 0.65209689,
      "epoch": 0.6825215626585489,
      "grad_norm": 5.28125,
      "learning_rate": 7.854552827802987e-06,
      "loss": 1.6268013,
      "memory(GiB)": 107.26,
      "step": 26905,
      "train_speed(iter/s)": 1.638646
    },
    {
      "acc": 0.66580715,
      "epoch": 0.682648401826484,
      "grad_norm": 12.375,
      "learning_rate": 7.853691832137547e-06,
      "loss": 1.656567,
      "memory(GiB)": 107.26,
      "step": 26910,
      "train_speed(iter/s)": 1.638678
    },
    {
      "acc": 0.66611404,
      "epoch": 0.6827752409944191,
      "grad_norm": 4.9375,
      "learning_rate": 7.852830710954003e-06,
      "loss": 1.56231728,
      "memory(GiB)": 107.26,
      "step": 26915,
      "train_speed(iter/s)": 1.638711
    },
    {
      "acc": 0.65471764,
      "epoch": 0.6829020801623541,
      "grad_norm": 6.15625,
      "learning_rate": 7.851969464290226e-06,
      "loss": 1.62660065,
      "memory(GiB)": 107.26,
      "step": 26920,
      "train_speed(iter/s)": 1.638745
    },
    {
      "acc": 0.65919824,
      "epoch": 0.6830289193302892,
      "grad_norm": 5.96875,
      "learning_rate": 7.851108092184099e-06,
      "loss": 1.58362312,
      "memory(GiB)": 107.26,
      "step": 26925,
      "train_speed(iter/s)": 1.638778
    },
    {
      "acc": 0.66248193,
      "epoch": 0.6831557584982243,
      "grad_norm": 5.21875,
      "learning_rate": 7.850246594673508e-06,
      "loss": 1.60354748,
      "memory(GiB)": 107.26,
      "step": 26930,
      "train_speed(iter/s)": 1.638813
    },
    {
      "acc": 0.65030179,
      "epoch": 0.6832825976661593,
      "grad_norm": 6.34375,
      "learning_rate": 7.849384971796346e-06,
      "loss": 1.69164639,
      "memory(GiB)": 107.26,
      "step": 26935,
      "train_speed(iter/s)": 1.638846
    },
    {
      "acc": 0.65637693,
      "epoch": 0.6834094368340944,
      "grad_norm": 5.625,
      "learning_rate": 7.848523223590514e-06,
      "loss": 1.6101099,
      "memory(GiB)": 107.26,
      "step": 26940,
      "train_speed(iter/s)": 1.638879
    },
    {
      "acc": 0.65389519,
      "epoch": 0.6835362760020294,
      "grad_norm": 6.40625,
      "learning_rate": 7.84766135009391e-06,
      "loss": 1.67608795,
      "memory(GiB)": 107.26,
      "step": 26945,
      "train_speed(iter/s)": 1.638912
    },
    {
      "acc": 0.64838867,
      "epoch": 0.6836631151699645,
      "grad_norm": 4.75,
      "learning_rate": 7.846799351344447e-06,
      "loss": 1.63493614,
      "memory(GiB)": 107.26,
      "step": 26950,
      "train_speed(iter/s)": 1.638945
    },
    {
      "acc": 0.66400805,
      "epoch": 0.6837899543378996,
      "grad_norm": 6.71875,
      "learning_rate": 7.845937227380038e-06,
      "loss": 1.59935093,
      "memory(GiB)": 107.26,
      "step": 26955,
      "train_speed(iter/s)": 1.638978
    },
    {
      "acc": 0.6427022,
      "epoch": 0.6839167935058346,
      "grad_norm": 5.21875,
      "learning_rate": 7.845074978238604e-06,
      "loss": 1.67486324,
      "memory(GiB)": 107.26,
      "step": 26960,
      "train_speed(iter/s)": 1.639009
    },
    {
      "acc": 0.65842757,
      "epoch": 0.6840436326737697,
      "grad_norm": 5.40625,
      "learning_rate": 7.84421260395807e-06,
      "loss": 1.61804733,
      "memory(GiB)": 107.26,
      "step": 26965,
      "train_speed(iter/s)": 1.639043
    },
    {
      "acc": 0.65147972,
      "epoch": 0.6841704718417048,
      "grad_norm": 5.75,
      "learning_rate": 7.84335010457637e-06,
      "loss": 1.62659683,
      "memory(GiB)": 107.26,
      "step": 26970,
      "train_speed(iter/s)": 1.639076
    },
    {
      "acc": 0.65960317,
      "epoch": 0.6842973110096398,
      "grad_norm": 6.34375,
      "learning_rate": 7.842487480131435e-06,
      "loss": 1.52411785,
      "memory(GiB)": 107.26,
      "step": 26975,
      "train_speed(iter/s)": 1.639109
    },
    {
      "acc": 0.65147829,
      "epoch": 0.6844241501775749,
      "grad_norm": 6.25,
      "learning_rate": 7.84162473066121e-06,
      "loss": 1.6766367,
      "memory(GiB)": 107.26,
      "step": 26980,
      "train_speed(iter/s)": 1.639141
    },
    {
      "acc": 0.66292338,
      "epoch": 0.6845509893455098,
      "grad_norm": 5.375,
      "learning_rate": 7.840761856203642e-06,
      "loss": 1.50713816,
      "memory(GiB)": 107.26,
      "step": 26985,
      "train_speed(iter/s)": 1.639172
    },
    {
      "acc": 0.64710188,
      "epoch": 0.6846778285134449,
      "grad_norm": 5.46875,
      "learning_rate": 7.839898856796685e-06,
      "loss": 1.67985497,
      "memory(GiB)": 107.26,
      "step": 26990,
      "train_speed(iter/s)": 1.639203
    },
    {
      "acc": 0.64517975,
      "epoch": 0.68480466768138,
      "grad_norm": 5.78125,
      "learning_rate": 7.839035732478297e-06,
      "loss": 1.64958305,
      "memory(GiB)": 107.26,
      "step": 26995,
      "train_speed(iter/s)": 1.639235
    },
    {
      "acc": 0.65153446,
      "epoch": 0.684931506849315,
      "grad_norm": 6.0,
      "learning_rate": 7.838172483286441e-06,
      "loss": 1.69107857,
      "memory(GiB)": 107.26,
      "step": 27000,
      "train_speed(iter/s)": 1.639269
    },
    {
      "epoch": 0.684931506849315,
      "eval_acc": 0.6450710754985308,
      "eval_loss": 1.5803025960922241,
      "eval_runtime": 58.4784,
      "eval_samples_per_second": 108.929,
      "eval_steps_per_second": 27.241,
      "step": 27000
    },
    {
      "acc": 0.66906567,
      "epoch": 0.6850583460172501,
      "grad_norm": 5.0,
      "learning_rate": 7.83730910925909e-06,
      "loss": 1.55417967,
      "memory(GiB)": 107.26,
      "step": 27005,
      "train_speed(iter/s)": 1.633082
    },
    {
      "acc": 0.6517766,
      "epoch": 0.6851851851851852,
      "grad_norm": 5.21875,
      "learning_rate": 7.836445610434215e-06,
      "loss": 1.59759665,
      "memory(GiB)": 107.26,
      "step": 27010,
      "train_speed(iter/s)": 1.633116
    },
    {
      "acc": 0.65137568,
      "epoch": 0.6853120243531202,
      "grad_norm": 5.8125,
      "learning_rate": 7.835581986849799e-06,
      "loss": 1.63237514,
      "memory(GiB)": 107.26,
      "step": 27015,
      "train_speed(iter/s)": 1.633151
    },
    {
      "acc": 0.64684496,
      "epoch": 0.6854388635210553,
      "grad_norm": 5.46875,
      "learning_rate": 7.834718238543827e-06,
      "loss": 1.60083122,
      "memory(GiB)": 107.26,
      "step": 27020,
      "train_speed(iter/s)": 1.633183
    },
    {
      "acc": 0.6389863,
      "epoch": 0.6855657026889903,
      "grad_norm": 4.84375,
      "learning_rate": 7.833854365554289e-06,
      "loss": 1.66159573,
      "memory(GiB)": 107.26,
      "step": 27025,
      "train_speed(iter/s)": 1.633217
    },
    {
      "acc": 0.64249711,
      "epoch": 0.6856925418569254,
      "grad_norm": 4.6875,
      "learning_rate": 7.832990367919186e-06,
      "loss": 1.66134071,
      "memory(GiB)": 107.26,
      "step": 27030,
      "train_speed(iter/s)": 1.63325
    },
    {
      "acc": 0.6603487,
      "epoch": 0.6858193810248605,
      "grad_norm": 6.4375,
      "learning_rate": 7.832126245676518e-06,
      "loss": 1.5967391,
      "memory(GiB)": 107.26,
      "step": 27035,
      "train_speed(iter/s)": 1.633281
    },
    {
      "acc": 0.65505686,
      "epoch": 0.6859462201927955,
      "grad_norm": 6.71875,
      "learning_rate": 7.831261998864293e-06,
      "loss": 1.59895248,
      "memory(GiB)": 107.26,
      "step": 27040,
      "train_speed(iter/s)": 1.633315
    },
    {
      "acc": 0.65722203,
      "epoch": 0.6860730593607306,
      "grad_norm": 5.34375,
      "learning_rate": 7.830397627520526e-06,
      "loss": 1.65112114,
      "memory(GiB)": 107.26,
      "step": 27045,
      "train_speed(iter/s)": 1.633348
    },
    {
      "acc": 0.64407406,
      "epoch": 0.6861998985286657,
      "grad_norm": 5.65625,
      "learning_rate": 7.82953313168323e-06,
      "loss": 1.63695679,
      "memory(GiB)": 107.26,
      "step": 27050,
      "train_speed(iter/s)": 1.633382
    },
    {
      "acc": 0.64654527,
      "epoch": 0.6863267376966007,
      "grad_norm": 6.09375,
      "learning_rate": 7.828668511390439e-06,
      "loss": 1.58244209,
      "memory(GiB)": 107.26,
      "step": 27055,
      "train_speed(iter/s)": 1.633414
    },
    {
      "acc": 0.64846325,
      "epoch": 0.6864535768645358,
      "grad_norm": 5.4375,
      "learning_rate": 7.827803766680176e-06,
      "loss": 1.63761368,
      "memory(GiB)": 107.26,
      "step": 27060,
      "train_speed(iter/s)": 1.633446
    },
    {
      "acc": 0.65124149,
      "epoch": 0.6865804160324708,
      "grad_norm": 6.3125,
      "learning_rate": 7.826938897590477e-06,
      "loss": 1.63812027,
      "memory(GiB)": 107.26,
      "step": 27065,
      "train_speed(iter/s)": 1.633477
    },
    {
      "acc": 0.666012,
      "epoch": 0.6867072552004059,
      "grad_norm": 5.28125,
      "learning_rate": 7.826073904159384e-06,
      "loss": 1.53024359,
      "memory(GiB)": 107.26,
      "step": 27070,
      "train_speed(iter/s)": 1.633508
    },
    {
      "acc": 0.66201282,
      "epoch": 0.686834094368341,
      "grad_norm": 5.71875,
      "learning_rate": 7.825208786424944e-06,
      "loss": 1.61423779,
      "memory(GiB)": 107.26,
      "step": 27075,
      "train_speed(iter/s)": 1.63354
    },
    {
      "acc": 0.6322577,
      "epoch": 0.686960933536276,
      "grad_norm": 8.9375,
      "learning_rate": 7.824343544425207e-06,
      "loss": 1.70796871,
      "memory(GiB)": 107.26,
      "step": 27080,
      "train_speed(iter/s)": 1.633576
    },
    {
      "acc": 0.65276732,
      "epoch": 0.6870877727042111,
      "grad_norm": 4.59375,
      "learning_rate": 7.823478178198234e-06,
      "loss": 1.59573755,
      "memory(GiB)": 107.26,
      "step": 27085,
      "train_speed(iter/s)": 1.633607
    },
    {
      "acc": 0.66387219,
      "epoch": 0.6872146118721462,
      "grad_norm": 5.21875,
      "learning_rate": 7.822612687782083e-06,
      "loss": 1.55407057,
      "memory(GiB)": 107.26,
      "step": 27090,
      "train_speed(iter/s)": 1.633641
    },
    {
      "acc": 0.66301851,
      "epoch": 0.6873414510400812,
      "grad_norm": 5.90625,
      "learning_rate": 7.821747073214823e-06,
      "loss": 1.56343193,
      "memory(GiB)": 107.26,
      "step": 27095,
      "train_speed(iter/s)": 1.633673
    },
    {
      "acc": 0.65670996,
      "epoch": 0.6874682902080163,
      "grad_norm": 5.03125,
      "learning_rate": 7.820881334534529e-06,
      "loss": 1.55156994,
      "memory(GiB)": 107.26,
      "step": 27100,
      "train_speed(iter/s)": 1.633706
    },
    {
      "acc": 0.66631594,
      "epoch": 0.6875951293759512,
      "grad_norm": 5.09375,
      "learning_rate": 7.820015471779278e-06,
      "loss": 1.53754244,
      "memory(GiB)": 107.26,
      "step": 27105,
      "train_speed(iter/s)": 1.633738
    },
    {
      "acc": 0.66593151,
      "epoch": 0.6877219685438863,
      "grad_norm": 5.78125,
      "learning_rate": 7.819149484987159e-06,
      "loss": 1.56529617,
      "memory(GiB)": 107.26,
      "step": 27110,
      "train_speed(iter/s)": 1.633772
    },
    {
      "acc": 0.6684948,
      "epoch": 0.6878488077118214,
      "grad_norm": 5.0,
      "learning_rate": 7.818283374196259e-06,
      "loss": 1.5542263,
      "memory(GiB)": 107.26,
      "step": 27115,
      "train_speed(iter/s)": 1.633803
    },
    {
      "acc": 0.65718846,
      "epoch": 0.6879756468797564,
      "grad_norm": 6.03125,
      "learning_rate": 7.817417139444671e-06,
      "loss": 1.65624275,
      "memory(GiB)": 107.26,
      "step": 27120,
      "train_speed(iter/s)": 1.633837
    },
    {
      "acc": 0.63353791,
      "epoch": 0.6881024860476915,
      "grad_norm": 7.0,
      "learning_rate": 7.8165507807705e-06,
      "loss": 1.70971642,
      "memory(GiB)": 107.26,
      "step": 27125,
      "train_speed(iter/s)": 1.633872
    },
    {
      "acc": 0.64279776,
      "epoch": 0.6882293252156266,
      "grad_norm": 5.375,
      "learning_rate": 7.81568429821185e-06,
      "loss": 1.68047504,
      "memory(GiB)": 107.26,
      "step": 27130,
      "train_speed(iter/s)": 1.633907
    },
    {
      "acc": 0.65463715,
      "epoch": 0.6883561643835616,
      "grad_norm": 5.8125,
      "learning_rate": 7.814817691806834e-06,
      "loss": 1.66218224,
      "memory(GiB)": 107.26,
      "step": 27135,
      "train_speed(iter/s)": 1.633939
    },
    {
      "acc": 0.65490894,
      "epoch": 0.6884830035514967,
      "grad_norm": 5.5,
      "learning_rate": 7.813950961593569e-06,
      "loss": 1.5835741,
      "memory(GiB)": 107.26,
      "step": 27140,
      "train_speed(iter/s)": 1.633971
    },
    {
      "acc": 0.64625416,
      "epoch": 0.6886098427194317,
      "grad_norm": 4.6875,
      "learning_rate": 7.813084107610175e-06,
      "loss": 1.61685715,
      "memory(GiB)": 107.26,
      "step": 27145,
      "train_speed(iter/s)": 1.634001
    },
    {
      "acc": 0.66330686,
      "epoch": 0.6887366818873668,
      "grad_norm": 5.125,
      "learning_rate": 7.812217129894785e-06,
      "loss": 1.4957881,
      "memory(GiB)": 107.26,
      "step": 27150,
      "train_speed(iter/s)": 1.634032
    },
    {
      "acc": 0.66132069,
      "epoch": 0.6888635210553019,
      "grad_norm": 5.28125,
      "learning_rate": 7.811350028485531e-06,
      "loss": 1.55517197,
      "memory(GiB)": 107.26,
      "step": 27155,
      "train_speed(iter/s)": 1.634066
    },
    {
      "acc": 0.65277162,
      "epoch": 0.6889903602232369,
      "grad_norm": 6.0,
      "learning_rate": 7.810482803420549e-06,
      "loss": 1.54844151,
      "memory(GiB)": 107.26,
      "step": 27160,
      "train_speed(iter/s)": 1.634098
    },
    {
      "acc": 0.65135374,
      "epoch": 0.689117199391172,
      "grad_norm": 5.8125,
      "learning_rate": 7.809615454737984e-06,
      "loss": 1.64623528,
      "memory(GiB)": 107.26,
      "step": 27165,
      "train_speed(iter/s)": 1.634129
    },
    {
      "acc": 0.64086094,
      "epoch": 0.6892440385591071,
      "grad_norm": 6.25,
      "learning_rate": 7.808747982475991e-06,
      "loss": 1.6480648,
      "memory(GiB)": 107.26,
      "step": 27170,
      "train_speed(iter/s)": 1.634164
    },
    {
      "acc": 0.64434471,
      "epoch": 0.6893708777270421,
      "grad_norm": 7.1875,
      "learning_rate": 7.807880386672718e-06,
      "loss": 1.64645977,
      "memory(GiB)": 107.26,
      "step": 27175,
      "train_speed(iter/s)": 1.634198
    },
    {
      "acc": 0.64159942,
      "epoch": 0.6894977168949772,
      "grad_norm": 6.1875,
      "learning_rate": 7.807012667366332e-06,
      "loss": 1.67221508,
      "memory(GiB)": 107.26,
      "step": 27180,
      "train_speed(iter/s)": 1.63423
    },
    {
      "acc": 0.65396795,
      "epoch": 0.6896245560629122,
      "grad_norm": 6.90625,
      "learning_rate": 7.806144824594994e-06,
      "loss": 1.64463844,
      "memory(GiB)": 107.26,
      "step": 27185,
      "train_speed(iter/s)": 1.634261
    },
    {
      "acc": 0.64494209,
      "epoch": 0.6897513952308473,
      "grad_norm": 5.5625,
      "learning_rate": 7.805276858396879e-06,
      "loss": 1.62344456,
      "memory(GiB)": 107.26,
      "step": 27190,
      "train_speed(iter/s)": 1.634296
    },
    {
      "acc": 0.66242733,
      "epoch": 0.6898782343987824,
      "grad_norm": 5.375,
      "learning_rate": 7.804408768810164e-06,
      "loss": 1.58216953,
      "memory(GiB)": 107.26,
      "step": 27195,
      "train_speed(iter/s)": 1.634328
    },
    {
      "acc": 0.64973612,
      "epoch": 0.6900050735667174,
      "grad_norm": 5.71875,
      "learning_rate": 7.80354055587303e-06,
      "loss": 1.68611679,
      "memory(GiB)": 107.26,
      "step": 27200,
      "train_speed(iter/s)": 1.634361
    },
    {
      "acc": 0.65159464,
      "epoch": 0.6901319127346525,
      "grad_norm": 5.59375,
      "learning_rate": 7.802672219623665e-06,
      "loss": 1.59653902,
      "memory(GiB)": 107.26,
      "step": 27205,
      "train_speed(iter/s)": 1.634391
    },
    {
      "acc": 0.66741424,
      "epoch": 0.6902587519025876,
      "grad_norm": 6.40625,
      "learning_rate": 7.801803760100264e-06,
      "loss": 1.60990677,
      "memory(GiB)": 107.26,
      "step": 27210,
      "train_speed(iter/s)": 1.634424
    },
    {
      "acc": 0.65118651,
      "epoch": 0.6903855910705226,
      "grad_norm": 7.3125,
      "learning_rate": 7.800935177341022e-06,
      "loss": 1.58042345,
      "memory(GiB)": 107.26,
      "step": 27215,
      "train_speed(iter/s)": 1.634458
    },
    {
      "acc": 0.66853809,
      "epoch": 0.6905124302384577,
      "grad_norm": 5.1875,
      "learning_rate": 7.800066471384149e-06,
      "loss": 1.59965324,
      "memory(GiB)": 107.26,
      "step": 27220,
      "train_speed(iter/s)": 1.634489
    },
    {
      "acc": 0.66857328,
      "epoch": 0.6906392694063926,
      "grad_norm": 6.125,
      "learning_rate": 7.799197642267848e-06,
      "loss": 1.60582314,
      "memory(GiB)": 107.26,
      "step": 27225,
      "train_speed(iter/s)": 1.634523
    },
    {
      "acc": 0.65664892,
      "epoch": 0.6907661085743277,
      "grad_norm": 5.0,
      "learning_rate": 7.79832869003034e-06,
      "loss": 1.57414627,
      "memory(GiB)": 107.26,
      "step": 27230,
      "train_speed(iter/s)": 1.634556
    },
    {
      "acc": 0.66003609,
      "epoch": 0.6908929477422628,
      "grad_norm": 6.03125,
      "learning_rate": 7.797459614709842e-06,
      "loss": 1.61059303,
      "memory(GiB)": 107.26,
      "step": 27235,
      "train_speed(iter/s)": 1.63459
    },
    {
      "acc": 0.65616622,
      "epoch": 0.6910197869101978,
      "grad_norm": 6.0,
      "learning_rate": 7.796590416344578e-06,
      "loss": 1.61198807,
      "memory(GiB)": 107.26,
      "step": 27240,
      "train_speed(iter/s)": 1.634623
    },
    {
      "acc": 0.64562421,
      "epoch": 0.6911466260781329,
      "grad_norm": 5.65625,
      "learning_rate": 7.795721094972783e-06,
      "loss": 1.68045845,
      "memory(GiB)": 107.26,
      "step": 27245,
      "train_speed(iter/s)": 1.634654
    },
    {
      "acc": 0.63632498,
      "epoch": 0.691273465246068,
      "grad_norm": 6.09375,
      "learning_rate": 7.794851650632693e-06,
      "loss": 1.70179195,
      "memory(GiB)": 107.26,
      "step": 27250,
      "train_speed(iter/s)": 1.63469
    },
    {
      "acc": 0.65938749,
      "epoch": 0.691400304414003,
      "grad_norm": 4.9375,
      "learning_rate": 7.793982083362548e-06,
      "loss": 1.50130119,
      "memory(GiB)": 107.26,
      "step": 27255,
      "train_speed(iter/s)": 1.634714
    },
    {
      "acc": 0.67015014,
      "epoch": 0.6915271435819381,
      "grad_norm": 6.65625,
      "learning_rate": 7.7931123932006e-06,
      "loss": 1.5763669,
      "memory(GiB)": 107.26,
      "step": 27260,
      "train_speed(iter/s)": 1.634746
    },
    {
      "acc": 0.64620943,
      "epoch": 0.6916539827498731,
      "grad_norm": 5.21875,
      "learning_rate": 7.792242580185095e-06,
      "loss": 1.59284096,
      "memory(GiB)": 107.26,
      "step": 27265,
      "train_speed(iter/s)": 1.634779
    },
    {
      "acc": 0.6555275,
      "epoch": 0.6917808219178082,
      "grad_norm": 7.53125,
      "learning_rate": 7.791372644354295e-06,
      "loss": 1.55669098,
      "memory(GiB)": 107.26,
      "step": 27270,
      "train_speed(iter/s)": 1.634814
    },
    {
      "acc": 0.64665003,
      "epoch": 0.6919076610857433,
      "grad_norm": 6.3125,
      "learning_rate": 7.790502585746464e-06,
      "loss": 1.64027348,
      "memory(GiB)": 107.26,
      "step": 27275,
      "train_speed(iter/s)": 1.634846
    },
    {
      "acc": 0.66824036,
      "epoch": 0.6920345002536783,
      "grad_norm": 5.71875,
      "learning_rate": 7.789632404399872e-06,
      "loss": 1.54462252,
      "memory(GiB)": 107.26,
      "step": 27280,
      "train_speed(iter/s)": 1.634878
    },
    {
      "acc": 0.6542613,
      "epoch": 0.6921613394216134,
      "grad_norm": 7.1875,
      "learning_rate": 7.788762100352791e-06,
      "loss": 1.6289505,
      "memory(GiB)": 107.26,
      "step": 27285,
      "train_speed(iter/s)": 1.634912
    },
    {
      "acc": 0.66637259,
      "epoch": 0.6922881785895485,
      "grad_norm": 5.75,
      "learning_rate": 7.787891673643501e-06,
      "loss": 1.5987524,
      "memory(GiB)": 107.26,
      "step": 27290,
      "train_speed(iter/s)": 1.634943
    },
    {
      "acc": 0.64926505,
      "epoch": 0.6924150177574835,
      "grad_norm": 6.65625,
      "learning_rate": 7.78702112431029e-06,
      "loss": 1.64018211,
      "memory(GiB)": 107.26,
      "step": 27295,
      "train_speed(iter/s)": 1.634974
    },
    {
      "acc": 0.65669932,
      "epoch": 0.6925418569254186,
      "grad_norm": 5.46875,
      "learning_rate": 7.786150452391446e-06,
      "loss": 1.57897968,
      "memory(GiB)": 107.26,
      "step": 27300,
      "train_speed(iter/s)": 1.635007
    },
    {
      "acc": 0.66868744,
      "epoch": 0.6926686960933536,
      "grad_norm": 5.96875,
      "learning_rate": 7.785279657925265e-06,
      "loss": 1.56595135,
      "memory(GiB)": 107.26,
      "step": 27305,
      "train_speed(iter/s)": 1.635041
    },
    {
      "acc": 0.64119015,
      "epoch": 0.6927955352612887,
      "grad_norm": 5.75,
      "learning_rate": 7.784408740950051e-06,
      "loss": 1.64192123,
      "memory(GiB)": 107.26,
      "step": 27310,
      "train_speed(iter/s)": 1.635073
    },
    {
      "acc": 0.66927886,
      "epoch": 0.6929223744292238,
      "grad_norm": 6.0,
      "learning_rate": 7.783537701504109e-06,
      "loss": 1.58278618,
      "memory(GiB)": 107.26,
      "step": 27315,
      "train_speed(iter/s)": 1.635106
    },
    {
      "acc": 0.66143999,
      "epoch": 0.6930492135971588,
      "grad_norm": 5.25,
      "learning_rate": 7.782666539625749e-06,
      "loss": 1.65351219,
      "memory(GiB)": 107.26,
      "step": 27320,
      "train_speed(iter/s)": 1.635141
    },
    {
      "acc": 0.66218901,
      "epoch": 0.6931760527650939,
      "grad_norm": 6.21875,
      "learning_rate": 7.781795255353293e-06,
      "loss": 1.56159458,
      "memory(GiB)": 107.26,
      "step": 27325,
      "train_speed(iter/s)": 1.635173
    },
    {
      "acc": 0.66439648,
      "epoch": 0.693302891933029,
      "grad_norm": 5.9375,
      "learning_rate": 7.780923848725061e-06,
      "loss": 1.57512035,
      "memory(GiB)": 107.26,
      "step": 27330,
      "train_speed(iter/s)": 1.635208
    },
    {
      "acc": 0.64612985,
      "epoch": 0.693429731100964,
      "grad_norm": 5.1875,
      "learning_rate": 7.780052319779382e-06,
      "loss": 1.58392906,
      "memory(GiB)": 107.26,
      "step": 27335,
      "train_speed(iter/s)": 1.635239
    },
    {
      "acc": 0.64886003,
      "epoch": 0.693556570268899,
      "grad_norm": 5.90625,
      "learning_rate": 7.779180668554591e-06,
      "loss": 1.66179276,
      "memory(GiB)": 107.26,
      "step": 27340,
      "train_speed(iter/s)": 1.63527
    },
    {
      "acc": 0.66692924,
      "epoch": 0.693683409436834,
      "grad_norm": 5.71875,
      "learning_rate": 7.778308895089024e-06,
      "loss": 1.56435852,
      "memory(GiB)": 107.26,
      "step": 27345,
      "train_speed(iter/s)": 1.635303
    },
    {
      "acc": 0.66735201,
      "epoch": 0.6938102486047691,
      "grad_norm": 5.34375,
      "learning_rate": 7.77743699942103e-06,
      "loss": 1.5706811,
      "memory(GiB)": 107.26,
      "step": 27350,
      "train_speed(iter/s)": 1.635337
    },
    {
      "acc": 0.66817093,
      "epoch": 0.6939370877727042,
      "grad_norm": 6.25,
      "learning_rate": 7.776564981588955e-06,
      "loss": 1.56316748,
      "memory(GiB)": 107.26,
      "step": 27355,
      "train_speed(iter/s)": 1.635369
    },
    {
      "acc": 0.65437531,
      "epoch": 0.6940639269406392,
      "grad_norm": 6.6875,
      "learning_rate": 7.775692841631154e-06,
      "loss": 1.69971409,
      "memory(GiB)": 107.26,
      "step": 27360,
      "train_speed(iter/s)": 1.635403
    },
    {
      "acc": 0.64340844,
      "epoch": 0.6941907661085743,
      "grad_norm": 4.34375,
      "learning_rate": 7.774820579585993e-06,
      "loss": 1.61681747,
      "memory(GiB)": 107.26,
      "step": 27365,
      "train_speed(iter/s)": 1.635437
    },
    {
      "acc": 0.65908537,
      "epoch": 0.6943176052765094,
      "grad_norm": 5.9375,
      "learning_rate": 7.773948195491831e-06,
      "loss": 1.58944664,
      "memory(GiB)": 107.26,
      "step": 27370,
      "train_speed(iter/s)": 1.635469
    },
    {
      "acc": 0.64517727,
      "epoch": 0.6944444444444444,
      "grad_norm": 5.65625,
      "learning_rate": 7.773075689387044e-06,
      "loss": 1.67690811,
      "memory(GiB)": 107.26,
      "step": 27375,
      "train_speed(iter/s)": 1.635504
    },
    {
      "acc": 0.66697702,
      "epoch": 0.6945712836123795,
      "grad_norm": 7.5625,
      "learning_rate": 7.772203061310008e-06,
      "loss": 1.56519604,
      "memory(GiB)": 107.26,
      "step": 27380,
      "train_speed(iter/s)": 1.635535
    },
    {
      "acc": 0.648909,
      "epoch": 0.6946981227803145,
      "grad_norm": 5.875,
      "learning_rate": 7.771330311299104e-06,
      "loss": 1.70168056,
      "memory(GiB)": 107.26,
      "step": 27385,
      "train_speed(iter/s)": 1.635567
    },
    {
      "acc": 0.65259457,
      "epoch": 0.6948249619482496,
      "grad_norm": 4.65625,
      "learning_rate": 7.770457439392719e-06,
      "loss": 1.60979671,
      "memory(GiB)": 107.26,
      "step": 27390,
      "train_speed(iter/s)": 1.6356
    },
    {
      "acc": 0.6404027,
      "epoch": 0.6949518011161847,
      "grad_norm": 5.03125,
      "learning_rate": 7.769584445629247e-06,
      "loss": 1.7007122,
      "memory(GiB)": 107.26,
      "step": 27395,
      "train_speed(iter/s)": 1.635633
    },
    {
      "acc": 0.65212255,
      "epoch": 0.6950786402841197,
      "grad_norm": 6.75,
      "learning_rate": 7.768711330047087e-06,
      "loss": 1.60189686,
      "memory(GiB)": 107.26,
      "step": 27400,
      "train_speed(iter/s)": 1.635666
    },
    {
      "acc": 0.65985851,
      "epoch": 0.6952054794520548,
      "grad_norm": 5.8125,
      "learning_rate": 7.767838092684638e-06,
      "loss": 1.62681122,
      "memory(GiB)": 107.26,
      "step": 27405,
      "train_speed(iter/s)": 1.635699
    },
    {
      "acc": 0.67035189,
      "epoch": 0.6953323186199899,
      "grad_norm": 4.34375,
      "learning_rate": 7.766964733580316e-06,
      "loss": 1.54356546,
      "memory(GiB)": 107.26,
      "step": 27410,
      "train_speed(iter/s)": 1.635729
    },
    {
      "acc": 0.66668577,
      "epoch": 0.6954591577879249,
      "grad_norm": 5.0,
      "learning_rate": 7.76609125277253e-06,
      "loss": 1.59488697,
      "memory(GiB)": 107.26,
      "step": 27415,
      "train_speed(iter/s)": 1.635762
    },
    {
      "acc": 0.6549706,
      "epoch": 0.69558599695586,
      "grad_norm": 7.9375,
      "learning_rate": 7.7652176502997e-06,
      "loss": 1.58501558,
      "memory(GiB)": 107.26,
      "step": 27420,
      "train_speed(iter/s)": 1.635793
    },
    {
      "acc": 0.66406651,
      "epoch": 0.695712836123795,
      "grad_norm": 5.25,
      "learning_rate": 7.764343926200254e-06,
      "loss": 1.53203926,
      "memory(GiB)": 107.26,
      "step": 27425,
      "train_speed(iter/s)": 1.635825
    },
    {
      "acc": 0.6538558,
      "epoch": 0.6958396752917301,
      "grad_norm": 5.375,
      "learning_rate": 7.763470080512617e-06,
      "loss": 1.60885811,
      "memory(GiB)": 107.26,
      "step": 27430,
      "train_speed(iter/s)": 1.635859
    },
    {
      "acc": 0.65442615,
      "epoch": 0.6959665144596652,
      "grad_norm": 6.5625,
      "learning_rate": 7.762596113275229e-06,
      "loss": 1.59839926,
      "memory(GiB)": 107.26,
      "step": 27435,
      "train_speed(iter/s)": 1.635892
    },
    {
      "acc": 0.65658865,
      "epoch": 0.6960933536276002,
      "grad_norm": 6.96875,
      "learning_rate": 7.761722024526533e-06,
      "loss": 1.64569206,
      "memory(GiB)": 107.26,
      "step": 27440,
      "train_speed(iter/s)": 1.635927
    },
    {
      "acc": 0.66237354,
      "epoch": 0.6962201927955353,
      "grad_norm": 5.375,
      "learning_rate": 7.760847814304969e-06,
      "loss": 1.61346474,
      "memory(GiB)": 107.26,
      "step": 27445,
      "train_speed(iter/s)": 1.635958
    },
    {
      "acc": 0.65442677,
      "epoch": 0.6963470319634704,
      "grad_norm": 8.3125,
      "learning_rate": 7.759973482648992e-06,
      "loss": 1.59562788,
      "memory(GiB)": 107.26,
      "step": 27450,
      "train_speed(iter/s)": 1.635991
    },
    {
      "acc": 0.65584602,
      "epoch": 0.6964738711314054,
      "grad_norm": 4.8125,
      "learning_rate": 7.75909902959706e-06,
      "loss": 1.65837135,
      "memory(GiB)": 107.26,
      "step": 27455,
      "train_speed(iter/s)": 1.636025
    },
    {
      "acc": 0.64058361,
      "epoch": 0.6966007102993405,
      "grad_norm": 5.21875,
      "learning_rate": 7.758224455187632e-06,
      "loss": 1.67555428,
      "memory(GiB)": 107.26,
      "step": 27460,
      "train_speed(iter/s)": 1.636058
    },
    {
      "acc": 0.65166955,
      "epoch": 0.6967275494672754,
      "grad_norm": 6.0625,
      "learning_rate": 7.75734975945918e-06,
      "loss": 1.67998333,
      "memory(GiB)": 107.26,
      "step": 27465,
      "train_speed(iter/s)": 1.636087
    },
    {
      "acc": 0.65085659,
      "epoch": 0.6968543886352105,
      "grad_norm": 7.15625,
      "learning_rate": 7.756474942450174e-06,
      "loss": 1.64790001,
      "memory(GiB)": 107.26,
      "step": 27470,
      "train_speed(iter/s)": 1.63612
    },
    {
      "acc": 0.66765256,
      "epoch": 0.6969812278031456,
      "grad_norm": 6.5625,
      "learning_rate": 7.755600004199094e-06,
      "loss": 1.54629698,
      "memory(GiB)": 107.26,
      "step": 27475,
      "train_speed(iter/s)": 1.636153
    },
    {
      "acc": 0.66047487,
      "epoch": 0.6971080669710806,
      "grad_norm": 6.0,
      "learning_rate": 7.754724944744423e-06,
      "loss": 1.60178394,
      "memory(GiB)": 107.26,
      "step": 27480,
      "train_speed(iter/s)": 1.636184
    },
    {
      "acc": 0.65821381,
      "epoch": 0.6972349061390157,
      "grad_norm": 6.5625,
      "learning_rate": 7.753849764124648e-06,
      "loss": 1.66496429,
      "memory(GiB)": 107.26,
      "step": 27485,
      "train_speed(iter/s)": 1.636215
    },
    {
      "acc": 0.64803553,
      "epoch": 0.6973617453069508,
      "grad_norm": 6.03125,
      "learning_rate": 7.752974462378268e-06,
      "loss": 1.63399925,
      "memory(GiB)": 107.26,
      "step": 27490,
      "train_speed(iter/s)": 1.636247
    },
    {
      "acc": 0.65273108,
      "epoch": 0.6974885844748858,
      "grad_norm": 5.0,
      "learning_rate": 7.752099039543778e-06,
      "loss": 1.62570057,
      "memory(GiB)": 107.26,
      "step": 27495,
      "train_speed(iter/s)": 1.636279
    },
    {
      "acc": 0.65321431,
      "epoch": 0.6976154236428209,
      "grad_norm": 6.03125,
      "learning_rate": 7.751223495659685e-06,
      "loss": 1.57066135,
      "memory(GiB)": 107.26,
      "step": 27500,
      "train_speed(iter/s)": 1.636314
    },
    {
      "acc": 0.65747986,
      "epoch": 0.6977422628107559,
      "grad_norm": 5.25,
      "learning_rate": 7.7503478307645e-06,
      "loss": 1.60095863,
      "memory(GiB)": 107.26,
      "step": 27505,
      "train_speed(iter/s)": 1.636347
    },
    {
      "acc": 0.66565981,
      "epoch": 0.697869101978691,
      "grad_norm": 5.34375,
      "learning_rate": 7.74947204489674e-06,
      "loss": 1.55961475,
      "memory(GiB)": 107.26,
      "step": 27510,
      "train_speed(iter/s)": 1.636379
    },
    {
      "acc": 0.64922576,
      "epoch": 0.6979959411466261,
      "grad_norm": 6.53125,
      "learning_rate": 7.748596138094922e-06,
      "loss": 1.66757259,
      "memory(GiB)": 107.26,
      "step": 27515,
      "train_speed(iter/s)": 1.636409
    },
    {
      "acc": 0.62690125,
      "epoch": 0.6981227803145611,
      "grad_norm": 5.84375,
      "learning_rate": 7.747720110397573e-06,
      "loss": 1.68180904,
      "memory(GiB)": 107.26,
      "step": 27520,
      "train_speed(iter/s)": 1.636439
    },
    {
      "acc": 0.65505948,
      "epoch": 0.6982496194824962,
      "grad_norm": 5.4375,
      "learning_rate": 7.746843961843226e-06,
      "loss": 1.64896736,
      "memory(GiB)": 107.26,
      "step": 27525,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.67042408,
      "epoch": 0.6983764586504313,
      "grad_norm": 6.28125,
      "learning_rate": 7.74596769247042e-06,
      "loss": 1.5555418,
      "memory(GiB)": 107.26,
      "step": 27530,
      "train_speed(iter/s)": 1.636502
    },
    {
      "acc": 0.66162472,
      "epoch": 0.6985032978183663,
      "grad_norm": 5.875,
      "learning_rate": 7.745091302317694e-06,
      "loss": 1.58649368,
      "memory(GiB)": 107.26,
      "step": 27535,
      "train_speed(iter/s)": 1.636536
    },
    {
      "acc": 0.66233625,
      "epoch": 0.6986301369863014,
      "grad_norm": 5.78125,
      "learning_rate": 7.744214791423597e-06,
      "loss": 1.6265234,
      "memory(GiB)": 107.26,
      "step": 27540,
      "train_speed(iter/s)": 1.636569
    },
    {
      "acc": 0.63770685,
      "epoch": 0.6987569761542364,
      "grad_norm": 5.65625,
      "learning_rate": 7.74333815982668e-06,
      "loss": 1.65318565,
      "memory(GiB)": 107.26,
      "step": 27545,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.64911213,
      "epoch": 0.6988838153221715,
      "grad_norm": 5.1875,
      "learning_rate": 7.742461407565504e-06,
      "loss": 1.58032379,
      "memory(GiB)": 107.26,
      "step": 27550,
      "train_speed(iter/s)": 1.636635
    },
    {
      "acc": 0.65556951,
      "epoch": 0.6990106544901066,
      "grad_norm": 5.03125,
      "learning_rate": 7.741584534678632e-06,
      "loss": 1.64268456,
      "memory(GiB)": 107.26,
      "step": 27555,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.66315279,
      "epoch": 0.6991374936580416,
      "grad_norm": 5.21875,
      "learning_rate": 7.74070754120463e-06,
      "loss": 1.6171772,
      "memory(GiB)": 107.26,
      "step": 27560,
      "train_speed(iter/s)": 1.636702
    },
    {
      "acc": 0.65602617,
      "epoch": 0.6992643328259767,
      "grad_norm": 5.65625,
      "learning_rate": 7.739830427182073e-06,
      "loss": 1.64129524,
      "memory(GiB)": 107.26,
      "step": 27565,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.65914369,
      "epoch": 0.6993911719939118,
      "grad_norm": 6.375,
      "learning_rate": 7.738953192649544e-06,
      "loss": 1.60509415,
      "memory(GiB)": 107.26,
      "step": 27570,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.65117669,
      "epoch": 0.6995180111618468,
      "grad_norm": 5.34375,
      "learning_rate": 7.738075837645625e-06,
      "loss": 1.62055225,
      "memory(GiB)": 107.26,
      "step": 27575,
      "train_speed(iter/s)": 1.636799
    },
    {
      "acc": 0.66142426,
      "epoch": 0.6996448503297819,
      "grad_norm": 6.28125,
      "learning_rate": 7.737198362208904e-06,
      "loss": 1.61629028,
      "memory(GiB)": 107.26,
      "step": 27580,
      "train_speed(iter/s)": 1.636829
    },
    {
      "acc": 0.65815783,
      "epoch": 0.6997716894977168,
      "grad_norm": 7.15625,
      "learning_rate": 7.736320766377978e-06,
      "loss": 1.6959465,
      "memory(GiB)": 107.26,
      "step": 27585,
      "train_speed(iter/s)": 1.636861
    },
    {
      "acc": 0.64031978,
      "epoch": 0.6998985286656519,
      "grad_norm": 5.5625,
      "learning_rate": 7.735443050191452e-06,
      "loss": 1.59045753,
      "memory(GiB)": 107.26,
      "step": 27590,
      "train_speed(iter/s)": 1.636893
    },
    {
      "acc": 0.64829602,
      "epoch": 0.700025367833587,
      "grad_norm": 5.5,
      "learning_rate": 7.734565213687923e-06,
      "loss": 1.58005152,
      "memory(GiB)": 107.26,
      "step": 27595,
      "train_speed(iter/s)": 1.636925
    },
    {
      "acc": 0.65510674,
      "epoch": 0.700152207001522,
      "grad_norm": 4.9375,
      "learning_rate": 7.733687256906009e-06,
      "loss": 1.64164505,
      "memory(GiB)": 107.26,
      "step": 27600,
      "train_speed(iter/s)": 1.636958
    },
    {
      "acc": 0.66025529,
      "epoch": 0.7002790461694571,
      "grad_norm": 5.15625,
      "learning_rate": 7.732809179884324e-06,
      "loss": 1.59567947,
      "memory(GiB)": 107.26,
      "step": 27605,
      "train_speed(iter/s)": 1.63699
    },
    {
      "acc": 0.6611599,
      "epoch": 0.7004058853373922,
      "grad_norm": 6.09375,
      "learning_rate": 7.73193098266149e-06,
      "loss": 1.58921146,
      "memory(GiB)": 107.26,
      "step": 27610,
      "train_speed(iter/s)": 1.637022
    },
    {
      "acc": 0.64529257,
      "epoch": 0.7005327245053272,
      "grad_norm": 6.71875,
      "learning_rate": 7.731052665276135e-06,
      "loss": 1.67083435,
      "memory(GiB)": 107.26,
      "step": 27615,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.66823578,
      "epoch": 0.7006595636732623,
      "grad_norm": 6.75,
      "learning_rate": 7.730174227766892e-06,
      "loss": 1.57269411,
      "memory(GiB)": 107.26,
      "step": 27620,
      "train_speed(iter/s)": 1.637089
    },
    {
      "acc": 0.65937729,
      "epoch": 0.7007864028411973,
      "grad_norm": 6.28125,
      "learning_rate": 7.729295670172394e-06,
      "loss": 1.59853535,
      "memory(GiB)": 107.26,
      "step": 27625,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.66014252,
      "epoch": 0.7009132420091324,
      "grad_norm": 6.09375,
      "learning_rate": 7.728416992531287e-06,
      "loss": 1.60750999,
      "memory(GiB)": 107.26,
      "step": 27630,
      "train_speed(iter/s)": 1.637153
    },
    {
      "acc": 0.65903234,
      "epoch": 0.7010400811770675,
      "grad_norm": 6.4375,
      "learning_rate": 7.72753819488222e-06,
      "loss": 1.64045296,
      "memory(GiB)": 107.26,
      "step": 27635,
      "train_speed(iter/s)": 1.637185
    },
    {
      "acc": 0.65994263,
      "epoch": 0.7011669203450025,
      "grad_norm": 5.3125,
      "learning_rate": 7.726659277263848e-06,
      "loss": 1.58435211,
      "memory(GiB)": 107.26,
      "step": 27640,
      "train_speed(iter/s)": 1.637217
    },
    {
      "acc": 0.65562024,
      "epoch": 0.7012937595129376,
      "grad_norm": 9.9375,
      "learning_rate": 7.725780239714824e-06,
      "loss": 1.66056213,
      "memory(GiB)": 107.26,
      "step": 27645,
      "train_speed(iter/s)": 1.637248
    },
    {
      "acc": 0.65208654,
      "epoch": 0.7014205986808727,
      "grad_norm": 6.1875,
      "learning_rate": 7.724901082273817e-06,
      "loss": 1.57872782,
      "memory(GiB)": 107.26,
      "step": 27650,
      "train_speed(iter/s)": 1.63728
    },
    {
      "acc": 0.65807223,
      "epoch": 0.7015474378488077,
      "grad_norm": 6.09375,
      "learning_rate": 7.724021804979493e-06,
      "loss": 1.57590771,
      "memory(GiB)": 107.26,
      "step": 27655,
      "train_speed(iter/s)": 1.63731
    },
    {
      "acc": 0.65873418,
      "epoch": 0.7016742770167428,
      "grad_norm": 6.15625,
      "learning_rate": 7.723142407870532e-06,
      "loss": 1.61198368,
      "memory(GiB)": 107.26,
      "step": 27660,
      "train_speed(iter/s)": 1.637341
    },
    {
      "acc": 0.67246671,
      "epoch": 0.7018011161846778,
      "grad_norm": 5.59375,
      "learning_rate": 7.722262890985605e-06,
      "loss": 1.60879211,
      "memory(GiB)": 107.26,
      "step": 27665,
      "train_speed(iter/s)": 1.637373
    },
    {
      "acc": 0.64914808,
      "epoch": 0.7019279553526129,
      "grad_norm": 5.5,
      "learning_rate": 7.721383254363407e-06,
      "loss": 1.65706635,
      "memory(GiB)": 107.26,
      "step": 27670,
      "train_speed(iter/s)": 1.637405
    },
    {
      "acc": 0.67284093,
      "epoch": 0.702054794520548,
      "grad_norm": 6.875,
      "learning_rate": 7.72050349804262e-06,
      "loss": 1.52197485,
      "memory(GiB)": 107.26,
      "step": 27675,
      "train_speed(iter/s)": 1.637436
    },
    {
      "acc": 0.66381187,
      "epoch": 0.702181633688483,
      "grad_norm": 5.09375,
      "learning_rate": 7.719623622061943e-06,
      "loss": 1.5586031,
      "memory(GiB)": 107.26,
      "step": 27680,
      "train_speed(iter/s)": 1.637468
    },
    {
      "acc": 0.64491067,
      "epoch": 0.7023084728564181,
      "grad_norm": 5.6875,
      "learning_rate": 7.718743626460076e-06,
      "loss": 1.66212597,
      "memory(GiB)": 107.26,
      "step": 27685,
      "train_speed(iter/s)": 1.6375
    },
    {
      "acc": 0.66848497,
      "epoch": 0.7024353120243532,
      "grad_norm": 5.28125,
      "learning_rate": 7.71786351127573e-06,
      "loss": 1.53633022,
      "memory(GiB)": 107.26,
      "step": 27690,
      "train_speed(iter/s)": 1.637532
    },
    {
      "acc": 0.64966307,
      "epoch": 0.7025621511922882,
      "grad_norm": 5.59375,
      "learning_rate": 7.71698327654761e-06,
      "loss": 1.59696503,
      "memory(GiB)": 107.26,
      "step": 27695,
      "train_speed(iter/s)": 1.637563
    },
    {
      "acc": 0.65377159,
      "epoch": 0.7026889903602233,
      "grad_norm": 6.0625,
      "learning_rate": 7.716102922314435e-06,
      "loss": 1.60128994,
      "memory(GiB)": 107.26,
      "step": 27700,
      "train_speed(iter/s)": 1.637593
    },
    {
      "acc": 0.6383997,
      "epoch": 0.7028158295281582,
      "grad_norm": 5.59375,
      "learning_rate": 7.715222448614926e-06,
      "loss": 1.7098875,
      "memory(GiB)": 107.26,
      "step": 27705,
      "train_speed(iter/s)": 1.637624
    },
    {
      "acc": 0.65988851,
      "epoch": 0.7029426686960933,
      "grad_norm": 6.84375,
      "learning_rate": 7.714341855487812e-06,
      "loss": 1.61131897,
      "memory(GiB)": 107.26,
      "step": 27710,
      "train_speed(iter/s)": 1.637653
    },
    {
      "acc": 0.64717531,
      "epoch": 0.7030695078640284,
      "grad_norm": 5.0625,
      "learning_rate": 7.713461142971824e-06,
      "loss": 1.5954669,
      "memory(GiB)": 107.26,
      "step": 27715,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.6708745,
      "epoch": 0.7031963470319634,
      "grad_norm": 5.9375,
      "learning_rate": 7.712580311105701e-06,
      "loss": 1.59094448,
      "memory(GiB)": 107.26,
      "step": 27720,
      "train_speed(iter/s)": 1.637716
    },
    {
      "acc": 0.65302525,
      "epoch": 0.7033231861998985,
      "grad_norm": 5.1875,
      "learning_rate": 7.711699359928184e-06,
      "loss": 1.55217676,
      "memory(GiB)": 107.26,
      "step": 27725,
      "train_speed(iter/s)": 1.637749
    },
    {
      "acc": 0.6635447,
      "epoch": 0.7034500253678336,
      "grad_norm": 6.5625,
      "learning_rate": 7.710818289478024e-06,
      "loss": 1.58479214,
      "memory(GiB)": 107.26,
      "step": 27730,
      "train_speed(iter/s)": 1.637783
    },
    {
      "acc": 0.65795183,
      "epoch": 0.7035768645357686,
      "grad_norm": 5.625,
      "learning_rate": 7.709937099793971e-06,
      "loss": 1.6035223,
      "memory(GiB)": 107.26,
      "step": 27735,
      "train_speed(iter/s)": 1.637814
    },
    {
      "acc": 0.64138708,
      "epoch": 0.7037037037037037,
      "grad_norm": 6.34375,
      "learning_rate": 7.709055790914787e-06,
      "loss": 1.64339561,
      "memory(GiB)": 107.26,
      "step": 27740,
      "train_speed(iter/s)": 1.637848
    },
    {
      "acc": 0.65108232,
      "epoch": 0.7038305428716387,
      "grad_norm": 5.84375,
      "learning_rate": 7.708174362879234e-06,
      "loss": 1.60807934,
      "memory(GiB)": 107.26,
      "step": 27745,
      "train_speed(iter/s)": 1.637878
    },
    {
      "acc": 0.64732895,
      "epoch": 0.7039573820395738,
      "grad_norm": 5.34375,
      "learning_rate": 7.70729281572608e-06,
      "loss": 1.66552467,
      "memory(GiB)": 107.26,
      "step": 27750,
      "train_speed(iter/s)": 1.637911
    },
    {
      "acc": 0.63947697,
      "epoch": 0.7040842212075089,
      "grad_norm": 5.375,
      "learning_rate": 7.706411149494102e-06,
      "loss": 1.6603775,
      "memory(GiB)": 107.26,
      "step": 27755,
      "train_speed(iter/s)": 1.637942
    },
    {
      "acc": 0.6579258,
      "epoch": 0.7042110603754439,
      "grad_norm": 6.03125,
      "learning_rate": 7.705529364222079e-06,
      "loss": 1.63553219,
      "memory(GiB)": 107.26,
      "step": 27760,
      "train_speed(iter/s)": 1.637975
    },
    {
      "acc": 0.65640154,
      "epoch": 0.704337899543379,
      "grad_norm": 6.84375,
      "learning_rate": 7.704647459948793e-06,
      "loss": 1.5791626,
      "memory(GiB)": 107.26,
      "step": 27765,
      "train_speed(iter/s)": 1.638007
    },
    {
      "acc": 0.65575662,
      "epoch": 0.7044647387113141,
      "grad_norm": 5.40625,
      "learning_rate": 7.703765436713038e-06,
      "loss": 1.60827332,
      "memory(GiB)": 107.26,
      "step": 27770,
      "train_speed(iter/s)": 1.638041
    },
    {
      "acc": 0.67407074,
      "epoch": 0.7045915778792491,
      "grad_norm": 5.75,
      "learning_rate": 7.702883294553607e-06,
      "loss": 1.53509302,
      "memory(GiB)": 107.26,
      "step": 27775,
      "train_speed(iter/s)": 1.638073
    },
    {
      "acc": 0.66179962,
      "epoch": 0.7047184170471842,
      "grad_norm": 5.34375,
      "learning_rate": 7.702001033509302e-06,
      "loss": 1.53914547,
      "memory(GiB)": 107.26,
      "step": 27780,
      "train_speed(iter/s)": 1.638104
    },
    {
      "acc": 0.63739076,
      "epoch": 0.7048452562151192,
      "grad_norm": 4.84375,
      "learning_rate": 7.701118653618927e-06,
      "loss": 1.63914642,
      "memory(GiB)": 107.26,
      "step": 27785,
      "train_speed(iter/s)": 1.638136
    },
    {
      "acc": 0.64568243,
      "epoch": 0.7049720953830543,
      "grad_norm": 5.78125,
      "learning_rate": 7.700236154921294e-06,
      "loss": 1.60901546,
      "memory(GiB)": 107.26,
      "step": 27790,
      "train_speed(iter/s)": 1.638167
    },
    {
      "acc": 0.64461689,
      "epoch": 0.7050989345509894,
      "grad_norm": 5.0625,
      "learning_rate": 7.699353537455222e-06,
      "loss": 1.63301888,
      "memory(GiB)": 107.26,
      "step": 27795,
      "train_speed(iter/s)": 1.638199
    },
    {
      "acc": 0.64643307,
      "epoch": 0.7052257737189244,
      "grad_norm": 5.15625,
      "learning_rate": 7.698470801259526e-06,
      "loss": 1.65585098,
      "memory(GiB)": 107.26,
      "step": 27800,
      "train_speed(iter/s)": 1.638232
    },
    {
      "acc": 0.66306343,
      "epoch": 0.7053526128868595,
      "grad_norm": 5.9375,
      "learning_rate": 7.697587946373037e-06,
      "loss": 1.63212433,
      "memory(GiB)": 107.26,
      "step": 27805,
      "train_speed(iter/s)": 1.638265
    },
    {
      "acc": 0.65107331,
      "epoch": 0.7054794520547946,
      "grad_norm": 5.0,
      "learning_rate": 7.696704972834589e-06,
      "loss": 1.59155254,
      "memory(GiB)": 107.26,
      "step": 27810,
      "train_speed(iter/s)": 1.638296
    },
    {
      "acc": 0.66268091,
      "epoch": 0.7056062912227296,
      "grad_norm": 5.375,
      "learning_rate": 7.695821880683012e-06,
      "loss": 1.62391586,
      "memory(GiB)": 107.26,
      "step": 27815,
      "train_speed(iter/s)": 1.638329
    },
    {
      "acc": 0.65237837,
      "epoch": 0.7057331303906647,
      "grad_norm": 6.59375,
      "learning_rate": 7.694938669957156e-06,
      "loss": 1.58336897,
      "memory(GiB)": 107.26,
      "step": 27820,
      "train_speed(iter/s)": 1.638361
    },
    {
      "acc": 0.6549633,
      "epoch": 0.7058599695585996,
      "grad_norm": 5.875,
      "learning_rate": 7.694055340695862e-06,
      "loss": 1.59736347,
      "memory(GiB)": 107.26,
      "step": 27825,
      "train_speed(iter/s)": 1.638391
    },
    {
      "acc": 0.66499166,
      "epoch": 0.7059868087265347,
      "grad_norm": 7.625,
      "learning_rate": 7.693171892937991e-06,
      "loss": 1.58155193,
      "memory(GiB)": 107.26,
      "step": 27830,
      "train_speed(iter/s)": 1.638425
    },
    {
      "acc": 0.65267334,
      "epoch": 0.7061136478944698,
      "grad_norm": 5.28125,
      "learning_rate": 7.692288326722393e-06,
      "loss": 1.65204754,
      "memory(GiB)": 107.26,
      "step": 27835,
      "train_speed(iter/s)": 1.638457
    },
    {
      "acc": 0.65091667,
      "epoch": 0.7062404870624048,
      "grad_norm": 6.125,
      "learning_rate": 7.691404642087933e-06,
      "loss": 1.59223194,
      "memory(GiB)": 107.26,
      "step": 27840,
      "train_speed(iter/s)": 1.637969
    },
    {
      "acc": 0.6709857,
      "epoch": 0.7063673262303399,
      "grad_norm": 6.125,
      "learning_rate": 7.690520839073484e-06,
      "loss": 1.57745066,
      "memory(GiB)": 107.26,
      "step": 27845,
      "train_speed(iter/s)": 1.637998
    },
    {
      "acc": 0.65187516,
      "epoch": 0.706494165398275,
      "grad_norm": 6.15625,
      "learning_rate": 7.689636917717913e-06,
      "loss": 1.5948163,
      "memory(GiB)": 107.26,
      "step": 27850,
      "train_speed(iter/s)": 1.637483
    },
    {
      "acc": 0.66029778,
      "epoch": 0.70662100456621,
      "grad_norm": 5.28125,
      "learning_rate": 7.688752878060103e-06,
      "loss": 1.62348709,
      "memory(GiB)": 107.26,
      "step": 27855,
      "train_speed(iter/s)": 1.637514
    },
    {
      "acc": 0.65179682,
      "epoch": 0.7067478437341451,
      "grad_norm": 5.0625,
      "learning_rate": 7.687868720138939e-06,
      "loss": 1.56717014,
      "memory(GiB)": 107.26,
      "step": 27860,
      "train_speed(iter/s)": 1.637547
    },
    {
      "acc": 0.65384226,
      "epoch": 0.7068746829020801,
      "grad_norm": 6.90625,
      "learning_rate": 7.686984443993304e-06,
      "loss": 1.6424366,
      "memory(GiB)": 107.26,
      "step": 27865,
      "train_speed(iter/s)": 1.637578
    },
    {
      "acc": 0.64773078,
      "epoch": 0.7070015220700152,
      "grad_norm": 5.40625,
      "learning_rate": 7.6861000496621e-06,
      "loss": 1.63153687,
      "memory(GiB)": 107.26,
      "step": 27870,
      "train_speed(iter/s)": 1.637609
    },
    {
      "acc": 0.65964775,
      "epoch": 0.7071283612379503,
      "grad_norm": 6.21875,
      "learning_rate": 7.685215537184223e-06,
      "loss": 1.65231895,
      "memory(GiB)": 107.26,
      "step": 27875,
      "train_speed(iter/s)": 1.637639
    },
    {
      "acc": 0.65603256,
      "epoch": 0.7072552004058853,
      "grad_norm": 6.0,
      "learning_rate": 7.684330906598577e-06,
      "loss": 1.66472816,
      "memory(GiB)": 107.26,
      "step": 27880,
      "train_speed(iter/s)": 1.63767
    },
    {
      "acc": 0.66592264,
      "epoch": 0.7073820395738204,
      "grad_norm": 5.4375,
      "learning_rate": 7.683446157944075e-06,
      "loss": 1.57878933,
      "memory(GiB)": 107.26,
      "step": 27885,
      "train_speed(iter/s)": 1.637699
    },
    {
      "acc": 0.6455842,
      "epoch": 0.7075088787417555,
      "grad_norm": 6.0625,
      "learning_rate": 7.682561291259628e-06,
      "loss": 1.69266052,
      "memory(GiB)": 107.26,
      "step": 27890,
      "train_speed(iter/s)": 1.637729
    },
    {
      "acc": 0.65307169,
      "epoch": 0.7076357179096905,
      "grad_norm": 5.625,
      "learning_rate": 7.681676306584159e-06,
      "loss": 1.61092491,
      "memory(GiB)": 107.26,
      "step": 27895,
      "train_speed(iter/s)": 1.637759
    },
    {
      "acc": 0.65158167,
      "epoch": 0.7077625570776256,
      "grad_norm": 6.21875,
      "learning_rate": 7.680791203956594e-06,
      "loss": 1.56169558,
      "memory(GiB)": 107.26,
      "step": 27900,
      "train_speed(iter/s)": 1.637792
    },
    {
      "acc": 0.6487205,
      "epoch": 0.7078893962455606,
      "grad_norm": 6.625,
      "learning_rate": 7.679905983415861e-06,
      "loss": 1.65474091,
      "memory(GiB)": 107.26,
      "step": 27905,
      "train_speed(iter/s)": 1.637823
    },
    {
      "acc": 0.66449122,
      "epoch": 0.7080162354134957,
      "grad_norm": 5.71875,
      "learning_rate": 7.6790206450009e-06,
      "loss": 1.61333084,
      "memory(GiB)": 107.26,
      "step": 27910,
      "train_speed(iter/s)": 1.637855
    },
    {
      "acc": 0.64561214,
      "epoch": 0.7081430745814308,
      "grad_norm": 5.21875,
      "learning_rate": 7.678135188750648e-06,
      "loss": 1.61060162,
      "memory(GiB)": 107.26,
      "step": 27915,
      "train_speed(iter/s)": 1.637886
    },
    {
      "acc": 0.65609264,
      "epoch": 0.7082699137493658,
      "grad_norm": 5.625,
      "learning_rate": 7.677249614704057e-06,
      "loss": 1.59790611,
      "memory(GiB)": 107.26,
      "step": 27920,
      "train_speed(iter/s)": 1.637917
    },
    {
      "acc": 0.66631536,
      "epoch": 0.7083967529173009,
      "grad_norm": 5.34375,
      "learning_rate": 7.676363922900073e-06,
      "loss": 1.57596111,
      "memory(GiB)": 107.26,
      "step": 27925,
      "train_speed(iter/s)": 1.637949
    },
    {
      "acc": 0.63721027,
      "epoch": 0.708523592085236,
      "grad_norm": 6.0,
      "learning_rate": 7.675478113377653e-06,
      "loss": 1.672649,
      "memory(GiB)": 107.26,
      "step": 27930,
      "train_speed(iter/s)": 1.63798
    },
    {
      "acc": 0.65707998,
      "epoch": 0.708650431253171,
      "grad_norm": 6.21875,
      "learning_rate": 7.674592186175762e-06,
      "loss": 1.58273659,
      "memory(GiB)": 107.26,
      "step": 27935,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.66473184,
      "epoch": 0.708777270421106,
      "grad_norm": 4.8125,
      "learning_rate": 7.673706141333365e-06,
      "loss": 1.50540161,
      "memory(GiB)": 107.26,
      "step": 27940,
      "train_speed(iter/s)": 1.638043
    },
    {
      "acc": 0.64969101,
      "epoch": 0.708904109589041,
      "grad_norm": 5.28125,
      "learning_rate": 7.672819978889435e-06,
      "loss": 1.63706322,
      "memory(GiB)": 107.26,
      "step": 27945,
      "train_speed(iter/s)": 1.638074
    },
    {
      "acc": 0.65129633,
      "epoch": 0.7090309487569761,
      "grad_norm": 6.4375,
      "learning_rate": 7.67193369888295e-06,
      "loss": 1.64601135,
      "memory(GiB)": 107.26,
      "step": 27950,
      "train_speed(iter/s)": 1.638106
    },
    {
      "acc": 0.66796207,
      "epoch": 0.7091577879249112,
      "grad_norm": 6.5,
      "learning_rate": 7.67104730135289e-06,
      "loss": 1.52657471,
      "memory(GiB)": 107.26,
      "step": 27955,
      "train_speed(iter/s)": 1.638139
    },
    {
      "acc": 0.66234288,
      "epoch": 0.7092846270928462,
      "grad_norm": 5.625,
      "learning_rate": 7.670160786338246e-06,
      "loss": 1.62004318,
      "memory(GiB)": 107.26,
      "step": 27960,
      "train_speed(iter/s)": 1.638171
    },
    {
      "acc": 0.65799379,
      "epoch": 0.7094114662607813,
      "grad_norm": 5.53125,
      "learning_rate": 7.669274153878006e-06,
      "loss": 1.58321896,
      "memory(GiB)": 107.26,
      "step": 27965,
      "train_speed(iter/s)": 1.638204
    },
    {
      "acc": 0.64791522,
      "epoch": 0.7095383054287164,
      "grad_norm": 5.3125,
      "learning_rate": 7.668387404011176e-06,
      "loss": 1.65013142,
      "memory(GiB)": 107.26,
      "step": 27970,
      "train_speed(iter/s)": 1.638237
    },
    {
      "acc": 0.6491189,
      "epoch": 0.7096651445966514,
      "grad_norm": 5.90625,
      "learning_rate": 7.667500536776748e-06,
      "loss": 1.67357464,
      "memory(GiB)": 107.26,
      "step": 27975,
      "train_speed(iter/s)": 1.638268
    },
    {
      "acc": 0.65121598,
      "epoch": 0.7097919837645865,
      "grad_norm": 4.875,
      "learning_rate": 7.666613552213742e-06,
      "loss": 1.61239452,
      "memory(GiB)": 107.26,
      "step": 27980,
      "train_speed(iter/s)": 1.6383
    },
    {
      "acc": 0.66499119,
      "epoch": 0.7099188229325215,
      "grad_norm": 5.90625,
      "learning_rate": 7.665726450361165e-06,
      "loss": 1.57723751,
      "memory(GiB)": 107.26,
      "step": 27985,
      "train_speed(iter/s)": 1.63833
    },
    {
      "acc": 0.66060958,
      "epoch": 0.7100456621004566,
      "grad_norm": 4.625,
      "learning_rate": 7.664839231258036e-06,
      "loss": 1.60058346,
      "memory(GiB)": 107.26,
      "step": 27990,
      "train_speed(iter/s)": 1.638361
    },
    {
      "acc": 0.66210179,
      "epoch": 0.7101725012683917,
      "grad_norm": 6.09375,
      "learning_rate": 7.663951894943383e-06,
      "loss": 1.6458725,
      "memory(GiB)": 107.26,
      "step": 27995,
      "train_speed(iter/s)": 1.638393
    },
    {
      "acc": 0.64717278,
      "epoch": 0.7102993404363267,
      "grad_norm": 6.5625,
      "learning_rate": 7.66306444145623e-06,
      "loss": 1.6352869,
      "memory(GiB)": 107.26,
      "step": 28000,
      "train_speed(iter/s)": 1.638427
    },
    {
      "epoch": 0.7102993404363267,
      "eval_acc": 0.6452535934527427,
      "eval_loss": 1.5794938802719116,
      "eval_runtime": 58.7015,
      "eval_samples_per_second": 108.515,
      "eval_steps_per_second": 27.137,
      "step": 28000
    },
    {
      "acc": 0.64372878,
      "epoch": 0.7104261796042618,
      "grad_norm": 6.59375,
      "learning_rate": 7.662176870835614e-06,
      "loss": 1.65335445,
      "memory(GiB)": 107.26,
      "step": 28005,
      "train_speed(iter/s)": 1.632447
    },
    {
      "acc": 0.66129799,
      "epoch": 0.7105530187721969,
      "grad_norm": 5.4375,
      "learning_rate": 7.661289183120572e-06,
      "loss": 1.58838711,
      "memory(GiB)": 107.26,
      "step": 28010,
      "train_speed(iter/s)": 1.632478
    },
    {
      "acc": 0.65851049,
      "epoch": 0.7106798579401319,
      "grad_norm": 6.0,
      "learning_rate": 7.66040137835015e-06,
      "loss": 1.63926582,
      "memory(GiB)": 107.26,
      "step": 28015,
      "train_speed(iter/s)": 1.632507
    },
    {
      "acc": 0.62963667,
      "epoch": 0.710806697108067,
      "grad_norm": 5.0,
      "learning_rate": 7.659513456563399e-06,
      "loss": 1.70481815,
      "memory(GiB)": 107.26,
      "step": 28020,
      "train_speed(iter/s)": 1.632538
    },
    {
      "acc": 0.66526031,
      "epoch": 0.710933536276002,
      "grad_norm": 6.25,
      "learning_rate": 7.658625417799372e-06,
      "loss": 1.58290081,
      "memory(GiB)": 107.26,
      "step": 28025,
      "train_speed(iter/s)": 1.632569
    },
    {
      "acc": 0.65998726,
      "epoch": 0.7110603754439371,
      "grad_norm": 5.71875,
      "learning_rate": 7.657737262097128e-06,
      "loss": 1.59762917,
      "memory(GiB)": 107.26,
      "step": 28030,
      "train_speed(iter/s)": 1.6326
    },
    {
      "acc": 0.66310759,
      "epoch": 0.7111872146118722,
      "grad_norm": 5.28125,
      "learning_rate": 7.656848989495733e-06,
      "loss": 1.58448563,
      "memory(GiB)": 107.26,
      "step": 28035,
      "train_speed(iter/s)": 1.632633
    },
    {
      "acc": 0.6720552,
      "epoch": 0.7113140537798072,
      "grad_norm": 5.75,
      "learning_rate": 7.65596060003426e-06,
      "loss": 1.62708893,
      "memory(GiB)": 107.26,
      "step": 28040,
      "train_speed(iter/s)": 1.632663
    },
    {
      "acc": 0.64487104,
      "epoch": 0.7114408929477423,
      "grad_norm": 8.125,
      "learning_rate": 7.655072093751779e-06,
      "loss": 1.7126873,
      "memory(GiB)": 107.26,
      "step": 28045,
      "train_speed(iter/s)": 1.632694
    },
    {
      "acc": 0.65357685,
      "epoch": 0.7115677321156774,
      "grad_norm": 6.5,
      "learning_rate": 7.654183470687375e-06,
      "loss": 1.61941929,
      "memory(GiB)": 107.26,
      "step": 28050,
      "train_speed(iter/s)": 1.632724
    },
    {
      "acc": 0.66873689,
      "epoch": 0.7116945712836124,
      "grad_norm": 6.53125,
      "learning_rate": 7.653294730880131e-06,
      "loss": 1.56383419,
      "memory(GiB)": 107.26,
      "step": 28055,
      "train_speed(iter/s)": 1.632757
    },
    {
      "acc": 0.66259618,
      "epoch": 0.7118214104515475,
      "grad_norm": 6.4375,
      "learning_rate": 7.65240587436914e-06,
      "loss": 1.52600517,
      "memory(GiB)": 107.26,
      "step": 28060,
      "train_speed(iter/s)": 1.63279
    },
    {
      "acc": 0.65123034,
      "epoch": 0.7119482496194824,
      "grad_norm": 5.71875,
      "learning_rate": 7.651516901193494e-06,
      "loss": 1.68130493,
      "memory(GiB)": 107.26,
      "step": 28065,
      "train_speed(iter/s)": 1.632822
    },
    {
      "acc": 0.66621528,
      "epoch": 0.7120750887874175,
      "grad_norm": 5.59375,
      "learning_rate": 7.650627811392298e-06,
      "loss": 1.5240799,
      "memory(GiB)": 107.26,
      "step": 28070,
      "train_speed(iter/s)": 1.632854
    },
    {
      "acc": 0.63887463,
      "epoch": 0.7122019279553526,
      "grad_norm": 5.625,
      "learning_rate": 7.649738605004658e-06,
      "loss": 1.63159866,
      "memory(GiB)": 107.26,
      "step": 28075,
      "train_speed(iter/s)": 1.632885
    },
    {
      "acc": 0.66605206,
      "epoch": 0.7123287671232876,
      "grad_norm": 5.0625,
      "learning_rate": 7.648849282069682e-06,
      "loss": 1.5066618,
      "memory(GiB)": 107.26,
      "step": 28080,
      "train_speed(iter/s)": 1.632915
    },
    {
      "acc": 0.65374641,
      "epoch": 0.7124556062912227,
      "grad_norm": 5.15625,
      "learning_rate": 7.647959842626489e-06,
      "loss": 1.63920403,
      "memory(GiB)": 107.26,
      "step": 28085,
      "train_speed(iter/s)": 1.632948
    },
    {
      "acc": 0.64618883,
      "epoch": 0.7125824454591578,
      "grad_norm": 5.28125,
      "learning_rate": 7.6470702867142e-06,
      "loss": 1.63928795,
      "memory(GiB)": 107.26,
      "step": 28090,
      "train_speed(iter/s)": 1.632978
    },
    {
      "acc": 0.6400157,
      "epoch": 0.7127092846270928,
      "grad_norm": 5.34375,
      "learning_rate": 7.646180614371941e-06,
      "loss": 1.64554939,
      "memory(GiB)": 107.26,
      "step": 28095,
      "train_speed(iter/s)": 1.633009
    },
    {
      "acc": 0.65022769,
      "epoch": 0.7128361237950279,
      "grad_norm": 6.875,
      "learning_rate": 7.645290825638845e-06,
      "loss": 1.62553387,
      "memory(GiB)": 107.26,
      "step": 28100,
      "train_speed(iter/s)": 1.633042
    },
    {
      "acc": 0.65212383,
      "epoch": 0.7129629629629629,
      "grad_norm": 7.25,
      "learning_rate": 7.644400920554048e-06,
      "loss": 1.61919861,
      "memory(GiB)": 107.26,
      "step": 28105,
      "train_speed(iter/s)": 1.633072
    },
    {
      "acc": 0.65708828,
      "epoch": 0.713089802130898,
      "grad_norm": 5.0,
      "learning_rate": 7.64351089915669e-06,
      "loss": 1.65823097,
      "memory(GiB)": 107.26,
      "step": 28110,
      "train_speed(iter/s)": 1.633105
    },
    {
      "acc": 0.68063121,
      "epoch": 0.7132166412988331,
      "grad_norm": 5.71875,
      "learning_rate": 7.642620761485921e-06,
      "loss": 1.51957417,
      "memory(GiB)": 107.26,
      "step": 28115,
      "train_speed(iter/s)": 1.633137
    },
    {
      "acc": 0.65642333,
      "epoch": 0.7133434804667681,
      "grad_norm": 5.3125,
      "learning_rate": 7.641730507580896e-06,
      "loss": 1.59378891,
      "memory(GiB)": 107.26,
      "step": 28120,
      "train_speed(iter/s)": 1.633171
    },
    {
      "acc": 0.64085526,
      "epoch": 0.7134703196347032,
      "grad_norm": 5.25,
      "learning_rate": 7.640840137480763e-06,
      "loss": 1.67423286,
      "memory(GiB)": 107.26,
      "step": 28125,
      "train_speed(iter/s)": 1.633202
    },
    {
      "acc": 0.65447989,
      "epoch": 0.7135971588026383,
      "grad_norm": 5.6875,
      "learning_rate": 7.639949651224697e-06,
      "loss": 1.56445322,
      "memory(GiB)": 107.26,
      "step": 28130,
      "train_speed(iter/s)": 1.633233
    },
    {
      "acc": 0.64223018,
      "epoch": 0.7137239979705733,
      "grad_norm": 5.625,
      "learning_rate": 7.639059048851853e-06,
      "loss": 1.67291393,
      "memory(GiB)": 107.26,
      "step": 28135,
      "train_speed(iter/s)": 1.633258
    },
    {
      "acc": 0.65768952,
      "epoch": 0.7138508371385084,
      "grad_norm": 5.4375,
      "learning_rate": 7.638168330401412e-06,
      "loss": 1.56363239,
      "memory(GiB)": 107.26,
      "step": 28140,
      "train_speed(iter/s)": 1.633288
    },
    {
      "acc": 0.65164928,
      "epoch": 0.7139776763064434,
      "grad_norm": 7.4375,
      "learning_rate": 7.637277495912548e-06,
      "loss": 1.58314877,
      "memory(GiB)": 107.26,
      "step": 28145,
      "train_speed(iter/s)": 1.633318
    },
    {
      "acc": 0.66595659,
      "epoch": 0.7141045154743785,
      "grad_norm": 5.1875,
      "learning_rate": 7.636386545424447e-06,
      "loss": 1.52089405,
      "memory(GiB)": 107.26,
      "step": 28150,
      "train_speed(iter/s)": 1.633349
    },
    {
      "acc": 0.64813285,
      "epoch": 0.7142313546423136,
      "grad_norm": 7.0,
      "learning_rate": 7.635495478976294e-06,
      "loss": 1.61723919,
      "memory(GiB)": 107.26,
      "step": 28155,
      "train_speed(iter/s)": 1.633382
    },
    {
      "acc": 0.64786263,
      "epoch": 0.7143581938102486,
      "grad_norm": 6.53125,
      "learning_rate": 7.634604296607285e-06,
      "loss": 1.70086632,
      "memory(GiB)": 107.26,
      "step": 28160,
      "train_speed(iter/s)": 1.633414
    },
    {
      "acc": 0.66287107,
      "epoch": 0.7144850329781837,
      "grad_norm": 5.625,
      "learning_rate": 7.633712998356612e-06,
      "loss": 1.52047501,
      "memory(GiB)": 107.26,
      "step": 28165,
      "train_speed(iter/s)": 1.633446
    },
    {
      "acc": 0.64548182,
      "epoch": 0.7146118721461188,
      "grad_norm": 6.5625,
      "learning_rate": 7.632821584263486e-06,
      "loss": 1.66940746,
      "memory(GiB)": 107.26,
      "step": 28170,
      "train_speed(iter/s)": 1.633479
    },
    {
      "acc": 0.65837064,
      "epoch": 0.7147387113140538,
      "grad_norm": 5.4375,
      "learning_rate": 7.631930054367112e-06,
      "loss": 1.5959218,
      "memory(GiB)": 107.26,
      "step": 28175,
      "train_speed(iter/s)": 1.633511
    },
    {
      "acc": 0.65959353,
      "epoch": 0.7148655504819889,
      "grad_norm": 5.1875,
      "learning_rate": 7.631038408706703e-06,
      "loss": 1.55693836,
      "memory(GiB)": 107.26,
      "step": 28180,
      "train_speed(iter/s)": 1.633544
    },
    {
      "acc": 0.64711881,
      "epoch": 0.7149923896499238,
      "grad_norm": 6.0625,
      "learning_rate": 7.630146647321476e-06,
      "loss": 1.60683556,
      "memory(GiB)": 107.26,
      "step": 28185,
      "train_speed(iter/s)": 1.633574
    },
    {
      "acc": 0.65769544,
      "epoch": 0.7151192288178589,
      "grad_norm": 5.0625,
      "learning_rate": 7.62925477025066e-06,
      "loss": 1.67220135,
      "memory(GiB)": 107.26,
      "step": 28190,
      "train_speed(iter/s)": 1.633606
    },
    {
      "acc": 0.64771824,
      "epoch": 0.715246067985794,
      "grad_norm": 5.53125,
      "learning_rate": 7.628362777533479e-06,
      "loss": 1.6699955,
      "memory(GiB)": 107.26,
      "step": 28195,
      "train_speed(iter/s)": 1.633638
    },
    {
      "acc": 0.64890823,
      "epoch": 0.715372907153729,
      "grad_norm": 6.0625,
      "learning_rate": 7.627470669209169e-06,
      "loss": 1.60386066,
      "memory(GiB)": 107.26,
      "step": 28200,
      "train_speed(iter/s)": 1.633671
    },
    {
      "acc": 0.66130986,
      "epoch": 0.7154997463216641,
      "grad_norm": 5.25,
      "learning_rate": 7.626578445316968e-06,
      "loss": 1.58195038,
      "memory(GiB)": 107.26,
      "step": 28205,
      "train_speed(iter/s)": 1.633704
    },
    {
      "acc": 0.65298753,
      "epoch": 0.7156265854895992,
      "grad_norm": 7.09375,
      "learning_rate": 7.62568610589612e-06,
      "loss": 1.62716141,
      "memory(GiB)": 107.26,
      "step": 28210,
      "train_speed(iter/s)": 1.633737
    },
    {
      "acc": 0.64660292,
      "epoch": 0.7157534246575342,
      "grad_norm": 7.25,
      "learning_rate": 7.624793650985873e-06,
      "loss": 1.65130424,
      "memory(GiB)": 107.26,
      "step": 28215,
      "train_speed(iter/s)": 1.633772
    },
    {
      "acc": 0.64226384,
      "epoch": 0.7158802638254693,
      "grad_norm": 5.3125,
      "learning_rate": 7.6239010806254835e-06,
      "loss": 1.62994423,
      "memory(GiB)": 107.26,
      "step": 28220,
      "train_speed(iter/s)": 1.633803
    },
    {
      "acc": 0.64463453,
      "epoch": 0.7160071029934043,
      "grad_norm": 5.78125,
      "learning_rate": 7.6230083948542084e-06,
      "loss": 1.65804558,
      "memory(GiB)": 107.26,
      "step": 28225,
      "train_speed(iter/s)": 1.633835
    },
    {
      "acc": 0.66637888,
      "epoch": 0.7161339421613394,
      "grad_norm": 6.125,
      "learning_rate": 7.622115593711314e-06,
      "loss": 1.53976002,
      "memory(GiB)": 107.26,
      "step": 28230,
      "train_speed(iter/s)": 1.633868
    },
    {
      "acc": 0.66092119,
      "epoch": 0.7162607813292745,
      "grad_norm": 5.15625,
      "learning_rate": 7.62122267723607e-06,
      "loss": 1.58150387,
      "memory(GiB)": 107.26,
      "step": 28235,
      "train_speed(iter/s)": 1.633901
    },
    {
      "acc": 0.66293249,
      "epoch": 0.7163876204972095,
      "grad_norm": 4.71875,
      "learning_rate": 7.620329645467748e-06,
      "loss": 1.51172228,
      "memory(GiB)": 107.26,
      "step": 28240,
      "train_speed(iter/s)": 1.633933
    },
    {
      "acc": 0.64189219,
      "epoch": 0.7165144596651446,
      "grad_norm": 5.625,
      "learning_rate": 7.61943649844563e-06,
      "loss": 1.72075176,
      "memory(GiB)": 107.26,
      "step": 28245,
      "train_speed(iter/s)": 1.633967
    },
    {
      "acc": 0.64627934,
      "epoch": 0.7166412988330797,
      "grad_norm": 5.34375,
      "learning_rate": 7.618543236209001e-06,
      "loss": 1.66811256,
      "memory(GiB)": 107.26,
      "step": 28250,
      "train_speed(iter/s)": 1.634
    },
    {
      "acc": 0.64912987,
      "epoch": 0.7167681380010147,
      "grad_norm": 5.3125,
      "learning_rate": 7.617649858797147e-06,
      "loss": 1.6408947,
      "memory(GiB)": 107.26,
      "step": 28255,
      "train_speed(iter/s)": 1.634034
    },
    {
      "acc": 0.65791235,
      "epoch": 0.7168949771689498,
      "grad_norm": 5.4375,
      "learning_rate": 7.616756366249367e-06,
      "loss": 1.6241497,
      "memory(GiB)": 107.26,
      "step": 28260,
      "train_speed(iter/s)": 1.634066
    },
    {
      "acc": 0.64235945,
      "epoch": 0.7170218163368848,
      "grad_norm": 4.71875,
      "learning_rate": 7.6158627586049586e-06,
      "loss": 1.68135319,
      "memory(GiB)": 107.26,
      "step": 28265,
      "train_speed(iter/s)": 1.634097
    },
    {
      "acc": 0.66701756,
      "epoch": 0.7171486555048199,
      "grad_norm": 5.21875,
      "learning_rate": 7.614969035903228e-06,
      "loss": 1.53536587,
      "memory(GiB)": 107.26,
      "step": 28270,
      "train_speed(iter/s)": 1.634129
    },
    {
      "acc": 0.6596734,
      "epoch": 0.717275494672755,
      "grad_norm": 5.6875,
      "learning_rate": 7.614075198183482e-06,
      "loss": 1.55323257,
      "memory(GiB)": 107.26,
      "step": 28275,
      "train_speed(iter/s)": 1.634161
    },
    {
      "acc": 0.65606174,
      "epoch": 0.71740233384069,
      "grad_norm": 8.0625,
      "learning_rate": 7.6131812454850406e-06,
      "loss": 1.6242775,
      "memory(GiB)": 107.26,
      "step": 28280,
      "train_speed(iter/s)": 1.634194
    },
    {
      "acc": 0.65724993,
      "epoch": 0.7175291730086251,
      "grad_norm": 5.96875,
      "learning_rate": 7.612287177847219e-06,
      "loss": 1.62805672,
      "memory(GiB)": 107.26,
      "step": 28285,
      "train_speed(iter/s)": 1.634226
    },
    {
      "acc": 0.6599793,
      "epoch": 0.7176560121765602,
      "grad_norm": 5.125,
      "learning_rate": 7.611392995309345e-06,
      "loss": 1.61015854,
      "memory(GiB)": 107.26,
      "step": 28290,
      "train_speed(iter/s)": 1.634256
    },
    {
      "acc": 0.66009569,
      "epoch": 0.7177828513444952,
      "grad_norm": 6.28125,
      "learning_rate": 7.610498697910748e-06,
      "loss": 1.55302458,
      "memory(GiB)": 107.26,
      "step": 28295,
      "train_speed(iter/s)": 1.634284
    },
    {
      "acc": 0.65794744,
      "epoch": 0.7179096905124303,
      "grad_norm": 5.03125,
      "learning_rate": 7.609604285690762e-06,
      "loss": 1.59754276,
      "memory(GiB)": 107.26,
      "step": 28300,
      "train_speed(iter/s)": 1.634315
    },
    {
      "acc": 0.67010798,
      "epoch": 0.7180365296803652,
      "grad_norm": 5.96875,
      "learning_rate": 7.608709758688731e-06,
      "loss": 1.56151638,
      "memory(GiB)": 107.26,
      "step": 28305,
      "train_speed(iter/s)": 1.634345
    },
    {
      "acc": 0.64606771,
      "epoch": 0.7181633688483003,
      "grad_norm": 5.5,
      "learning_rate": 7.607815116943995e-06,
      "loss": 1.55978374,
      "memory(GiB)": 107.26,
      "step": 28310,
      "train_speed(iter/s)": 1.634379
    },
    {
      "acc": 0.65833216,
      "epoch": 0.7182902080162354,
      "grad_norm": 5.25,
      "learning_rate": 7.606920360495908e-06,
      "loss": 1.54789133,
      "memory(GiB)": 107.26,
      "step": 28315,
      "train_speed(iter/s)": 1.634413
    },
    {
      "acc": 0.63946843,
      "epoch": 0.7184170471841704,
      "grad_norm": 5.1875,
      "learning_rate": 7.6060254893838255e-06,
      "loss": 1.65668297,
      "memory(GiB)": 107.26,
      "step": 28320,
      "train_speed(iter/s)": 1.634444
    },
    {
      "acc": 0.63760405,
      "epoch": 0.7185438863521055,
      "grad_norm": 5.28125,
      "learning_rate": 7.6051305036471065e-06,
      "loss": 1.6331768,
      "memory(GiB)": 107.26,
      "step": 28325,
      "train_speed(iter/s)": 1.634473
    },
    {
      "acc": 0.66723285,
      "epoch": 0.7186707255200406,
      "grad_norm": 5.875,
      "learning_rate": 7.604235403325117e-06,
      "loss": 1.54969244,
      "memory(GiB)": 107.26,
      "step": 28330,
      "train_speed(iter/s)": 1.634502
    },
    {
      "acc": 0.65138001,
      "epoch": 0.7187975646879756,
      "grad_norm": 5.65625,
      "learning_rate": 7.603340188457227e-06,
      "loss": 1.64608841,
      "memory(GiB)": 107.26,
      "step": 28335,
      "train_speed(iter/s)": 1.634534
    },
    {
      "acc": 0.64432344,
      "epoch": 0.7189244038559107,
      "grad_norm": 5.5625,
      "learning_rate": 7.602444859082814e-06,
      "loss": 1.63461361,
      "memory(GiB)": 107.26,
      "step": 28340,
      "train_speed(iter/s)": 1.634567
    },
    {
      "acc": 0.65408378,
      "epoch": 0.7190512430238457,
      "grad_norm": 6.4375,
      "learning_rate": 7.601549415241254e-06,
      "loss": 1.59898453,
      "memory(GiB)": 107.26,
      "step": 28345,
      "train_speed(iter/s)": 1.634601
    },
    {
      "acc": 0.65668678,
      "epoch": 0.7191780821917808,
      "grad_norm": 6.8125,
      "learning_rate": 7.6006538569719375e-06,
      "loss": 1.57138348,
      "memory(GiB)": 107.26,
      "step": 28350,
      "train_speed(iter/s)": 1.63463
    },
    {
      "acc": 0.64599156,
      "epoch": 0.7193049213597159,
      "grad_norm": 5.6875,
      "learning_rate": 7.599758184314252e-06,
      "loss": 1.62103386,
      "memory(GiB)": 107.26,
      "step": 28355,
      "train_speed(iter/s)": 1.634659
    },
    {
      "acc": 0.65033569,
      "epoch": 0.7194317605276509,
      "grad_norm": 6.28125,
      "learning_rate": 7.598862397307596e-06,
      "loss": 1.61566505,
      "memory(GiB)": 107.26,
      "step": 28360,
      "train_speed(iter/s)": 1.63469
    },
    {
      "acc": 0.68914576,
      "epoch": 0.719558599695586,
      "grad_norm": 5.4375,
      "learning_rate": 7.597966495991368e-06,
      "loss": 1.48199768,
      "memory(GiB)": 107.26,
      "step": 28365,
      "train_speed(iter/s)": 1.634723
    },
    {
      "acc": 0.66563497,
      "epoch": 0.7196854388635211,
      "grad_norm": 5.5,
      "learning_rate": 7.597070480404974e-06,
      "loss": 1.49406185,
      "memory(GiB)": 107.26,
      "step": 28370,
      "train_speed(iter/s)": 1.634754
    },
    {
      "acc": 0.65118771,
      "epoch": 0.7198122780314561,
      "grad_norm": 4.9375,
      "learning_rate": 7.596174350587826e-06,
      "loss": 1.6017231,
      "memory(GiB)": 107.26,
      "step": 28375,
      "train_speed(iter/s)": 1.634785
    },
    {
      "acc": 0.64284024,
      "epoch": 0.7199391171993912,
      "grad_norm": 5.21875,
      "learning_rate": 7.595278106579339e-06,
      "loss": 1.62861099,
      "memory(GiB)": 107.26,
      "step": 28380,
      "train_speed(iter/s)": 1.634815
    },
    {
      "acc": 0.66137543,
      "epoch": 0.7200659563673262,
      "grad_norm": 8.3125,
      "learning_rate": 7.594381748418933e-06,
      "loss": 1.56671133,
      "memory(GiB)": 107.26,
      "step": 28385,
      "train_speed(iter/s)": 1.634845
    },
    {
      "acc": 0.65421228,
      "epoch": 0.7201927955352613,
      "grad_norm": 5.0625,
      "learning_rate": 7.593485276146035e-06,
      "loss": 1.57735558,
      "memory(GiB)": 107.26,
      "step": 28390,
      "train_speed(iter/s)": 1.634878
    },
    {
      "acc": 0.67311559,
      "epoch": 0.7203196347031964,
      "grad_norm": 8.4375,
      "learning_rate": 7.592588689800077e-06,
      "loss": 1.57889423,
      "memory(GiB)": 107.26,
      "step": 28395,
      "train_speed(iter/s)": 1.634909
    },
    {
      "acc": 0.64280615,
      "epoch": 0.7204464738711314,
      "grad_norm": 5.5625,
      "learning_rate": 7.591691989420491e-06,
      "loss": 1.66243401,
      "memory(GiB)": 107.26,
      "step": 28400,
      "train_speed(iter/s)": 1.634941
    },
    {
      "acc": 0.66686802,
      "epoch": 0.7205733130390665,
      "grad_norm": 5.8125,
      "learning_rate": 7.590795175046721e-06,
      "loss": 1.56688204,
      "memory(GiB)": 107.26,
      "step": 28405,
      "train_speed(iter/s)": 1.634969
    },
    {
      "acc": 0.65321946,
      "epoch": 0.7207001522070016,
      "grad_norm": 6.125,
      "learning_rate": 7.5898982467182125e-06,
      "loss": 1.58327923,
      "memory(GiB)": 107.26,
      "step": 28410,
      "train_speed(iter/s)": 1.635
    },
    {
      "acc": 0.64529729,
      "epoch": 0.7208269913749366,
      "grad_norm": 7.9375,
      "learning_rate": 7.589001204474416e-06,
      "loss": 1.61412697,
      "memory(GiB)": 107.26,
      "step": 28415,
      "train_speed(iter/s)": 1.635028
    },
    {
      "acc": 0.63404508,
      "epoch": 0.7209538305428717,
      "grad_norm": 5.25,
      "learning_rate": 7.588104048354787e-06,
      "loss": 1.6536087,
      "memory(GiB)": 107.26,
      "step": 28420,
      "train_speed(iter/s)": 1.635061
    },
    {
      "acc": 0.65113173,
      "epoch": 0.7210806697108066,
      "grad_norm": 6.1875,
      "learning_rate": 7.587206778398788e-06,
      "loss": 1.585532,
      "memory(GiB)": 107.26,
      "step": 28425,
      "train_speed(iter/s)": 1.635094
    },
    {
      "acc": 0.66793399,
      "epoch": 0.7212075088787417,
      "grad_norm": 7.71875,
      "learning_rate": 7.586309394645882e-06,
      "loss": 1.53728447,
      "memory(GiB)": 107.26,
      "step": 28430,
      "train_speed(iter/s)": 1.635129
    },
    {
      "acc": 0.66511078,
      "epoch": 0.7213343480466768,
      "grad_norm": 6.125,
      "learning_rate": 7.585411897135544e-06,
      "loss": 1.5370821,
      "memory(GiB)": 107.26,
      "step": 28435,
      "train_speed(iter/s)": 1.63516
    },
    {
      "acc": 0.65755339,
      "epoch": 0.7214611872146118,
      "grad_norm": 6.59375,
      "learning_rate": 7.584514285907245e-06,
      "loss": 1.52565784,
      "memory(GiB)": 107.26,
      "step": 28440,
      "train_speed(iter/s)": 1.635193
    },
    {
      "acc": 0.66906776,
      "epoch": 0.7215880263825469,
      "grad_norm": 5.78125,
      "learning_rate": 7.583616561000471e-06,
      "loss": 1.57781048,
      "memory(GiB)": 107.26,
      "step": 28445,
      "train_speed(iter/s)": 1.635225
    },
    {
      "acc": 0.63835506,
      "epoch": 0.721714865550482,
      "grad_norm": 5.0625,
      "learning_rate": 7.582718722454705e-06,
      "loss": 1.66925316,
      "memory(GiB)": 107.26,
      "step": 28450,
      "train_speed(iter/s)": 1.635258
    },
    {
      "acc": 0.65034003,
      "epoch": 0.721841704718417,
      "grad_norm": 5.5,
      "learning_rate": 7.581820770309438e-06,
      "loss": 1.65013161,
      "memory(GiB)": 107.26,
      "step": 28455,
      "train_speed(iter/s)": 1.635291
    },
    {
      "acc": 0.67539387,
      "epoch": 0.7219685438863521,
      "grad_norm": 5.71875,
      "learning_rate": 7.580922704604168e-06,
      "loss": 1.54581757,
      "memory(GiB)": 107.26,
      "step": 28460,
      "train_speed(iter/s)": 1.635323
    },
    {
      "acc": 0.64145727,
      "epoch": 0.7220953830542871,
      "grad_norm": 5.6875,
      "learning_rate": 7.5800245253783935e-06,
      "loss": 1.61393929,
      "memory(GiB)": 107.26,
      "step": 28465,
      "train_speed(iter/s)": 1.635358
    },
    {
      "acc": 0.63770723,
      "epoch": 0.7222222222222222,
      "grad_norm": 5.65625,
      "learning_rate": 7.579126232671621e-06,
      "loss": 1.62833481,
      "memory(GiB)": 107.26,
      "step": 28470,
      "train_speed(iter/s)": 1.63539
    },
    {
      "acc": 0.66014361,
      "epoch": 0.7223490613901573,
      "grad_norm": 7.21875,
      "learning_rate": 7.578227826523361e-06,
      "loss": 1.65746651,
      "memory(GiB)": 107.26,
      "step": 28475,
      "train_speed(iter/s)": 1.635422
    },
    {
      "acc": 0.67401967,
      "epoch": 0.7224759005580923,
      "grad_norm": 5.15625,
      "learning_rate": 7.577329306973132e-06,
      "loss": 1.55169582,
      "memory(GiB)": 107.26,
      "step": 28480,
      "train_speed(iter/s)": 1.635454
    },
    {
      "acc": 0.66695127,
      "epoch": 0.7226027397260274,
      "grad_norm": 5.71875,
      "learning_rate": 7.576430674060452e-06,
      "loss": 1.53620176,
      "memory(GiB)": 107.26,
      "step": 28485,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.64144616,
      "epoch": 0.7227295788939625,
      "grad_norm": 6.0625,
      "learning_rate": 7.575531927824849e-06,
      "loss": 1.67516003,
      "memory(GiB)": 107.26,
      "step": 28490,
      "train_speed(iter/s)": 1.63552
    },
    {
      "acc": 0.66256618,
      "epoch": 0.7228564180618975,
      "grad_norm": 5.96875,
      "learning_rate": 7.574633068305852e-06,
      "loss": 1.64587593,
      "memory(GiB)": 107.26,
      "step": 28495,
      "train_speed(iter/s)": 1.63555
    },
    {
      "acc": 0.65183434,
      "epoch": 0.7229832572298326,
      "grad_norm": 6.4375,
      "learning_rate": 7.5737340955429995e-06,
      "loss": 1.62679482,
      "memory(GiB)": 107.26,
      "step": 28500,
      "train_speed(iter/s)": 1.635583
    },
    {
      "acc": 0.66222229,
      "epoch": 0.7231100963977676,
      "grad_norm": 5.53125,
      "learning_rate": 7.572835009575828e-06,
      "loss": 1.53670635,
      "memory(GiB)": 107.26,
      "step": 28505,
      "train_speed(iter/s)": 1.635615
    },
    {
      "acc": 0.65891342,
      "epoch": 0.7232369355657027,
      "grad_norm": 6.6875,
      "learning_rate": 7.571935810443886e-06,
      "loss": 1.60377464,
      "memory(GiB)": 107.26,
      "step": 28510,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.6671082,
      "epoch": 0.7233637747336378,
      "grad_norm": 4.8125,
      "learning_rate": 7.571036498186727e-06,
      "loss": 1.55178967,
      "memory(GiB)": 107.26,
      "step": 28515,
      "train_speed(iter/s)": 1.635679
    },
    {
      "acc": 0.65232339,
      "epoch": 0.7234906139015728,
      "grad_norm": 6.375,
      "learning_rate": 7.570137072843902e-06,
      "loss": 1.6531435,
      "memory(GiB)": 107.26,
      "step": 28520,
      "train_speed(iter/s)": 1.635712
    },
    {
      "acc": 0.65039473,
      "epoch": 0.7236174530695079,
      "grad_norm": 5.78125,
      "learning_rate": 7.569237534454974e-06,
      "loss": 1.63271866,
      "memory(GiB)": 107.26,
      "step": 28525,
      "train_speed(iter/s)": 1.635743
    },
    {
      "acc": 0.64541965,
      "epoch": 0.723744292237443,
      "grad_norm": 5.40625,
      "learning_rate": 7.568337883059509e-06,
      "loss": 1.65945778,
      "memory(GiB)": 107.26,
      "step": 28530,
      "train_speed(iter/s)": 1.635774
    },
    {
      "acc": 0.64750261,
      "epoch": 0.723871131405378,
      "grad_norm": 5.96875,
      "learning_rate": 7.567438118697077e-06,
      "loss": 1.63611412,
      "memory(GiB)": 107.26,
      "step": 28535,
      "train_speed(iter/s)": 1.635808
    },
    {
      "acc": 0.6540668,
      "epoch": 0.723997970573313,
      "grad_norm": 6.09375,
      "learning_rate": 7.566538241407253e-06,
      "loss": 1.5664588,
      "memory(GiB)": 107.26,
      "step": 28540,
      "train_speed(iter/s)": 1.635842
    },
    {
      "acc": 0.64244623,
      "epoch": 0.724124809741248,
      "grad_norm": 5.15625,
      "learning_rate": 7.565638251229617e-06,
      "loss": 1.6360487,
      "memory(GiB)": 107.26,
      "step": 28545,
      "train_speed(iter/s)": 1.635873
    },
    {
      "acc": 0.66653395,
      "epoch": 0.7242516489091831,
      "grad_norm": 5.28125,
      "learning_rate": 7.5647381482037585e-06,
      "loss": 1.5413662,
      "memory(GiB)": 107.26,
      "step": 28550,
      "train_speed(iter/s)": 1.635906
    },
    {
      "acc": 0.66212358,
      "epoch": 0.7243784880771182,
      "grad_norm": 4.96875,
      "learning_rate": 7.563837932369264e-06,
      "loss": 1.62485027,
      "memory(GiB)": 107.26,
      "step": 28555,
      "train_speed(iter/s)": 1.635937
    },
    {
      "acc": 0.65409918,
      "epoch": 0.7245053272450532,
      "grad_norm": 6.34375,
      "learning_rate": 7.562937603765732e-06,
      "loss": 1.60686321,
      "memory(GiB)": 107.26,
      "step": 28560,
      "train_speed(iter/s)": 1.635967
    },
    {
      "acc": 0.65579839,
      "epoch": 0.7246321664129883,
      "grad_norm": 5.3125,
      "learning_rate": 7.562037162432761e-06,
      "loss": 1.60123482,
      "memory(GiB)": 107.26,
      "step": 28565,
      "train_speed(iter/s)": 1.635996
    },
    {
      "acc": 0.64640474,
      "epoch": 0.7247590055809234,
      "grad_norm": 5.65625,
      "learning_rate": 7.561136608409956e-06,
      "loss": 1.60928974,
      "memory(GiB)": 107.26,
      "step": 28570,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.63884611,
      "epoch": 0.7248858447488584,
      "grad_norm": 5.375,
      "learning_rate": 7.560235941736929e-06,
      "loss": 1.70201988,
      "memory(GiB)": 107.26,
      "step": 28575,
      "train_speed(iter/s)": 1.636059
    },
    {
      "acc": 0.65022783,
      "epoch": 0.7250126839167935,
      "grad_norm": 5.6875,
      "learning_rate": 7.559335162453294e-06,
      "loss": 1.64949226,
      "memory(GiB)": 107.26,
      "step": 28580,
      "train_speed(iter/s)": 1.63609
    },
    {
      "acc": 0.66325626,
      "epoch": 0.7251395230847285,
      "grad_norm": 7.5625,
      "learning_rate": 7.558434270598672e-06,
      "loss": 1.60072556,
      "memory(GiB)": 107.26,
      "step": 28585,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.65244141,
      "epoch": 0.7252663622526636,
      "grad_norm": 5.625,
      "learning_rate": 7.5575332662126885e-06,
      "loss": 1.60335426,
      "memory(GiB)": 107.26,
      "step": 28590,
      "train_speed(iter/s)": 1.636152
    },
    {
      "acc": 0.6560441,
      "epoch": 0.7253932014205987,
      "grad_norm": 5.5625,
      "learning_rate": 7.556632149334975e-06,
      "loss": 1.5482873,
      "memory(GiB)": 107.26,
      "step": 28595,
      "train_speed(iter/s)": 1.636186
    },
    {
      "acc": 0.65472507,
      "epoch": 0.7255200405885337,
      "grad_norm": 5.375,
      "learning_rate": 7.555730920005163e-06,
      "loss": 1.54491882,
      "memory(GiB)": 107.26,
      "step": 28600,
      "train_speed(iter/s)": 1.636217
    },
    {
      "acc": 0.62411947,
      "epoch": 0.7256468797564688,
      "grad_norm": 6.3125,
      "learning_rate": 7.554829578262894e-06,
      "loss": 1.70723457,
      "memory(GiB)": 107.26,
      "step": 28605,
      "train_speed(iter/s)": 1.636248
    },
    {
      "acc": 0.6647521,
      "epoch": 0.7257737189244039,
      "grad_norm": 7.28125,
      "learning_rate": 7.5539281241478155e-06,
      "loss": 1.63543358,
      "memory(GiB)": 107.26,
      "step": 28610,
      "train_speed(iter/s)": 1.63628
    },
    {
      "acc": 0.66804781,
      "epoch": 0.7259005580923389,
      "grad_norm": 5.4375,
      "learning_rate": 7.5530265576995756e-06,
      "loss": 1.55517902,
      "memory(GiB)": 107.26,
      "step": 28615,
      "train_speed(iter/s)": 1.636311
    },
    {
      "acc": 0.64884882,
      "epoch": 0.726027397260274,
      "grad_norm": 5.125,
      "learning_rate": 7.552124878957829e-06,
      "loss": 1.64047585,
      "memory(GiB)": 107.26,
      "step": 28620,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65027685,
      "epoch": 0.726154236428209,
      "grad_norm": 5.78125,
      "learning_rate": 7.551223087962234e-06,
      "loss": 1.60495262,
      "memory(GiB)": 107.26,
      "step": 28625,
      "train_speed(iter/s)": 1.63637
    },
    {
      "acc": 0.66777925,
      "epoch": 0.7262810755961441,
      "grad_norm": 4.96875,
      "learning_rate": 7.55032118475246e-06,
      "loss": 1.53380642,
      "memory(GiB)": 107.26,
      "step": 28630,
      "train_speed(iter/s)": 1.6364
    },
    {
      "acc": 0.65362878,
      "epoch": 0.7264079147640792,
      "grad_norm": 4.5625,
      "learning_rate": 7.549419169368171e-06,
      "loss": 1.6411747,
      "memory(GiB)": 107.26,
      "step": 28635,
      "train_speed(iter/s)": 1.636431
    },
    {
      "acc": 0.63941097,
      "epoch": 0.7265347539320142,
      "grad_norm": 5.9375,
      "learning_rate": 7.548517041849048e-06,
      "loss": 1.67578945,
      "memory(GiB)": 107.26,
      "step": 28640,
      "train_speed(iter/s)": 1.636464
    },
    {
      "acc": 0.65995092,
      "epoch": 0.7266615930999493,
      "grad_norm": 5.40625,
      "learning_rate": 7.547614802234764e-06,
      "loss": 1.58281345,
      "memory(GiB)": 107.26,
      "step": 28645,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.66975508,
      "epoch": 0.7267884322678844,
      "grad_norm": 4.78125,
      "learning_rate": 7.546712450565008e-06,
      "loss": 1.61544456,
      "memory(GiB)": 107.26,
      "step": 28650,
      "train_speed(iter/s)": 1.636526
    },
    {
      "acc": 0.66985846,
      "epoch": 0.7269152714358194,
      "grad_norm": 4.90625,
      "learning_rate": 7.545809986879469e-06,
      "loss": 1.49771652,
      "memory(GiB)": 107.26,
      "step": 28655,
      "train_speed(iter/s)": 1.636555
    },
    {
      "acc": 0.65896926,
      "epoch": 0.7270421106037545,
      "grad_norm": 6.875,
      "learning_rate": 7.5449074112178385e-06,
      "loss": 1.60677738,
      "memory(GiB)": 107.26,
      "step": 28660,
      "train_speed(iter/s)": 1.636587
    },
    {
      "acc": 0.66010389,
      "epoch": 0.7271689497716894,
      "grad_norm": 7.125,
      "learning_rate": 7.54400472361982e-06,
      "loss": 1.69140244,
      "memory(GiB)": 107.26,
      "step": 28665,
      "train_speed(iter/s)": 1.63662
    },
    {
      "acc": 0.63945427,
      "epoch": 0.7272957889396245,
      "grad_norm": 5.53125,
      "learning_rate": 7.543101924125115e-06,
      "loss": 1.61796341,
      "memory(GiB)": 107.26,
      "step": 28670,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.65030837,
      "epoch": 0.7274226281075596,
      "grad_norm": 6.28125,
      "learning_rate": 7.542199012773432e-06,
      "loss": 1.69965439,
      "memory(GiB)": 107.26,
      "step": 28675,
      "train_speed(iter/s)": 1.636679
    },
    {
      "acc": 0.65872974,
      "epoch": 0.7275494672754946,
      "grad_norm": 6.0,
      "learning_rate": 7.541295989604488e-06,
      "loss": 1.60749512,
      "memory(GiB)": 107.26,
      "step": 28680,
      "train_speed(iter/s)": 1.636711
    },
    {
      "acc": 0.6657393,
      "epoch": 0.7276763064434297,
      "grad_norm": 6.09375,
      "learning_rate": 7.540392854657999e-06,
      "loss": 1.54372215,
      "memory(GiB)": 107.26,
      "step": 28685,
      "train_speed(iter/s)": 1.636741
    },
    {
      "acc": 0.66332645,
      "epoch": 0.7278031456113648,
      "grad_norm": 6.9375,
      "learning_rate": 7.539489607973691e-06,
      "loss": 1.59117022,
      "memory(GiB)": 107.26,
      "step": 28690,
      "train_speed(iter/s)": 1.636775
    },
    {
      "acc": 0.65675788,
      "epoch": 0.7279299847792998,
      "grad_norm": 6.59375,
      "learning_rate": 7.5385862495912905e-06,
      "loss": 1.66924763,
      "memory(GiB)": 107.26,
      "step": 28695,
      "train_speed(iter/s)": 1.636806
    },
    {
      "acc": 0.65276146,
      "epoch": 0.7280568239472349,
      "grad_norm": 6.78125,
      "learning_rate": 7.537682779550537e-06,
      "loss": 1.63707638,
      "memory(GiB)": 107.26,
      "step": 28700,
      "train_speed(iter/s)": 1.636836
    },
    {
      "acc": 0.6351037,
      "epoch": 0.7281836631151699,
      "grad_norm": 6.21875,
      "learning_rate": 7.536779197891159e-06,
      "loss": 1.66475887,
      "memory(GiB)": 107.26,
      "step": 28705,
      "train_speed(iter/s)": 1.636867
    },
    {
      "acc": 0.65997095,
      "epoch": 0.728310502283105,
      "grad_norm": 5.9375,
      "learning_rate": 7.535875504652912e-06,
      "loss": 1.55873117,
      "memory(GiB)": 107.26,
      "step": 28710,
      "train_speed(iter/s)": 1.6369
    },
    {
      "acc": 0.65323343,
      "epoch": 0.7284373414510401,
      "grad_norm": 6.03125,
      "learning_rate": 7.534971699875534e-06,
      "loss": 1.64260521,
      "memory(GiB)": 107.26,
      "step": 28715,
      "train_speed(iter/s)": 1.63693
    },
    {
      "acc": 0.65670843,
      "epoch": 0.7285641806189751,
      "grad_norm": 5.46875,
      "learning_rate": 7.534067783598784e-06,
      "loss": 1.58925095,
      "memory(GiB)": 107.26,
      "step": 28720,
      "train_speed(iter/s)": 1.63696
    },
    {
      "acc": 0.63327646,
      "epoch": 0.7286910197869102,
      "grad_norm": 5.625,
      "learning_rate": 7.533163755862419e-06,
      "loss": 1.65086842,
      "memory(GiB)": 107.26,
      "step": 28725,
      "train_speed(iter/s)": 1.636987
    },
    {
      "acc": 0.65334597,
      "epoch": 0.7288178589548453,
      "grad_norm": 5.25,
      "learning_rate": 7.5322596167062035e-06,
      "loss": 1.60484161,
      "memory(GiB)": 107.26,
      "step": 28730,
      "train_speed(iter/s)": 1.637018
    },
    {
      "acc": 0.6649353,
      "epoch": 0.7289446981227803,
      "grad_norm": 5.84375,
      "learning_rate": 7.5313553661699035e-06,
      "loss": 1.56646852,
      "memory(GiB)": 107.26,
      "step": 28735,
      "train_speed(iter/s)": 1.637049
    },
    {
      "acc": 0.6701273,
      "epoch": 0.7290715372907154,
      "grad_norm": 20.625,
      "learning_rate": 7.530451004293292e-06,
      "loss": 1.5336462,
      "memory(GiB)": 107.26,
      "step": 28740,
      "train_speed(iter/s)": 1.637081
    },
    {
      "acc": 0.64840803,
      "epoch": 0.7291983764586504,
      "grad_norm": 5.78125,
      "learning_rate": 7.5295465311161485e-06,
      "loss": 1.58825636,
      "memory(GiB)": 107.26,
      "step": 28745,
      "train_speed(iter/s)": 1.637111
    },
    {
      "acc": 0.66824765,
      "epoch": 0.7293252156265855,
      "grad_norm": 5.375,
      "learning_rate": 7.5286419466782546e-06,
      "loss": 1.52523804,
      "memory(GiB)": 107.26,
      "step": 28750,
      "train_speed(iter/s)": 1.637144
    },
    {
      "acc": 0.6580842,
      "epoch": 0.7294520547945206,
      "grad_norm": 10.75,
      "learning_rate": 7.527737251019399e-06,
      "loss": 1.61860924,
      "memory(GiB)": 107.26,
      "step": 28755,
      "train_speed(iter/s)": 1.637176
    },
    {
      "acc": 0.63688421,
      "epoch": 0.7295788939624556,
      "grad_norm": 7.0625,
      "learning_rate": 7.526832444179373e-06,
      "loss": 1.65304832,
      "memory(GiB)": 107.26,
      "step": 28760,
      "train_speed(iter/s)": 1.637206
    },
    {
      "acc": 0.65494881,
      "epoch": 0.7297057331303907,
      "grad_norm": 6.5625,
      "learning_rate": 7.525927526197974e-06,
      "loss": 1.61950989,
      "memory(GiB)": 107.26,
      "step": 28765,
      "train_speed(iter/s)": 1.63724
    },
    {
      "acc": 0.64320092,
      "epoch": 0.7298325722983258,
      "grad_norm": 5.03125,
      "learning_rate": 7.5250224971150065e-06,
      "loss": 1.63848495,
      "memory(GiB)": 107.26,
      "step": 28770,
      "train_speed(iter/s)": 1.63727
    },
    {
      "acc": 0.66441946,
      "epoch": 0.7299594114662608,
      "grad_norm": 4.96875,
      "learning_rate": 7.524117356970275e-06,
      "loss": 1.58541851,
      "memory(GiB)": 107.26,
      "step": 28775,
      "train_speed(iter/s)": 1.637301
    },
    {
      "acc": 0.64897156,
      "epoch": 0.7300862506341959,
      "grad_norm": 5.75,
      "learning_rate": 7.523212105803594e-06,
      "loss": 1.697052,
      "memory(GiB)": 107.26,
      "step": 28780,
      "train_speed(iter/s)": 1.637332
    },
    {
      "acc": 0.65576429,
      "epoch": 0.7302130898021308,
      "grad_norm": 6.75,
      "learning_rate": 7.522306743654777e-06,
      "loss": 1.58193026,
      "memory(GiB)": 107.26,
      "step": 28785,
      "train_speed(iter/s)": 1.637365
    },
    {
      "acc": 0.65727882,
      "epoch": 0.7303399289700659,
      "grad_norm": 5.1875,
      "learning_rate": 7.521401270563651e-06,
      "loss": 1.62872696,
      "memory(GiB)": 107.26,
      "step": 28790,
      "train_speed(iter/s)": 1.637396
    },
    {
      "acc": 0.64559159,
      "epoch": 0.730466768138001,
      "grad_norm": 5.46875,
      "learning_rate": 7.520495686570037e-06,
      "loss": 1.6176899,
      "memory(GiB)": 107.26,
      "step": 28795,
      "train_speed(iter/s)": 1.637428
    },
    {
      "acc": 0.6717761,
      "epoch": 0.730593607305936,
      "grad_norm": 6.0,
      "learning_rate": 7.5195899917137716e-06,
      "loss": 1.55495958,
      "memory(GiB)": 107.26,
      "step": 28800,
      "train_speed(iter/s)": 1.637459
    },
    {
      "acc": 0.64469399,
      "epoch": 0.7307204464738711,
      "grad_norm": 6.625,
      "learning_rate": 7.518684186034688e-06,
      "loss": 1.5902565,
      "memory(GiB)": 107.26,
      "step": 28805,
      "train_speed(iter/s)": 1.637489
    },
    {
      "acc": 0.66833234,
      "epoch": 0.7308472856418062,
      "grad_norm": 4.4375,
      "learning_rate": 7.51777826957263e-06,
      "loss": 1.58796854,
      "memory(GiB)": 107.26,
      "step": 28810,
      "train_speed(iter/s)": 1.637521
    },
    {
      "acc": 0.65012321,
      "epoch": 0.7309741248097412,
      "grad_norm": 5.53125,
      "learning_rate": 7.516872242367441e-06,
      "loss": 1.59206676,
      "memory(GiB)": 107.26,
      "step": 28815,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.65565753,
      "epoch": 0.7311009639776763,
      "grad_norm": 5.875,
      "learning_rate": 7.5159661044589745e-06,
      "loss": 1.64078064,
      "memory(GiB)": 107.26,
      "step": 28820,
      "train_speed(iter/s)": 1.637582
    },
    {
      "acc": 0.66105814,
      "epoch": 0.7312278031456113,
      "grad_norm": 7.4375,
      "learning_rate": 7.515059855887087e-06,
      "loss": 1.58720894,
      "memory(GiB)": 107.26,
      "step": 28825,
      "train_speed(iter/s)": 1.637614
    },
    {
      "acc": 0.65079846,
      "epoch": 0.7313546423135464,
      "grad_norm": 6.40625,
      "learning_rate": 7.514153496691636e-06,
      "loss": 1.53527317,
      "memory(GiB)": 107.26,
      "step": 28830,
      "train_speed(iter/s)": 1.637644
    },
    {
      "acc": 0.66522713,
      "epoch": 0.7314814814814815,
      "grad_norm": 5.28125,
      "learning_rate": 7.513247026912491e-06,
      "loss": 1.50655642,
      "memory(GiB)": 107.26,
      "step": 28835,
      "train_speed(iter/s)": 1.637675
    },
    {
      "acc": 0.66022797,
      "epoch": 0.7316083206494165,
      "grad_norm": 6.40625,
      "learning_rate": 7.512340446589521e-06,
      "loss": 1.61134663,
      "memory(GiB)": 107.26,
      "step": 28840,
      "train_speed(iter/s)": 1.637707
    },
    {
      "acc": 0.66343188,
      "epoch": 0.7317351598173516,
      "grad_norm": 4.9375,
      "learning_rate": 7.5114337557625985e-06,
      "loss": 1.56014729,
      "memory(GiB)": 107.26,
      "step": 28845,
      "train_speed(iter/s)": 1.637738
    },
    {
      "acc": 0.67496748,
      "epoch": 0.7318619989852867,
      "grad_norm": 7.0,
      "learning_rate": 7.510526954471611e-06,
      "loss": 1.50279789,
      "memory(GiB)": 107.26,
      "step": 28850,
      "train_speed(iter/s)": 1.637769
    },
    {
      "acc": 0.6495604,
      "epoch": 0.7319888381532217,
      "grad_norm": 10.3125,
      "learning_rate": 7.509620042756436e-06,
      "loss": 1.66697559,
      "memory(GiB)": 107.26,
      "step": 28855,
      "train_speed(iter/s)": 1.637801
    },
    {
      "acc": 0.64409056,
      "epoch": 0.7321156773211568,
      "grad_norm": 5.5,
      "learning_rate": 7.508713020656968e-06,
      "loss": 1.67663002,
      "memory(GiB)": 107.26,
      "step": 28860,
      "train_speed(iter/s)": 1.637834
    },
    {
      "acc": 0.64763794,
      "epoch": 0.7322425164890918,
      "grad_norm": 5.71875,
      "learning_rate": 7.5078058882131e-06,
      "loss": 1.64753304,
      "memory(GiB)": 107.26,
      "step": 28865,
      "train_speed(iter/s)": 1.637865
    },
    {
      "acc": 0.64455328,
      "epoch": 0.7323693556570269,
      "grad_norm": 5.4375,
      "learning_rate": 7.506898645464733e-06,
      "loss": 1.64705353,
      "memory(GiB)": 107.26,
      "step": 28870,
      "train_speed(iter/s)": 1.637898
    },
    {
      "acc": 0.66780729,
      "epoch": 0.732496194824962,
      "grad_norm": 6.03125,
      "learning_rate": 7.505991292451772e-06,
      "loss": 1.53682671,
      "memory(GiB)": 107.26,
      "step": 28875,
      "train_speed(iter/s)": 1.637929
    },
    {
      "acc": 0.64855089,
      "epoch": 0.732623033992897,
      "grad_norm": 4.53125,
      "learning_rate": 7.505083829214125e-06,
      "loss": 1.6743021,
      "memory(GiB)": 107.26,
      "step": 28880,
      "train_speed(iter/s)": 1.63796
    },
    {
      "acc": 0.65300374,
      "epoch": 0.7327498731608321,
      "grad_norm": 4.84375,
      "learning_rate": 7.5041762557917065e-06,
      "loss": 1.67922287,
      "memory(GiB)": 107.26,
      "step": 28885,
      "train_speed(iter/s)": 1.63799
    },
    {
      "acc": 0.64687438,
      "epoch": 0.7328767123287672,
      "grad_norm": 5.0625,
      "learning_rate": 7.5032685722244355e-06,
      "loss": 1.67658386,
      "memory(GiB)": 107.26,
      "step": 28890,
      "train_speed(iter/s)": 1.638022
    },
    {
      "acc": 0.64731216,
      "epoch": 0.7330035514967022,
      "grad_norm": 5.84375,
      "learning_rate": 7.502360778552238e-06,
      "loss": 1.60681381,
      "memory(GiB)": 107.26,
      "step": 28895,
      "train_speed(iter/s)": 1.638054
    },
    {
      "acc": 0.64578061,
      "epoch": 0.7331303906646373,
      "grad_norm": 5.75,
      "learning_rate": 7.5014528748150405e-06,
      "loss": 1.64231052,
      "memory(GiB)": 107.26,
      "step": 28900,
      "train_speed(iter/s)": 1.638085
    },
    {
      "acc": 0.64699678,
      "epoch": 0.7332572298325722,
      "grad_norm": 6.71875,
      "learning_rate": 7.5005448610527765e-06,
      "loss": 1.59507742,
      "memory(GiB)": 107.26,
      "step": 28905,
      "train_speed(iter/s)": 1.638118
    },
    {
      "acc": 0.66919632,
      "epoch": 0.7333840690005073,
      "grad_norm": 5.28125,
      "learning_rate": 7.499636737305386e-06,
      "loss": 1.52741604,
      "memory(GiB)": 107.26,
      "step": 28910,
      "train_speed(iter/s)": 1.638149
    },
    {
      "acc": 0.68215356,
      "epoch": 0.7335109081684424,
      "grad_norm": 6.0,
      "learning_rate": 7.498728503612811e-06,
      "loss": 1.53564415,
      "memory(GiB)": 107.26,
      "step": 28915,
      "train_speed(iter/s)": 1.638181
    },
    {
      "acc": 0.65912724,
      "epoch": 0.7336377473363774,
      "grad_norm": 6.34375,
      "learning_rate": 7.497820160015002e-06,
      "loss": 1.60964584,
      "memory(GiB)": 107.26,
      "step": 28920,
      "train_speed(iter/s)": 1.638213
    },
    {
      "acc": 0.63588533,
      "epoch": 0.7337645865043125,
      "grad_norm": 4.8125,
      "learning_rate": 7.496911706551908e-06,
      "loss": 1.64821682,
      "memory(GiB)": 107.26,
      "step": 28925,
      "train_speed(iter/s)": 1.638244
    },
    {
      "acc": 0.6685626,
      "epoch": 0.7338914256722476,
      "grad_norm": 5.78125,
      "learning_rate": 7.496003143263492e-06,
      "loss": 1.57344646,
      "memory(GiB)": 107.26,
      "step": 28930,
      "train_speed(iter/s)": 1.638275
    },
    {
      "acc": 0.64720635,
      "epoch": 0.7340182648401826,
      "grad_norm": 5.34375,
      "learning_rate": 7.495094470189712e-06,
      "loss": 1.68701668,
      "memory(GiB)": 107.26,
      "step": 28935,
      "train_speed(iter/s)": 1.638304
    },
    {
      "acc": 0.63717484,
      "epoch": 0.7341451040081177,
      "grad_norm": 6.78125,
      "learning_rate": 7.4941856873705376e-06,
      "loss": 1.75105724,
      "memory(GiB)": 107.26,
      "step": 28940,
      "train_speed(iter/s)": 1.638336
    },
    {
      "acc": 0.65002036,
      "epoch": 0.7342719431760527,
      "grad_norm": 5.21875,
      "learning_rate": 7.493276794845941e-06,
      "loss": 1.62092533,
      "memory(GiB)": 107.26,
      "step": 28945,
      "train_speed(iter/s)": 1.638368
    },
    {
      "acc": 0.6571394,
      "epoch": 0.7343987823439878,
      "grad_norm": 5.78125,
      "learning_rate": 7.4923677926559005e-06,
      "loss": 1.54039307,
      "memory(GiB)": 107.26,
      "step": 28950,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.66165857,
      "epoch": 0.7345256215119229,
      "grad_norm": 5.53125,
      "learning_rate": 7.491458680840396e-06,
      "loss": 1.59092617,
      "memory(GiB)": 107.26,
      "step": 28955,
      "train_speed(iter/s)": 1.638433
    },
    {
      "acc": 0.65214124,
      "epoch": 0.7346524606798579,
      "grad_norm": 5.46875,
      "learning_rate": 7.490549459439415e-06,
      "loss": 1.59545469,
      "memory(GiB)": 107.26,
      "step": 28960,
      "train_speed(iter/s)": 1.638463
    },
    {
      "acc": 0.65226622,
      "epoch": 0.734779299847793,
      "grad_norm": 4.59375,
      "learning_rate": 7.48964012849295e-06,
      "loss": 1.61961594,
      "memory(GiB)": 107.26,
      "step": 28965,
      "train_speed(iter/s)": 1.638495
    },
    {
      "acc": 0.64480858,
      "epoch": 0.7349061390157281,
      "grad_norm": 5.78125,
      "learning_rate": 7.488730688040995e-06,
      "loss": 1.61559258,
      "memory(GiB)": 107.26,
      "step": 28970,
      "train_speed(iter/s)": 1.638526
    },
    {
      "acc": 0.64775119,
      "epoch": 0.7350329781836631,
      "grad_norm": 7.96875,
      "learning_rate": 7.487821138123554e-06,
      "loss": 1.6876297,
      "memory(GiB)": 107.26,
      "step": 28975,
      "train_speed(iter/s)": 1.638557
    },
    {
      "acc": 0.68063836,
      "epoch": 0.7351598173515982,
      "grad_norm": 5.59375,
      "learning_rate": 7.486911478780633e-06,
      "loss": 1.51250477,
      "memory(GiB)": 107.26,
      "step": 28980,
      "train_speed(iter/s)": 1.638588
    },
    {
      "acc": 0.66190429,
      "epoch": 0.7352866565195332,
      "grad_norm": 5.5,
      "learning_rate": 7.4860017100522395e-06,
      "loss": 1.58536749,
      "memory(GiB)": 107.26,
      "step": 28985,
      "train_speed(iter/s)": 1.638619
    },
    {
      "acc": 0.64551468,
      "epoch": 0.7354134956874683,
      "grad_norm": 7.0625,
      "learning_rate": 7.485091831978394e-06,
      "loss": 1.66205559,
      "memory(GiB)": 107.26,
      "step": 28990,
      "train_speed(iter/s)": 1.638651
    },
    {
      "acc": 0.66126914,
      "epoch": 0.7355403348554034,
      "grad_norm": 5.40625,
      "learning_rate": 7.484181844599113e-06,
      "loss": 1.63732357,
      "memory(GiB)": 107.26,
      "step": 28995,
      "train_speed(iter/s)": 1.638682
    },
    {
      "acc": 0.66213698,
      "epoch": 0.7356671740233384,
      "grad_norm": 6.34375,
      "learning_rate": 7.483271747954425e-06,
      "loss": 1.556464,
      "memory(GiB)": 107.26,
      "step": 29000,
      "train_speed(iter/s)": 1.638712
    },
    {
      "epoch": 0.7356671740233384,
      "eval_acc": 0.645303712799437,
      "eval_loss": 1.5776091814041138,
      "eval_runtime": 58.7302,
      "eval_samples_per_second": 108.462,
      "eval_steps_per_second": 27.124,
      "step": 29000
    },
    {
      "acc": 0.65302205,
      "epoch": 0.7357940131912735,
      "grad_norm": 6.0625,
      "learning_rate": 7.482361542084356e-06,
      "loss": 1.59565334,
      "memory(GiB)": 107.26,
      "step": 29005,
      "train_speed(iter/s)": 1.632929
    },
    {
      "acc": 0.65489759,
      "epoch": 0.7359208523592086,
      "grad_norm": 4.78125,
      "learning_rate": 7.481451227028946e-06,
      "loss": 1.55446205,
      "memory(GiB)": 107.26,
      "step": 29010,
      "train_speed(iter/s)": 1.63296
    },
    {
      "acc": 0.67156148,
      "epoch": 0.7360476915271436,
      "grad_norm": 6.375,
      "learning_rate": 7.4805408028282316e-06,
      "loss": 1.54848623,
      "memory(GiB)": 107.26,
      "step": 29015,
      "train_speed(iter/s)": 1.632984
    },
    {
      "acc": 0.66563883,
      "epoch": 0.7361745306950787,
      "grad_norm": 4.75,
      "learning_rate": 7.479630269522257e-06,
      "loss": 1.57929354,
      "memory(GiB)": 107.26,
      "step": 29020,
      "train_speed(iter/s)": 1.633013
    },
    {
      "acc": 0.64647994,
      "epoch": 0.7363013698630136,
      "grad_norm": 5.84375,
      "learning_rate": 7.478719627151073e-06,
      "loss": 1.72253551,
      "memory(GiB)": 107.26,
      "step": 29025,
      "train_speed(iter/s)": 1.63304
    },
    {
      "acc": 0.64691334,
      "epoch": 0.7364282090309487,
      "grad_norm": 6.21875,
      "learning_rate": 7.4778088757547325e-06,
      "loss": 1.640555,
      "memory(GiB)": 107.26,
      "step": 29030,
      "train_speed(iter/s)": 1.633066
    },
    {
      "acc": 0.65285029,
      "epoch": 0.7365550481988838,
      "grad_norm": 6.8125,
      "learning_rate": 7.476898015373296e-06,
      "loss": 1.64556732,
      "memory(GiB)": 107.26,
      "step": 29035,
      "train_speed(iter/s)": 1.633094
    },
    {
      "acc": 0.66230888,
      "epoch": 0.7366818873668188,
      "grad_norm": 5.53125,
      "learning_rate": 7.4759870460468256e-06,
      "loss": 1.55623093,
      "memory(GiB)": 107.26,
      "step": 29040,
      "train_speed(iter/s)": 1.633124
    },
    {
      "acc": 0.6492528,
      "epoch": 0.7368087265347539,
      "grad_norm": 6.5,
      "learning_rate": 7.475075967815391e-06,
      "loss": 1.57962933,
      "memory(GiB)": 107.26,
      "step": 29045,
      "train_speed(iter/s)": 1.633156
    },
    {
      "acc": 0.65094757,
      "epoch": 0.736935565702689,
      "grad_norm": 5.34375,
      "learning_rate": 7.474164780719064e-06,
      "loss": 1.57986946,
      "memory(GiB)": 107.26,
      "step": 29050,
      "train_speed(iter/s)": 1.633184
    },
    {
      "acc": 0.6516325,
      "epoch": 0.737062404870624,
      "grad_norm": 7.09375,
      "learning_rate": 7.473253484797924e-06,
      "loss": 1.62027225,
      "memory(GiB)": 107.26,
      "step": 29055,
      "train_speed(iter/s)": 1.632782
    },
    {
      "acc": 0.66782169,
      "epoch": 0.7371892440385591,
      "grad_norm": 7.5,
      "learning_rate": 7.4723420800920545e-06,
      "loss": 1.57625542,
      "memory(GiB)": 107.26,
      "step": 29060,
      "train_speed(iter/s)": 1.632815
    },
    {
      "acc": 0.6600131,
      "epoch": 0.7373160832064941,
      "grad_norm": 5.53125,
      "learning_rate": 7.47143056664154e-06,
      "loss": 1.569203,
      "memory(GiB)": 107.26,
      "step": 29065,
      "train_speed(iter/s)": 1.632844
    },
    {
      "acc": 0.65921316,
      "epoch": 0.7374429223744292,
      "grad_norm": 6.15625,
      "learning_rate": 7.470518944486476e-06,
      "loss": 1.59611015,
      "memory(GiB)": 107.26,
      "step": 29070,
      "train_speed(iter/s)": 1.632874
    },
    {
      "acc": 0.6511754,
      "epoch": 0.7375697615423643,
      "grad_norm": 6.65625,
      "learning_rate": 7.469607213666958e-06,
      "loss": 1.63686409,
      "memory(GiB)": 107.26,
      "step": 29075,
      "train_speed(iter/s)": 1.632903
    },
    {
      "acc": 0.65454955,
      "epoch": 0.7376966007102993,
      "grad_norm": 6.0625,
      "learning_rate": 7.468695374223092e-06,
      "loss": 1.56224289,
      "memory(GiB)": 107.26,
      "step": 29080,
      "train_speed(iter/s)": 1.632931
    },
    {
      "acc": 0.65945535,
      "epoch": 0.7378234398782344,
      "grad_norm": 4.90625,
      "learning_rate": 7.4677834261949765e-06,
      "loss": 1.59688416,
      "memory(GiB)": 107.26,
      "step": 29085,
      "train_speed(iter/s)": 1.632962
    },
    {
      "acc": 0.67590871,
      "epoch": 0.7379502790461695,
      "grad_norm": 5.6875,
      "learning_rate": 7.466871369622731e-06,
      "loss": 1.5191824,
      "memory(GiB)": 107.26,
      "step": 29090,
      "train_speed(iter/s)": 1.632995
    },
    {
      "acc": 0.64535332,
      "epoch": 0.7380771182141045,
      "grad_norm": 6.5,
      "learning_rate": 7.465959204546469e-06,
      "loss": 1.64760017,
      "memory(GiB)": 107.26,
      "step": 29095,
      "train_speed(iter/s)": 1.633024
    },
    {
      "acc": 0.64834518,
      "epoch": 0.7382039573820396,
      "grad_norm": 5.96875,
      "learning_rate": 7.465046931006311e-06,
      "loss": 1.62377567,
      "memory(GiB)": 107.26,
      "step": 29100,
      "train_speed(iter/s)": 1.633054
    },
    {
      "acc": 0.66090288,
      "epoch": 0.7383307965499746,
      "grad_norm": 5.40625,
      "learning_rate": 7.464134549042383e-06,
      "loss": 1.57544804,
      "memory(GiB)": 107.26,
      "step": 29105,
      "train_speed(iter/s)": 1.633085
    },
    {
      "acc": 0.65764174,
      "epoch": 0.7384576357179097,
      "grad_norm": 5.125,
      "learning_rate": 7.463222058694817e-06,
      "loss": 1.634412,
      "memory(GiB)": 107.26,
      "step": 29110,
      "train_speed(iter/s)": 1.633115
    },
    {
      "acc": 0.6506928,
      "epoch": 0.7385844748858448,
      "grad_norm": 4.75,
      "learning_rate": 7.462309460003747e-06,
      "loss": 1.63622303,
      "memory(GiB)": 107.26,
      "step": 29115,
      "train_speed(iter/s)": 1.633147
    },
    {
      "acc": 0.65546932,
      "epoch": 0.7387113140537798,
      "grad_norm": 6.90625,
      "learning_rate": 7.461396753009314e-06,
      "loss": 1.61398411,
      "memory(GiB)": 107.26,
      "step": 29120,
      "train_speed(iter/s)": 1.633179
    },
    {
      "acc": 0.66466231,
      "epoch": 0.7388381532217149,
      "grad_norm": 5.125,
      "learning_rate": 7.460483937751662e-06,
      "loss": 1.56808968,
      "memory(GiB)": 107.26,
      "step": 29125,
      "train_speed(iter/s)": 1.633211
    },
    {
      "acc": 0.64946513,
      "epoch": 0.73896499238965,
      "grad_norm": 5.0,
      "learning_rate": 7.45957101427094e-06,
      "loss": 1.57433662,
      "memory(GiB)": 107.26,
      "step": 29130,
      "train_speed(iter/s)": 1.633241
    },
    {
      "acc": 0.64844632,
      "epoch": 0.739091831557585,
      "grad_norm": 4.90625,
      "learning_rate": 7.458657982607303e-06,
      "loss": 1.61761398,
      "memory(GiB)": 107.26,
      "step": 29135,
      "train_speed(iter/s)": 1.633273
    },
    {
      "acc": 0.64636641,
      "epoch": 0.73921867072552,
      "grad_norm": 7.375,
      "learning_rate": 7.457744842800913e-06,
      "loss": 1.65024185,
      "memory(GiB)": 107.26,
      "step": 29140,
      "train_speed(iter/s)": 1.633302
    },
    {
      "acc": 0.63608418,
      "epoch": 0.739345509893455,
      "grad_norm": 5.625,
      "learning_rate": 7.45683159489193e-06,
      "loss": 1.72340813,
      "memory(GiB)": 107.26,
      "step": 29145,
      "train_speed(iter/s)": 1.633333
    },
    {
      "acc": 0.64535031,
      "epoch": 0.7394723490613901,
      "grad_norm": 6.0,
      "learning_rate": 7.455918238920526e-06,
      "loss": 1.60442944,
      "memory(GiB)": 107.26,
      "step": 29150,
      "train_speed(iter/s)": 1.633363
    },
    {
      "acc": 0.65186863,
      "epoch": 0.7395991882293252,
      "grad_norm": 5.59375,
      "learning_rate": 7.455004774926873e-06,
      "loss": 1.64565983,
      "memory(GiB)": 107.26,
      "step": 29155,
      "train_speed(iter/s)": 1.633391
    },
    {
      "acc": 0.65645924,
      "epoch": 0.7397260273972602,
      "grad_norm": 7.65625,
      "learning_rate": 7.454091202951148e-06,
      "loss": 1.59208612,
      "memory(GiB)": 107.26,
      "step": 29160,
      "train_speed(iter/s)": 1.633419
    },
    {
      "acc": 0.64907517,
      "epoch": 0.7398528665651953,
      "grad_norm": 6.0,
      "learning_rate": 7.453177523033536e-06,
      "loss": 1.61463699,
      "memory(GiB)": 107.26,
      "step": 29165,
      "train_speed(iter/s)": 1.633448
    },
    {
      "acc": 0.64362574,
      "epoch": 0.7399797057331304,
      "grad_norm": 6.5,
      "learning_rate": 7.452263735214223e-06,
      "loss": 1.62020454,
      "memory(GiB)": 107.26,
      "step": 29170,
      "train_speed(iter/s)": 1.633476
    },
    {
      "acc": 0.65544157,
      "epoch": 0.7401065449010654,
      "grad_norm": 6.25,
      "learning_rate": 7.451349839533404e-06,
      "loss": 1.61742306,
      "memory(GiB)": 107.26,
      "step": 29175,
      "train_speed(iter/s)": 1.633508
    },
    {
      "acc": 0.64561186,
      "epoch": 0.7402333840690005,
      "grad_norm": 5.78125,
      "learning_rate": 7.450435836031273e-06,
      "loss": 1.57471981,
      "memory(GiB)": 107.26,
      "step": 29180,
      "train_speed(iter/s)": 1.633539
    },
    {
      "acc": 0.65780139,
      "epoch": 0.7403602232369355,
      "grad_norm": 5.59375,
      "learning_rate": 7.449521724748034e-06,
      "loss": 1.57947617,
      "memory(GiB)": 107.26,
      "step": 29185,
      "train_speed(iter/s)": 1.633568
    },
    {
      "acc": 0.65712581,
      "epoch": 0.7404870624048706,
      "grad_norm": 6.25,
      "learning_rate": 7.4486075057238936e-06,
      "loss": 1.58286419,
      "memory(GiB)": 107.26,
      "step": 29190,
      "train_speed(iter/s)": 1.633598
    },
    {
      "acc": 0.66068292,
      "epoch": 0.7406139015728057,
      "grad_norm": 6.0625,
      "learning_rate": 7.447693178999062e-06,
      "loss": 1.62818146,
      "memory(GiB)": 107.26,
      "step": 29195,
      "train_speed(iter/s)": 1.633628
    },
    {
      "acc": 0.65259008,
      "epoch": 0.7407407407407407,
      "grad_norm": 7.0,
      "learning_rate": 7.446778744613759e-06,
      "loss": 1.63442802,
      "memory(GiB)": 107.26,
      "step": 29200,
      "train_speed(iter/s)": 1.633658
    },
    {
      "acc": 0.67077045,
      "epoch": 0.7408675799086758,
      "grad_norm": 5.53125,
      "learning_rate": 7.445864202608198e-06,
      "loss": 1.53851881,
      "memory(GiB)": 107.26,
      "step": 29205,
      "train_speed(iter/s)": 1.633688
    },
    {
      "acc": 0.6462534,
      "epoch": 0.7409944190766109,
      "grad_norm": 6.28125,
      "learning_rate": 7.444949553022613e-06,
      "loss": 1.61572952,
      "memory(GiB)": 107.26,
      "step": 29210,
      "train_speed(iter/s)": 1.633719
    },
    {
      "acc": 0.65515347,
      "epoch": 0.7411212582445459,
      "grad_norm": 6.125,
      "learning_rate": 7.444034795897229e-06,
      "loss": 1.56886559,
      "memory(GiB)": 107.26,
      "step": 29215,
      "train_speed(iter/s)": 1.63375
    },
    {
      "acc": 0.66997108,
      "epoch": 0.741248097412481,
      "grad_norm": 5.40625,
      "learning_rate": 7.443119931272285e-06,
      "loss": 1.54643269,
      "memory(GiB)": 107.26,
      "step": 29220,
      "train_speed(iter/s)": 1.633778
    },
    {
      "acc": 0.66247816,
      "epoch": 0.741374936580416,
      "grad_norm": 5.8125,
      "learning_rate": 7.442204959188016e-06,
      "loss": 1.5858345,
      "memory(GiB)": 107.26,
      "step": 29225,
      "train_speed(iter/s)": 1.633808
    },
    {
      "acc": 0.64899673,
      "epoch": 0.7415017757483511,
      "grad_norm": 5.53125,
      "learning_rate": 7.4412898796846724e-06,
      "loss": 1.62201366,
      "memory(GiB)": 107.26,
      "step": 29230,
      "train_speed(iter/s)": 1.633839
    },
    {
      "acc": 0.66947298,
      "epoch": 0.7416286149162862,
      "grad_norm": 5.1875,
      "learning_rate": 7.440374692802497e-06,
      "loss": 1.56868095,
      "memory(GiB)": 107.26,
      "step": 29235,
      "train_speed(iter/s)": 1.633869
    },
    {
      "acc": 0.66679707,
      "epoch": 0.7417554540842212,
      "grad_norm": 6.59375,
      "learning_rate": 7.439459398581747e-06,
      "loss": 1.56801071,
      "memory(GiB)": 107.26,
      "step": 29240,
      "train_speed(iter/s)": 1.633901
    },
    {
      "acc": 0.65908208,
      "epoch": 0.7418822932521563,
      "grad_norm": 7.71875,
      "learning_rate": 7.438543997062684e-06,
      "loss": 1.53008347,
      "memory(GiB)": 107.26,
      "step": 29245,
      "train_speed(iter/s)": 1.63393
    },
    {
      "acc": 0.63522816,
      "epoch": 0.7420091324200914,
      "grad_norm": 5.5,
      "learning_rate": 7.437628488285568e-06,
      "loss": 1.59189148,
      "memory(GiB)": 107.26,
      "step": 29250,
      "train_speed(iter/s)": 1.633961
    },
    {
      "acc": 0.65244327,
      "epoch": 0.7421359715880264,
      "grad_norm": 4.875,
      "learning_rate": 7.4367128722906665e-06,
      "loss": 1.57065783,
      "memory(GiB)": 107.26,
      "step": 29255,
      "train_speed(iter/s)": 1.633992
    },
    {
      "acc": 0.65026116,
      "epoch": 0.7422628107559615,
      "grad_norm": 5.78125,
      "learning_rate": 7.435797149118255e-06,
      "loss": 1.65320244,
      "memory(GiB)": 107.26,
      "step": 29260,
      "train_speed(iter/s)": 1.634024
    },
    {
      "acc": 0.65269237,
      "epoch": 0.7423896499238964,
      "grad_norm": 6.25,
      "learning_rate": 7.434881318808609e-06,
      "loss": 1.64191818,
      "memory(GiB)": 107.26,
      "step": 29265,
      "train_speed(iter/s)": 1.634054
    },
    {
      "acc": 0.65367908,
      "epoch": 0.7425164890918315,
      "grad_norm": 5.3125,
      "learning_rate": 7.433965381402013e-06,
      "loss": 1.57841244,
      "memory(GiB)": 107.26,
      "step": 29270,
      "train_speed(iter/s)": 1.634085
    },
    {
      "acc": 0.65718813,
      "epoch": 0.7426433282597666,
      "grad_norm": 6.25,
      "learning_rate": 7.4330493369387514e-06,
      "loss": 1.57849865,
      "memory(GiB)": 107.26,
      "step": 29275,
      "train_speed(iter/s)": 1.634114
    },
    {
      "acc": 0.63749423,
      "epoch": 0.7427701674277016,
      "grad_norm": 5.34375,
      "learning_rate": 7.432133185459117e-06,
      "loss": 1.61236401,
      "memory(GiB)": 107.26,
      "step": 29280,
      "train_speed(iter/s)": 1.634143
    },
    {
      "acc": 0.65103292,
      "epoch": 0.7428970065956367,
      "grad_norm": 5.53125,
      "learning_rate": 7.431216927003406e-06,
      "loss": 1.5567421,
      "memory(GiB)": 107.26,
      "step": 29285,
      "train_speed(iter/s)": 1.634173
    },
    {
      "acc": 0.65703173,
      "epoch": 0.7430238457635718,
      "grad_norm": 6.5625,
      "learning_rate": 7.430300561611922e-06,
      "loss": 1.58064222,
      "memory(GiB)": 107.26,
      "step": 29290,
      "train_speed(iter/s)": 1.634204
    },
    {
      "acc": 0.66567316,
      "epoch": 0.7431506849315068,
      "grad_norm": 5.09375,
      "learning_rate": 7.429384089324967e-06,
      "loss": 1.55421238,
      "memory(GiB)": 107.26,
      "step": 29295,
      "train_speed(iter/s)": 1.634235
    },
    {
      "acc": 0.65256748,
      "epoch": 0.7432775240994419,
      "grad_norm": 5.875,
      "learning_rate": 7.428467510182854e-06,
      "loss": 1.61065369,
      "memory(GiB)": 107.26,
      "step": 29300,
      "train_speed(iter/s)": 1.634265
    },
    {
      "acc": 0.66901436,
      "epoch": 0.7434043632673769,
      "grad_norm": 5.75,
      "learning_rate": 7.427550824225896e-06,
      "loss": 1.52973232,
      "memory(GiB)": 107.26,
      "step": 29305,
      "train_speed(iter/s)": 1.634295
    },
    {
      "acc": 0.65624475,
      "epoch": 0.743531202435312,
      "grad_norm": 6.375,
      "learning_rate": 7.426634031494417e-06,
      "loss": 1.60223827,
      "memory(GiB)": 107.26,
      "step": 29310,
      "train_speed(iter/s)": 1.634325
    },
    {
      "acc": 0.66822181,
      "epoch": 0.7436580416032471,
      "grad_norm": 6.0625,
      "learning_rate": 7.425717132028738e-06,
      "loss": 1.54658632,
      "memory(GiB)": 107.26,
      "step": 29315,
      "train_speed(iter/s)": 1.634356
    },
    {
      "acc": 0.64067836,
      "epoch": 0.7437848807711821,
      "grad_norm": 5.1875,
      "learning_rate": 7.42480012586919e-06,
      "loss": 1.67891979,
      "memory(GiB)": 107.26,
      "step": 29320,
      "train_speed(iter/s)": 1.634387
    },
    {
      "acc": 0.65336094,
      "epoch": 0.7439117199391172,
      "grad_norm": 6.375,
      "learning_rate": 7.423883013056106e-06,
      "loss": 1.53660908,
      "memory(GiB)": 107.26,
      "step": 29325,
      "train_speed(iter/s)": 1.634417
    },
    {
      "acc": 0.65465679,
      "epoch": 0.7440385591070523,
      "grad_norm": 6.125,
      "learning_rate": 7.422965793629825e-06,
      "loss": 1.58391066,
      "memory(GiB)": 107.26,
      "step": 29330,
      "train_speed(iter/s)": 1.634448
    },
    {
      "acc": 0.65347052,
      "epoch": 0.7441653982749873,
      "grad_norm": 6.65625,
      "learning_rate": 7.422048467630691e-06,
      "loss": 1.64281731,
      "memory(GiB)": 107.26,
      "step": 29335,
      "train_speed(iter/s)": 1.634478
    },
    {
      "acc": 0.66338549,
      "epoch": 0.7442922374429224,
      "grad_norm": 5.8125,
      "learning_rate": 7.421131035099052e-06,
      "loss": 1.5602623,
      "memory(GiB)": 107.26,
      "step": 29340,
      "train_speed(iter/s)": 1.634508
    },
    {
      "acc": 0.65835791,
      "epoch": 0.7444190766108574,
      "grad_norm": 5.40625,
      "learning_rate": 7.42021349607526e-06,
      "loss": 1.58160629,
      "memory(GiB)": 107.26,
      "step": 29345,
      "train_speed(iter/s)": 1.634537
    },
    {
      "acc": 0.67601261,
      "epoch": 0.7445459157787925,
      "grad_norm": 4.6875,
      "learning_rate": 7.419295850599673e-06,
      "loss": 1.48615532,
      "memory(GiB)": 107.26,
      "step": 29350,
      "train_speed(iter/s)": 1.634567
    },
    {
      "acc": 0.65748472,
      "epoch": 0.7446727549467276,
      "grad_norm": 6.3125,
      "learning_rate": 7.418378098712653e-06,
      "loss": 1.61027641,
      "memory(GiB)": 107.26,
      "step": 29355,
      "train_speed(iter/s)": 1.634597
    },
    {
      "acc": 0.64197178,
      "epoch": 0.7447995941146626,
      "grad_norm": 5.1875,
      "learning_rate": 7.417460240454568e-06,
      "loss": 1.74327393,
      "memory(GiB)": 107.26,
      "step": 29360,
      "train_speed(iter/s)": 1.634626
    },
    {
      "acc": 0.66285787,
      "epoch": 0.7449264332825977,
      "grad_norm": 5.75,
      "learning_rate": 7.4165422758657865e-06,
      "loss": 1.55088511,
      "memory(GiB)": 107.26,
      "step": 29365,
      "train_speed(iter/s)": 1.634655
    },
    {
      "acc": 0.66351409,
      "epoch": 0.7450532724505328,
      "grad_norm": 5.9375,
      "learning_rate": 7.415624204986689e-06,
      "loss": 1.55484562,
      "memory(GiB)": 107.26,
      "step": 29370,
      "train_speed(iter/s)": 1.634685
    },
    {
      "acc": 0.65254173,
      "epoch": 0.7451801116184678,
      "grad_norm": 5.15625,
      "learning_rate": 7.4147060278576525e-06,
      "loss": 1.61883831,
      "memory(GiB)": 107.26,
      "step": 29375,
      "train_speed(iter/s)": 1.634715
    },
    {
      "acc": 0.65611467,
      "epoch": 0.7453069507864029,
      "grad_norm": 6.09375,
      "learning_rate": 7.413787744519064e-06,
      "loss": 1.57823658,
      "memory(GiB)": 107.26,
      "step": 29380,
      "train_speed(iter/s)": 1.634745
    },
    {
      "acc": 0.64709883,
      "epoch": 0.7454337899543378,
      "grad_norm": 5.46875,
      "learning_rate": 7.412869355011314e-06,
      "loss": 1.59464245,
      "memory(GiB)": 107.26,
      "step": 29385,
      "train_speed(iter/s)": 1.634774
    },
    {
      "acc": 0.64943509,
      "epoch": 0.7455606291222729,
      "grad_norm": 5.71875,
      "learning_rate": 7.411950859374797e-06,
      "loss": 1.66105232,
      "memory(GiB)": 107.26,
      "step": 29390,
      "train_speed(iter/s)": 1.634801
    },
    {
      "acc": 0.66782761,
      "epoch": 0.745687468290208,
      "grad_norm": 6.34375,
      "learning_rate": 7.411032257649913e-06,
      "loss": 1.58989353,
      "memory(GiB)": 107.26,
      "step": 29395,
      "train_speed(iter/s)": 1.634832
    },
    {
      "acc": 0.66166725,
      "epoch": 0.745814307458143,
      "grad_norm": 5.59375,
      "learning_rate": 7.410113549877065e-06,
      "loss": 1.57638054,
      "memory(GiB)": 107.26,
      "step": 29400,
      "train_speed(iter/s)": 1.634863
    },
    {
      "acc": 0.65150476,
      "epoch": 0.7459411466260781,
      "grad_norm": 4.875,
      "learning_rate": 7.409194736096663e-06,
      "loss": 1.62246895,
      "memory(GiB)": 107.26,
      "step": 29405,
      "train_speed(iter/s)": 1.634893
    },
    {
      "acc": 0.66954308,
      "epoch": 0.7460679857940132,
      "grad_norm": 5.84375,
      "learning_rate": 7.408275816349121e-06,
      "loss": 1.58283529,
      "memory(GiB)": 107.26,
      "step": 29410,
      "train_speed(iter/s)": 1.634923
    },
    {
      "acc": 0.65509677,
      "epoch": 0.7461948249619482,
      "grad_norm": 5.15625,
      "learning_rate": 7.4073567906748555e-06,
      "loss": 1.6054512,
      "memory(GiB)": 107.26,
      "step": 29415,
      "train_speed(iter/s)": 1.634954
    },
    {
      "acc": 0.65568814,
      "epoch": 0.7463216641298833,
      "grad_norm": 5.34375,
      "learning_rate": 7.406437659114291e-06,
      "loss": 1.57774544,
      "memory(GiB)": 107.26,
      "step": 29420,
      "train_speed(iter/s)": 1.634985
    },
    {
      "acc": 0.65638022,
      "epoch": 0.7464485032978183,
      "grad_norm": 5.59375,
      "learning_rate": 7.405518421707854e-06,
      "loss": 1.63415298,
      "memory(GiB)": 107.26,
      "step": 29425,
      "train_speed(iter/s)": 1.635015
    },
    {
      "acc": 0.65783281,
      "epoch": 0.7465753424657534,
      "grad_norm": 5.875,
      "learning_rate": 7.404599078495977e-06,
      "loss": 1.57890148,
      "memory(GiB)": 107.26,
      "step": 29430,
      "train_speed(iter/s)": 1.635047
    },
    {
      "acc": 0.64815598,
      "epoch": 0.7467021816336885,
      "grad_norm": 4.90625,
      "learning_rate": 7.403679629519096e-06,
      "loss": 1.62392502,
      "memory(GiB)": 107.26,
      "step": 29435,
      "train_speed(iter/s)": 1.635076
    },
    {
      "acc": 0.64220324,
      "epoch": 0.7468290208016235,
      "grad_norm": 5.34375,
      "learning_rate": 7.402760074817654e-06,
      "loss": 1.72198009,
      "memory(GiB)": 107.26,
      "step": 29440,
      "train_speed(iter/s)": 1.635108
    },
    {
      "acc": 0.68102574,
      "epoch": 0.7469558599695586,
      "grad_norm": 5.65625,
      "learning_rate": 7.4018404144320955e-06,
      "loss": 1.56236458,
      "memory(GiB)": 107.26,
      "step": 29445,
      "train_speed(iter/s)": 1.635138
    },
    {
      "acc": 0.66697998,
      "epoch": 0.7470826991374937,
      "grad_norm": 6.0,
      "learning_rate": 7.4009206484028735e-06,
      "loss": 1.61778507,
      "memory(GiB)": 107.26,
      "step": 29450,
      "train_speed(iter/s)": 1.635168
    },
    {
      "acc": 0.65659909,
      "epoch": 0.7472095383054287,
      "grad_norm": 5.59375,
      "learning_rate": 7.400000776770441e-06,
      "loss": 1.56847897,
      "memory(GiB)": 107.26,
      "step": 29455,
      "train_speed(iter/s)": 1.635198
    },
    {
      "acc": 0.64287806,
      "epoch": 0.7473363774733638,
      "grad_norm": 6.375,
      "learning_rate": 7.39908079957526e-06,
      "loss": 1.62988739,
      "memory(GiB)": 107.26,
      "step": 29460,
      "train_speed(iter/s)": 1.635227
    },
    {
      "acc": 0.65236793,
      "epoch": 0.7474632166412988,
      "grad_norm": 8.4375,
      "learning_rate": 7.398160716857794e-06,
      "loss": 1.61483688,
      "memory(GiB)": 107.26,
      "step": 29465,
      "train_speed(iter/s)": 1.635255
    },
    {
      "acc": 0.66173334,
      "epoch": 0.7475900558092339,
      "grad_norm": 5.40625,
      "learning_rate": 7.397240528658513e-06,
      "loss": 1.56382313,
      "memory(GiB)": 107.26,
      "step": 29470,
      "train_speed(iter/s)": 1.635285
    },
    {
      "acc": 0.66082401,
      "epoch": 0.747716894977169,
      "grad_norm": 5.78125,
      "learning_rate": 7.39632023501789e-06,
      "loss": 1.59990225,
      "memory(GiB)": 107.26,
      "step": 29475,
      "train_speed(iter/s)": 1.635314
    },
    {
      "acc": 0.64387903,
      "epoch": 0.747843734145104,
      "grad_norm": 5.5,
      "learning_rate": 7.3953998359764036e-06,
      "loss": 1.66633511,
      "memory(GiB)": 107.26,
      "step": 29480,
      "train_speed(iter/s)": 1.635343
    },
    {
      "acc": 0.65288548,
      "epoch": 0.7479705733130391,
      "grad_norm": 6.28125,
      "learning_rate": 7.394479331574539e-06,
      "loss": 1.62439461,
      "memory(GiB)": 107.26,
      "step": 29485,
      "train_speed(iter/s)": 1.635372
    },
    {
      "acc": 0.64740705,
      "epoch": 0.7480974124809742,
      "grad_norm": 5.21875,
      "learning_rate": 7.393558721852783e-06,
      "loss": 1.67519226,
      "memory(GiB)": 107.26,
      "step": 29490,
      "train_speed(iter/s)": 1.635402
    },
    {
      "acc": 0.63736835,
      "epoch": 0.7482242516489092,
      "grad_norm": 6.25,
      "learning_rate": 7.392638006851627e-06,
      "loss": 1.64093533,
      "memory(GiB)": 107.26,
      "step": 29495,
      "train_speed(iter/s)": 1.635433
    },
    {
      "acc": 0.66596212,
      "epoch": 0.7483510908168443,
      "grad_norm": 6.59375,
      "learning_rate": 7.391717186611569e-06,
      "loss": 1.57376862,
      "memory(GiB)": 107.26,
      "step": 29500,
      "train_speed(iter/s)": 1.635464
    },
    {
      "acc": 0.66979008,
      "epoch": 0.7484779299847792,
      "grad_norm": 6.46875,
      "learning_rate": 7.39079626117311e-06,
      "loss": 1.55795383,
      "memory(GiB)": 107.26,
      "step": 29505,
      "train_speed(iter/s)": 1.635493
    },
    {
      "acc": 0.64057879,
      "epoch": 0.7486047691527143,
      "grad_norm": 6.34375,
      "learning_rate": 7.3898752305767595e-06,
      "loss": 1.64928398,
      "memory(GiB)": 107.26,
      "step": 29510,
      "train_speed(iter/s)": 1.635523
    },
    {
      "acc": 0.64141364,
      "epoch": 0.7487316083206494,
      "grad_norm": 5.6875,
      "learning_rate": 7.3889540948630245e-06,
      "loss": 1.62794495,
      "memory(GiB)": 107.26,
      "step": 29515,
      "train_speed(iter/s)": 1.635553
    },
    {
      "acc": 0.65392351,
      "epoch": 0.7488584474885844,
      "grad_norm": 7.1875,
      "learning_rate": 7.388032854072424e-06,
      "loss": 1.58461208,
      "memory(GiB)": 107.26,
      "step": 29520,
      "train_speed(iter/s)": 1.635583
    },
    {
      "acc": 0.64792252,
      "epoch": 0.7489852866565195,
      "grad_norm": 6.09375,
      "learning_rate": 7.387111508245476e-06,
      "loss": 1.63074455,
      "memory(GiB)": 107.26,
      "step": 29525,
      "train_speed(iter/s)": 1.635616
    },
    {
      "acc": 0.66881089,
      "epoch": 0.7491121258244546,
      "grad_norm": 5.96875,
      "learning_rate": 7.386190057422706e-06,
      "loss": 1.49496088,
      "memory(GiB)": 107.26,
      "step": 29530,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.64808435,
      "epoch": 0.7492389649923896,
      "grad_norm": 5.1875,
      "learning_rate": 7.385268501644645e-06,
      "loss": 1.57918186,
      "memory(GiB)": 107.26,
      "step": 29535,
      "train_speed(iter/s)": 1.635676
    },
    {
      "acc": 0.65629177,
      "epoch": 0.7493658041603247,
      "grad_norm": 5.28125,
      "learning_rate": 7.384346840951824e-06,
      "loss": 1.61358967,
      "memory(GiB)": 107.26,
      "step": 29540,
      "train_speed(iter/s)": 1.635706
    },
    {
      "acc": 0.66148424,
      "epoch": 0.7494926433282597,
      "grad_norm": 5.875,
      "learning_rate": 7.383425075384785e-06,
      "loss": 1.54641685,
      "memory(GiB)": 107.26,
      "step": 29545,
      "train_speed(iter/s)": 1.635735
    },
    {
      "acc": 0.65984583,
      "epoch": 0.7496194824961948,
      "grad_norm": 5.5625,
      "learning_rate": 7.382503204984069e-06,
      "loss": 1.57003193,
      "memory(GiB)": 107.26,
      "step": 29550,
      "train_speed(iter/s)": 1.635765
    },
    {
      "acc": 0.64785705,
      "epoch": 0.7497463216641299,
      "grad_norm": 5.71875,
      "learning_rate": 7.381581229790226e-06,
      "loss": 1.56552334,
      "memory(GiB)": 107.26,
      "step": 29555,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.66115417,
      "epoch": 0.7498731608320649,
      "grad_norm": 5.75,
      "learning_rate": 7.380659149843806e-06,
      "loss": 1.55136147,
      "memory(GiB)": 107.26,
      "step": 29560,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.65330138,
      "epoch": 0.75,
      "grad_norm": 5.34375,
      "learning_rate": 7.379736965185369e-06,
      "loss": 1.60111237,
      "memory(GiB)": 107.26,
      "step": 29565,
      "train_speed(iter/s)": 1.635857
    },
    {
      "acc": 0.63867292,
      "epoch": 0.7501268391679351,
      "grad_norm": 4.65625,
      "learning_rate": 7.378814675855475e-06,
      "loss": 1.67757683,
      "memory(GiB)": 107.26,
      "step": 29570,
      "train_speed(iter/s)": 1.635885
    },
    {
      "acc": 0.66674709,
      "epoch": 0.7502536783358701,
      "grad_norm": 6.625,
      "learning_rate": 7.37789228189469e-06,
      "loss": 1.58155918,
      "memory(GiB)": 107.26,
      "step": 29575,
      "train_speed(iter/s)": 1.635915
    },
    {
      "acc": 0.65364809,
      "epoch": 0.7503805175038052,
      "grad_norm": 4.75,
      "learning_rate": 7.376969783343588e-06,
      "loss": 1.59277039,
      "memory(GiB)": 107.26,
      "step": 29580,
      "train_speed(iter/s)": 1.635945
    },
    {
      "acc": 0.66334772,
      "epoch": 0.7505073566717403,
      "grad_norm": 5.625,
      "learning_rate": 7.37604718024274e-06,
      "loss": 1.5441288,
      "memory(GiB)": 107.26,
      "step": 29585,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.66174784,
      "epoch": 0.7506341958396753,
      "grad_norm": 6.46875,
      "learning_rate": 7.375124472632732e-06,
      "loss": 1.6192318,
      "memory(GiB)": 107.26,
      "step": 29590,
      "train_speed(iter/s)": 1.636007
    },
    {
      "acc": 0.64309864,
      "epoch": 0.7507610350076104,
      "grad_norm": 5.0625,
      "learning_rate": 7.374201660554142e-06,
      "loss": 1.66559067,
      "memory(GiB)": 107.26,
      "step": 29595,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.65074983,
      "epoch": 0.7508878741755454,
      "grad_norm": 6.5,
      "learning_rate": 7.373278744047565e-06,
      "loss": 1.62208939,
      "memory(GiB)": 107.26,
      "step": 29600,
      "train_speed(iter/s)": 1.63607
    },
    {
      "acc": 0.64657602,
      "epoch": 0.7510147133434805,
      "grad_norm": 7.09375,
      "learning_rate": 7.372355723153593e-06,
      "loss": 1.67632256,
      "memory(GiB)": 107.26,
      "step": 29605,
      "train_speed(iter/s)": 1.6361
    },
    {
      "acc": 0.64342175,
      "epoch": 0.7511415525114156,
      "grad_norm": 6.65625,
      "learning_rate": 7.371432597912824e-06,
      "loss": 1.62645016,
      "memory(GiB)": 107.26,
      "step": 29610,
      "train_speed(iter/s)": 1.63613
    },
    {
      "acc": 0.66082139,
      "epoch": 0.7512683916793506,
      "grad_norm": 5.96875,
      "learning_rate": 7.3705093683658616e-06,
      "loss": 1.62099457,
      "memory(GiB)": 107.26,
      "step": 29615,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.65664968,
      "epoch": 0.7513952308472857,
      "grad_norm": 5.84375,
      "learning_rate": 7.369586034553313e-06,
      "loss": 1.61133804,
      "memory(GiB)": 107.26,
      "step": 29620,
      "train_speed(iter/s)": 1.636195
    },
    {
      "acc": 0.64428134,
      "epoch": 0.7515220700152208,
      "grad_norm": 6.4375,
      "learning_rate": 7.368662596515792e-06,
      "loss": 1.69118843,
      "memory(GiB)": 107.26,
      "step": 29625,
      "train_speed(iter/s)": 1.636225
    },
    {
      "acc": 0.65153189,
      "epoch": 0.7516489091831557,
      "grad_norm": 6.625,
      "learning_rate": 7.367739054293914e-06,
      "loss": 1.64189262,
      "memory(GiB)": 107.26,
      "step": 29630,
      "train_speed(iter/s)": 1.636256
    },
    {
      "acc": 0.6485333,
      "epoch": 0.7517757483510908,
      "grad_norm": 5.375,
      "learning_rate": 7.366815407928302e-06,
      "loss": 1.62105923,
      "memory(GiB)": 107.26,
      "step": 29635,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.65996079,
      "epoch": 0.7519025875190258,
      "grad_norm": 5.3125,
      "learning_rate": 7.365891657459582e-06,
      "loss": 1.59371748,
      "memory(GiB)": 107.26,
      "step": 29640,
      "train_speed(iter/s)": 1.636314
    },
    {
      "acc": 0.64206958,
      "epoch": 0.7520294266869609,
      "grad_norm": 5.6875,
      "learning_rate": 7.3649678029283825e-06,
      "loss": 1.65756855,
      "memory(GiB)": 107.26,
      "step": 29645,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.64436665,
      "epoch": 0.752156265854896,
      "grad_norm": 6.34375,
      "learning_rate": 7.364043844375342e-06,
      "loss": 1.65251083,
      "memory(GiB)": 107.26,
      "step": 29650,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.64730539,
      "epoch": 0.752283105022831,
      "grad_norm": 5.4375,
      "learning_rate": 7.363119781841095e-06,
      "loss": 1.65012016,
      "memory(GiB)": 107.26,
      "step": 29655,
      "train_speed(iter/s)": 1.636403
    },
    {
      "acc": 0.64375935,
      "epoch": 0.7524099441907661,
      "grad_norm": 5.78125,
      "learning_rate": 7.362195615366293e-06,
      "loss": 1.60773811,
      "memory(GiB)": 107.26,
      "step": 29660,
      "train_speed(iter/s)": 1.636432
    },
    {
      "acc": 0.67046642,
      "epoch": 0.7525367833587012,
      "grad_norm": 6.625,
      "learning_rate": 7.361271344991579e-06,
      "loss": 1.5541666,
      "memory(GiB)": 107.26,
      "step": 29665,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.64899292,
      "epoch": 0.7526636225266362,
      "grad_norm": 4.96875,
      "learning_rate": 7.36034697075761e-06,
      "loss": 1.62179127,
      "memory(GiB)": 107.26,
      "step": 29670,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.65920362,
      "epoch": 0.7527904616945713,
      "grad_norm": 6.875,
      "learning_rate": 7.359422492705043e-06,
      "loss": 1.59457026,
      "memory(GiB)": 107.26,
      "step": 29675,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.6616468,
      "epoch": 0.7529173008625063,
      "grad_norm": 7.1875,
      "learning_rate": 7.3584979108745405e-06,
      "loss": 1.56224499,
      "memory(GiB)": 107.26,
      "step": 29680,
      "train_speed(iter/s)": 1.63655
    },
    {
      "acc": 0.66282234,
      "epoch": 0.7530441400304414,
      "grad_norm": 5.625,
      "learning_rate": 7.357573225306771e-06,
      "loss": 1.61290379,
      "memory(GiB)": 107.26,
      "step": 29685,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.65243807,
      "epoch": 0.7531709791983765,
      "grad_norm": 6.59375,
      "learning_rate": 7.356648436042404e-06,
      "loss": 1.63054199,
      "memory(GiB)": 107.26,
      "step": 29690,
      "train_speed(iter/s)": 1.636609
    },
    {
      "acc": 0.64831009,
      "epoch": 0.7532978183663115,
      "grad_norm": 6.46875,
      "learning_rate": 7.355723543122118e-06,
      "loss": 1.65382385,
      "memory(GiB)": 107.26,
      "step": 29695,
      "train_speed(iter/s)": 1.636639
    },
    {
      "acc": 0.67311249,
      "epoch": 0.7534246575342466,
      "grad_norm": 4.90625,
      "learning_rate": 7.354798546586592e-06,
      "loss": 1.5371871,
      "memory(GiB)": 107.26,
      "step": 29700,
      "train_speed(iter/s)": 1.63667
    },
    {
      "acc": 0.66026783,
      "epoch": 0.7535514967021817,
      "grad_norm": 7.03125,
      "learning_rate": 7.353873446476512e-06,
      "loss": 1.57704239,
      "memory(GiB)": 107.26,
      "step": 29705,
      "train_speed(iter/s)": 1.636698
    },
    {
      "acc": 0.65831127,
      "epoch": 0.7536783358701167,
      "grad_norm": 5.9375,
      "learning_rate": 7.3529482428325705e-06,
      "loss": 1.65265522,
      "memory(GiB)": 107.26,
      "step": 29710,
      "train_speed(iter/s)": 1.636728
    },
    {
      "acc": 0.65506353,
      "epoch": 0.7538051750380518,
      "grad_norm": 6.09375,
      "learning_rate": 7.35202293569546e-06,
      "loss": 1.62467575,
      "memory(GiB)": 107.26,
      "step": 29715,
      "train_speed(iter/s)": 1.636759
    },
    {
      "acc": 0.65700579,
      "epoch": 0.7539320142059868,
      "grad_norm": 5.375,
      "learning_rate": 7.351097525105878e-06,
      "loss": 1.59740696,
      "memory(GiB)": 107.26,
      "step": 29720,
      "train_speed(iter/s)": 1.636784
    },
    {
      "acc": 0.65579414,
      "epoch": 0.7540588533739219,
      "grad_norm": 4.90625,
      "learning_rate": 7.35017201110453e-06,
      "loss": 1.55782127,
      "memory(GiB)": 107.26,
      "step": 29725,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.64052753,
      "epoch": 0.754185692541857,
      "grad_norm": 6.5,
      "learning_rate": 7.349246393732126e-06,
      "loss": 1.65632668,
      "memory(GiB)": 107.26,
      "step": 29730,
      "train_speed(iter/s)": 1.636841
    },
    {
      "acc": 0.65518627,
      "epoch": 0.754312531709792,
      "grad_norm": 7.59375,
      "learning_rate": 7.3483206730293755e-06,
      "loss": 1.61497421,
      "memory(GiB)": 107.26,
      "step": 29735,
      "train_speed(iter/s)": 1.63687
    },
    {
      "acc": 0.66390858,
      "epoch": 0.7544393708777271,
      "grad_norm": 5.46875,
      "learning_rate": 7.347394849036998e-06,
      "loss": 1.59423027,
      "memory(GiB)": 107.26,
      "step": 29740,
      "train_speed(iter/s)": 1.6369
    },
    {
      "acc": 0.66273823,
      "epoch": 0.7545662100456622,
      "grad_norm": 5.53125,
      "learning_rate": 7.346468921795714e-06,
      "loss": 1.53641481,
      "memory(GiB)": 107.26,
      "step": 29745,
      "train_speed(iter/s)": 1.636932
    },
    {
      "acc": 0.65079851,
      "epoch": 0.7546930492135971,
      "grad_norm": 5.15625,
      "learning_rate": 7.345542891346251e-06,
      "loss": 1.6226469,
      "memory(GiB)": 107.26,
      "step": 29750,
      "train_speed(iter/s)": 1.636961
    },
    {
      "acc": 0.65375557,
      "epoch": 0.7548198883815322,
      "grad_norm": 5.46875,
      "learning_rate": 7.344616757729341e-06,
      "loss": 1.60310822,
      "memory(GiB)": 107.26,
      "step": 29755,
      "train_speed(iter/s)": 1.636991
    },
    {
      "acc": 0.66145716,
      "epoch": 0.7549467275494672,
      "grad_norm": 5.65625,
      "learning_rate": 7.343690520985716e-06,
      "loss": 1.61966534,
      "memory(GiB)": 107.26,
      "step": 29760,
      "train_speed(iter/s)": 1.637021
    },
    {
      "acc": 0.64625902,
      "epoch": 0.7550735667174023,
      "grad_norm": 5.5,
      "learning_rate": 7.342764181156119e-06,
      "loss": 1.59873657,
      "memory(GiB)": 107.26,
      "step": 29765,
      "train_speed(iter/s)": 1.63705
    },
    {
      "acc": 0.65413418,
      "epoch": 0.7552004058853374,
      "grad_norm": 5.03125,
      "learning_rate": 7.341837738281293e-06,
      "loss": 1.59818268,
      "memory(GiB)": 107.26,
      "step": 29770,
      "train_speed(iter/s)": 1.63708
    },
    {
      "acc": 0.66173258,
      "epoch": 0.7553272450532724,
      "grad_norm": 8.25,
      "learning_rate": 7.3409111924019885e-06,
      "loss": 1.62648888,
      "memory(GiB)": 107.26,
      "step": 29775,
      "train_speed(iter/s)": 1.637109
    },
    {
      "acc": 0.65616388,
      "epoch": 0.7554540842212075,
      "grad_norm": 6.375,
      "learning_rate": 7.3399845435589574e-06,
      "loss": 1.63140087,
      "memory(GiB)": 107.26,
      "step": 29780,
      "train_speed(iter/s)": 1.63714
    },
    {
      "acc": 0.6688632,
      "epoch": 0.7555809233891426,
      "grad_norm": 5.40625,
      "learning_rate": 7.33905779179296e-06,
      "loss": 1.5481226,
      "memory(GiB)": 107.26,
      "step": 29785,
      "train_speed(iter/s)": 1.637172
    },
    {
      "acc": 0.65429974,
      "epoch": 0.7557077625570776,
      "grad_norm": 5.125,
      "learning_rate": 7.338130937144756e-06,
      "loss": 1.62104034,
      "memory(GiB)": 107.26,
      "step": 29790,
      "train_speed(iter/s)": 1.6372
    },
    {
      "acc": 0.67015171,
      "epoch": 0.7558346017250127,
      "grad_norm": 5.71875,
      "learning_rate": 7.3372039796551156e-06,
      "loss": 1.5546936,
      "memory(GiB)": 107.26,
      "step": 29795,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.65639257,
      "epoch": 0.7559614408929477,
      "grad_norm": 5.21875,
      "learning_rate": 7.33627691936481e-06,
      "loss": 1.59825134,
      "memory(GiB)": 107.26,
      "step": 29800,
      "train_speed(iter/s)": 1.637261
    },
    {
      "acc": 0.68607998,
      "epoch": 0.7560882800608828,
      "grad_norm": 4.53125,
      "learning_rate": 7.335349756314614e-06,
      "loss": 1.48566208,
      "memory(GiB)": 107.26,
      "step": 29805,
      "train_speed(iter/s)": 1.637289
    },
    {
      "acc": 0.64377146,
      "epoch": 0.7562151192288179,
      "grad_norm": 5.0625,
      "learning_rate": 7.33442249054531e-06,
      "loss": 1.66704712,
      "memory(GiB)": 107.26,
      "step": 29810,
      "train_speed(iter/s)": 1.63732
    },
    {
      "acc": 0.64667444,
      "epoch": 0.7563419583967529,
      "grad_norm": 5.0625,
      "learning_rate": 7.33349512209768e-06,
      "loss": 1.62487621,
      "memory(GiB)": 107.26,
      "step": 29815,
      "train_speed(iter/s)": 1.637351
    },
    {
      "acc": 0.64919128,
      "epoch": 0.756468797564688,
      "grad_norm": 5.78125,
      "learning_rate": 7.332567651012518e-06,
      "loss": 1.62173386,
      "memory(GiB)": 107.26,
      "step": 29820,
      "train_speed(iter/s)": 1.63738
    },
    {
      "acc": 0.66280093,
      "epoch": 0.7565956367326231,
      "grad_norm": 5.4375,
      "learning_rate": 7.331640077330616e-06,
      "loss": 1.59632034,
      "memory(GiB)": 107.26,
      "step": 29825,
      "train_speed(iter/s)": 1.637411
    },
    {
      "acc": 0.66249547,
      "epoch": 0.7567224759005581,
      "grad_norm": 6.28125,
      "learning_rate": 7.330712401092773e-06,
      "loss": 1.64506664,
      "memory(GiB)": 107.26,
      "step": 29830,
      "train_speed(iter/s)": 1.637441
    },
    {
      "acc": 0.65253234,
      "epoch": 0.7568493150684932,
      "grad_norm": 5.46875,
      "learning_rate": 7.329784622339794e-06,
      "loss": 1.59413738,
      "memory(GiB)": 107.26,
      "step": 29835,
      "train_speed(iter/s)": 1.637472
    },
    {
      "acc": 0.64974833,
      "epoch": 0.7569761542364282,
      "grad_norm": 6.28125,
      "learning_rate": 7.328856741112484e-06,
      "loss": 1.61772957,
      "memory(GiB)": 107.26,
      "step": 29840,
      "train_speed(iter/s)": 1.637502
    },
    {
      "acc": 0.64665971,
      "epoch": 0.7571029934043633,
      "grad_norm": 6.46875,
      "learning_rate": 7.327928757451659e-06,
      "loss": 1.61196213,
      "memory(GiB)": 107.26,
      "step": 29845,
      "train_speed(iter/s)": 1.637533
    },
    {
      "acc": 0.67542439,
      "epoch": 0.7572298325722984,
      "grad_norm": 4.96875,
      "learning_rate": 7.3270006713981325e-06,
      "loss": 1.54231443,
      "memory(GiB)": 107.26,
      "step": 29850,
      "train_speed(iter/s)": 1.637563
    },
    {
      "acc": 0.65324078,
      "epoch": 0.7573566717402334,
      "grad_norm": 5.3125,
      "learning_rate": 7.326072482992728e-06,
      "loss": 1.65557213,
      "memory(GiB)": 107.26,
      "step": 29855,
      "train_speed(iter/s)": 1.637593
    },
    {
      "acc": 0.64249907,
      "epoch": 0.7574835109081685,
      "grad_norm": 6.40625,
      "learning_rate": 7.325144192276269e-06,
      "loss": 1.67062187,
      "memory(GiB)": 107.26,
      "step": 29860,
      "train_speed(iter/s)": 1.637623
    },
    {
      "acc": 0.65190935,
      "epoch": 0.7576103500761036,
      "grad_norm": 6.21875,
      "learning_rate": 7.324215799289588e-06,
      "loss": 1.66738281,
      "memory(GiB)": 107.26,
      "step": 29865,
      "train_speed(iter/s)": 1.637655
    },
    {
      "acc": 0.66650739,
      "epoch": 0.7577371892440385,
      "grad_norm": 7.59375,
      "learning_rate": 7.3232873040735194e-06,
      "loss": 1.56114635,
      "memory(GiB)": 107.26,
      "step": 29870,
      "train_speed(iter/s)": 1.637685
    },
    {
      "acc": 0.65164852,
      "epoch": 0.7578640284119736,
      "grad_norm": 5.46875,
      "learning_rate": 7.322358706668901e-06,
      "loss": 1.64391556,
      "memory(GiB)": 107.26,
      "step": 29875,
      "train_speed(iter/s)": 1.637715
    },
    {
      "acc": 0.65894613,
      "epoch": 0.7579908675799086,
      "grad_norm": 4.9375,
      "learning_rate": 7.321430007116582e-06,
      "loss": 1.60768604,
      "memory(GiB)": 107.26,
      "step": 29880,
      "train_speed(iter/s)": 1.637744
    },
    {
      "acc": 0.63830967,
      "epoch": 0.7581177067478437,
      "grad_norm": 6.46875,
      "learning_rate": 7.320501205457403e-06,
      "loss": 1.67277222,
      "memory(GiB)": 107.26,
      "step": 29885,
      "train_speed(iter/s)": 1.637773
    },
    {
      "acc": 0.65324621,
      "epoch": 0.7582445459157788,
      "grad_norm": 6.65625,
      "learning_rate": 7.319572301732224e-06,
      "loss": 1.59981041,
      "memory(GiB)": 107.26,
      "step": 29890,
      "train_speed(iter/s)": 1.637805
    },
    {
      "acc": 0.64784412,
      "epoch": 0.7583713850837138,
      "grad_norm": 4.59375,
      "learning_rate": 7.3186432959818956e-06,
      "loss": 1.62317047,
      "memory(GiB)": 107.26,
      "step": 29895,
      "train_speed(iter/s)": 1.637368
    },
    {
      "acc": 0.66519856,
      "epoch": 0.7584982242516489,
      "grad_norm": 5.75,
      "learning_rate": 7.317714188247285e-06,
      "loss": 1.56719046,
      "memory(GiB)": 107.26,
      "step": 29900,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.64895034,
      "epoch": 0.758625063419584,
      "grad_norm": 5.53125,
      "learning_rate": 7.316784978569256e-06,
      "loss": 1.61393585,
      "memory(GiB)": 107.26,
      "step": 29905,
      "train_speed(iter/s)": 1.637429
    },
    {
      "acc": 0.6667388,
      "epoch": 0.758751902587519,
      "grad_norm": 5.5,
      "learning_rate": 7.31585566698868e-06,
      "loss": 1.49736147,
      "memory(GiB)": 107.26,
      "step": 29910,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.66874523,
      "epoch": 0.7588787417554541,
      "grad_norm": 6.15625,
      "learning_rate": 7.314926253546433e-06,
      "loss": 1.50971413,
      "memory(GiB)": 107.26,
      "step": 29915,
      "train_speed(iter/s)": 1.637488
    },
    {
      "acc": 0.65503187,
      "epoch": 0.7590055809233891,
      "grad_norm": 5.40625,
      "learning_rate": 7.313996738283393e-06,
      "loss": 1.61468105,
      "memory(GiB)": 107.26,
      "step": 29920,
      "train_speed(iter/s)": 1.637518
    },
    {
      "acc": 0.65964861,
      "epoch": 0.7591324200913242,
      "grad_norm": 6.5,
      "learning_rate": 7.3130671212404455e-06,
      "loss": 1.63434219,
      "memory(GiB)": 107.26,
      "step": 29925,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.66120968,
      "epoch": 0.7592592592592593,
      "grad_norm": 5.0,
      "learning_rate": 7.312137402458479e-06,
      "loss": 1.59386234,
      "memory(GiB)": 107.26,
      "step": 29930,
      "train_speed(iter/s)": 1.637581
    },
    {
      "acc": 0.65374956,
      "epoch": 0.7593860984271943,
      "grad_norm": 5.03125,
      "learning_rate": 7.3112075819783864e-06,
      "loss": 1.61699638,
      "memory(GiB)": 107.26,
      "step": 29935,
      "train_speed(iter/s)": 1.637611
    },
    {
      "acc": 0.63833857,
      "epoch": 0.7595129375951294,
      "grad_norm": 7.96875,
      "learning_rate": 7.310277659841066e-06,
      "loss": 1.71940956,
      "memory(GiB)": 107.26,
      "step": 29940,
      "train_speed(iter/s)": 1.63764
    },
    {
      "acc": 0.63947573,
      "epoch": 0.7596397767630645,
      "grad_norm": 5.6875,
      "learning_rate": 7.309347636087418e-06,
      "loss": 1.63878136,
      "memory(GiB)": 107.26,
      "step": 29945,
      "train_speed(iter/s)": 1.637669
    },
    {
      "acc": 0.63901157,
      "epoch": 0.7597666159309995,
      "grad_norm": 4.59375,
      "learning_rate": 7.308417510758353e-06,
      "loss": 1.66148605,
      "memory(GiB)": 107.26,
      "step": 29950,
      "train_speed(iter/s)": 1.637699
    },
    {
      "acc": 0.63761992,
      "epoch": 0.7598934550989346,
      "grad_norm": 5.21875,
      "learning_rate": 7.307487283894777e-06,
      "loss": 1.69525356,
      "memory(GiB)": 107.26,
      "step": 29955,
      "train_speed(iter/s)": 1.637728
    },
    {
      "acc": 0.67261648,
      "epoch": 0.7600202942668696,
      "grad_norm": 6.0625,
      "learning_rate": 7.30655695553761e-06,
      "loss": 1.51520844,
      "memory(GiB)": 107.26,
      "step": 29960,
      "train_speed(iter/s)": 1.63776
    },
    {
      "acc": 0.66503682,
      "epoch": 0.7601471334348047,
      "grad_norm": 6.375,
      "learning_rate": 7.305626525727769e-06,
      "loss": 1.58030739,
      "memory(GiB)": 107.26,
      "step": 29965,
      "train_speed(iter/s)": 1.637791
    },
    {
      "acc": 0.65106478,
      "epoch": 0.7602739726027398,
      "grad_norm": 4.9375,
      "learning_rate": 7.30469599450618e-06,
      "loss": 1.70409756,
      "memory(GiB)": 107.26,
      "step": 29970,
      "train_speed(iter/s)": 1.637822
    },
    {
      "acc": 0.65443373,
      "epoch": 0.7604008117706748,
      "grad_norm": 4.5,
      "learning_rate": 7.30376536191377e-06,
      "loss": 1.58743601,
      "memory(GiB)": 107.26,
      "step": 29975,
      "train_speed(iter/s)": 1.63785
    },
    {
      "acc": 0.65636206,
      "epoch": 0.7605276509386099,
      "grad_norm": 4.71875,
      "learning_rate": 7.302834627991477e-06,
      "loss": 1.61941223,
      "memory(GiB)": 107.26,
      "step": 29980,
      "train_speed(iter/s)": 1.637877
    },
    {
      "acc": 0.65274434,
      "epoch": 0.760654490106545,
      "grad_norm": 5.125,
      "learning_rate": 7.301903792780233e-06,
      "loss": 1.6437767,
      "memory(GiB)": 107.26,
      "step": 29985,
      "train_speed(iter/s)": 1.637908
    },
    {
      "acc": 0.66308994,
      "epoch": 0.76078132927448,
      "grad_norm": 5.125,
      "learning_rate": 7.300972856320984e-06,
      "loss": 1.54835691,
      "memory(GiB)": 107.26,
      "step": 29990,
      "train_speed(iter/s)": 1.637938
    },
    {
      "acc": 0.64939518,
      "epoch": 0.760908168442415,
      "grad_norm": 6.4375,
      "learning_rate": 7.3000418186546754e-06,
      "loss": 1.61641197,
      "memory(GiB)": 107.26,
      "step": 29995,
      "train_speed(iter/s)": 1.637969
    },
    {
      "acc": 0.67032819,
      "epoch": 0.76103500761035,
      "grad_norm": 4.875,
      "learning_rate": 7.299110679822258e-06,
      "loss": 1.51219215,
      "memory(GiB)": 107.26,
      "step": 30000,
      "train_speed(iter/s)": 1.637998
    },
    {
      "epoch": 0.76103500761035,
      "eval_acc": 0.6454574121292995,
      "eval_loss": 1.5773452520370483,
      "eval_runtime": 58.4813,
      "eval_samples_per_second": 108.924,
      "eval_steps_per_second": 27.239,
      "step": 30000
    },
    {
      "acc": 0.64851847,
      "epoch": 0.7611618467782851,
      "grad_norm": 5.59375,
      "learning_rate": 7.298179439864689e-06,
      "loss": 1.64854012,
      "memory(GiB)": 107.26,
      "step": 30005,
      "train_speed(iter/s)": 1.632436
    },
    {
      "acc": 0.65962582,
      "epoch": 0.7612886859462202,
      "grad_norm": 5.15625,
      "learning_rate": 7.297248098822926e-06,
      "loss": 1.59453669,
      "memory(GiB)": 107.26,
      "step": 30010,
      "train_speed(iter/s)": 1.632465
    },
    {
      "acc": 0.65277605,
      "epoch": 0.7614155251141552,
      "grad_norm": 4.78125,
      "learning_rate": 7.296316656737936e-06,
      "loss": 1.59894285,
      "memory(GiB)": 107.26,
      "step": 30015,
      "train_speed(iter/s)": 1.632497
    },
    {
      "acc": 0.67766733,
      "epoch": 0.7615423642820903,
      "grad_norm": 5.25,
      "learning_rate": 7.295385113650689e-06,
      "loss": 1.56710539,
      "memory(GiB)": 107.26,
      "step": 30020,
      "train_speed(iter/s)": 1.632526
    },
    {
      "acc": 0.65055079,
      "epoch": 0.7616692034500254,
      "grad_norm": 5.65625,
      "learning_rate": 7.294453469602154e-06,
      "loss": 1.5896925,
      "memory(GiB)": 107.26,
      "step": 30025,
      "train_speed(iter/s)": 1.632557
    },
    {
      "acc": 0.67665253,
      "epoch": 0.7617960426179604,
      "grad_norm": 4.96875,
      "learning_rate": 7.293521724633313e-06,
      "loss": 1.54910517,
      "memory(GiB)": 107.26,
      "step": 30030,
      "train_speed(iter/s)": 1.632589
    },
    {
      "acc": 0.6501286,
      "epoch": 0.7619228817858955,
      "grad_norm": 5.34375,
      "learning_rate": 7.2925898787851455e-06,
      "loss": 1.60441208,
      "memory(GiB)": 107.26,
      "step": 30035,
      "train_speed(iter/s)": 1.632618
    },
    {
      "acc": 0.65851078,
      "epoch": 0.7620497209538305,
      "grad_norm": 5.4375,
      "learning_rate": 7.2916579320986415e-06,
      "loss": 1.57115822,
      "memory(GiB)": 107.26,
      "step": 30040,
      "train_speed(iter/s)": 1.632648
    },
    {
      "acc": 0.66216612,
      "epoch": 0.7621765601217656,
      "grad_norm": 5.9375,
      "learning_rate": 7.290725884614787e-06,
      "loss": 1.62082214,
      "memory(GiB)": 107.26,
      "step": 30045,
      "train_speed(iter/s)": 1.632679
    },
    {
      "acc": 0.65391493,
      "epoch": 0.7623033992897007,
      "grad_norm": 5.0,
      "learning_rate": 7.2897937363745844e-06,
      "loss": 1.59441214,
      "memory(GiB)": 107.26,
      "step": 30050,
      "train_speed(iter/s)": 1.632709
    },
    {
      "acc": 0.64813786,
      "epoch": 0.7624302384576357,
      "grad_norm": 6.53125,
      "learning_rate": 7.2888614874190276e-06,
      "loss": 1.64497299,
      "memory(GiB)": 107.26,
      "step": 30055,
      "train_speed(iter/s)": 1.632739
    },
    {
      "acc": 0.65853043,
      "epoch": 0.7625570776255708,
      "grad_norm": 5.96875,
      "learning_rate": 7.287929137789124e-06,
      "loss": 1.60991402,
      "memory(GiB)": 107.26,
      "step": 30060,
      "train_speed(iter/s)": 1.632767
    },
    {
      "acc": 0.66545115,
      "epoch": 0.7626839167935059,
      "grad_norm": 6.1875,
      "learning_rate": 7.286996687525882e-06,
      "loss": 1.59634819,
      "memory(GiB)": 107.26,
      "step": 30065,
      "train_speed(iter/s)": 1.632797
    },
    {
      "acc": 0.64672623,
      "epoch": 0.7628107559614409,
      "grad_norm": 5.0,
      "learning_rate": 7.2860641366703155e-06,
      "loss": 1.61122017,
      "memory(GiB)": 107.26,
      "step": 30070,
      "train_speed(iter/s)": 1.632828
    },
    {
      "acc": 0.64658146,
      "epoch": 0.762937595129376,
      "grad_norm": 5.53125,
      "learning_rate": 7.285131485263441e-06,
      "loss": 1.63771935,
      "memory(GiB)": 107.26,
      "step": 30075,
      "train_speed(iter/s)": 1.632857
    },
    {
      "acc": 0.65267591,
      "epoch": 0.763064434297311,
      "grad_norm": 5.3125,
      "learning_rate": 7.2841987333462815e-06,
      "loss": 1.68563614,
      "memory(GiB)": 107.26,
      "step": 30080,
      "train_speed(iter/s)": 1.632887
    },
    {
      "acc": 0.65449533,
      "epoch": 0.7631912734652461,
      "grad_norm": 6.0625,
      "learning_rate": 7.283265880959863e-06,
      "loss": 1.59625053,
      "memory(GiB)": 107.26,
      "step": 30085,
      "train_speed(iter/s)": 1.632915
    },
    {
      "acc": 0.65416627,
      "epoch": 0.7633181126331812,
      "grad_norm": 5.03125,
      "learning_rate": 7.282332928145219e-06,
      "loss": 1.60183277,
      "memory(GiB)": 107.26,
      "step": 30090,
      "train_speed(iter/s)": 1.632944
    },
    {
      "acc": 0.65568857,
      "epoch": 0.7634449518011162,
      "grad_norm": 5.4375,
      "learning_rate": 7.281399874943381e-06,
      "loss": 1.55371704,
      "memory(GiB)": 107.26,
      "step": 30095,
      "train_speed(iter/s)": 1.632973
    },
    {
      "acc": 0.65219216,
      "epoch": 0.7635717909690513,
      "grad_norm": 6.03125,
      "learning_rate": 7.280466721395393e-06,
      "loss": 1.67741356,
      "memory(GiB)": 107.26,
      "step": 30100,
      "train_speed(iter/s)": 1.633002
    },
    {
      "acc": 0.64180088,
      "epoch": 0.7636986301369864,
      "grad_norm": 5.1875,
      "learning_rate": 7.279533467542295e-06,
      "loss": 1.63484802,
      "memory(GiB)": 107.26,
      "step": 30105,
      "train_speed(iter/s)": 1.633032
    },
    {
      "acc": 0.64799442,
      "epoch": 0.7638254693049213,
      "grad_norm": 5.78125,
      "learning_rate": 7.2786001134251385e-06,
      "loss": 1.65406303,
      "memory(GiB)": 107.26,
      "step": 30110,
      "train_speed(iter/s)": 1.633065
    },
    {
      "acc": 0.6596735,
      "epoch": 0.7639523084728564,
      "grad_norm": 6.25,
      "learning_rate": 7.2776666590849744e-06,
      "loss": 1.58586216,
      "memory(GiB)": 107.26,
      "step": 30115,
      "train_speed(iter/s)": 1.633096
    },
    {
      "acc": 0.66249547,
      "epoch": 0.7640791476407914,
      "grad_norm": 5.0,
      "learning_rate": 7.276733104562863e-06,
      "loss": 1.5884119,
      "memory(GiB)": 107.26,
      "step": 30120,
      "train_speed(iter/s)": 1.633124
    },
    {
      "acc": 0.65575652,
      "epoch": 0.7642059868087265,
      "grad_norm": 4.9375,
      "learning_rate": 7.275799449899865e-06,
      "loss": 1.62027416,
      "memory(GiB)": 107.26,
      "step": 30125,
      "train_speed(iter/s)": 1.633154
    },
    {
      "acc": 0.65884333,
      "epoch": 0.7643328259766616,
      "grad_norm": 5.8125,
      "learning_rate": 7.274865695137046e-06,
      "loss": 1.63180504,
      "memory(GiB)": 107.26,
      "step": 30130,
      "train_speed(iter/s)": 1.633184
    },
    {
      "acc": 0.65675788,
      "epoch": 0.7644596651445966,
      "grad_norm": 5.71875,
      "learning_rate": 7.273931840315477e-06,
      "loss": 1.59313021,
      "memory(GiB)": 107.26,
      "step": 30135,
      "train_speed(iter/s)": 1.633214
    },
    {
      "acc": 0.66940193,
      "epoch": 0.7645865043125317,
      "grad_norm": 4.96875,
      "learning_rate": 7.272997885476234e-06,
      "loss": 1.57314262,
      "memory(GiB)": 107.26,
      "step": 30140,
      "train_speed(iter/s)": 1.633244
    },
    {
      "acc": 0.64023342,
      "epoch": 0.7647133434804668,
      "grad_norm": 5.09375,
      "learning_rate": 7.272063830660395e-06,
      "loss": 1.68861885,
      "memory(GiB)": 107.26,
      "step": 30145,
      "train_speed(iter/s)": 1.633273
    },
    {
      "acc": 0.65716252,
      "epoch": 0.7648401826484018,
      "grad_norm": 5.28125,
      "learning_rate": 7.271129675909046e-06,
      "loss": 1.59184484,
      "memory(GiB)": 107.26,
      "step": 30150,
      "train_speed(iter/s)": 1.633304
    },
    {
      "acc": 0.66924739,
      "epoch": 0.7649670218163369,
      "grad_norm": 6.21875,
      "learning_rate": 7.270195421263271e-06,
      "loss": 1.53166142,
      "memory(GiB)": 107.26,
      "step": 30155,
      "train_speed(iter/s)": 1.633328
    },
    {
      "acc": 0.65546036,
      "epoch": 0.7650938609842719,
      "grad_norm": 6.1875,
      "learning_rate": 7.269261066764169e-06,
      "loss": 1.61620693,
      "memory(GiB)": 107.26,
      "step": 30160,
      "train_speed(iter/s)": 1.633357
    },
    {
      "acc": 0.63966408,
      "epoch": 0.765220700152207,
      "grad_norm": 6.875,
      "learning_rate": 7.268326612452832e-06,
      "loss": 1.6111599,
      "memory(GiB)": 107.26,
      "step": 30165,
      "train_speed(iter/s)": 1.633392
    },
    {
      "acc": 0.66058455,
      "epoch": 0.7653475393201421,
      "grad_norm": 7.09375,
      "learning_rate": 7.267392058370364e-06,
      "loss": 1.57169638,
      "memory(GiB)": 107.26,
      "step": 30170,
      "train_speed(iter/s)": 1.633422
    },
    {
      "acc": 0.65681047,
      "epoch": 0.7654743784880771,
      "grad_norm": 5.59375,
      "learning_rate": 7.2664574045578685e-06,
      "loss": 1.61545181,
      "memory(GiB)": 107.26,
      "step": 30175,
      "train_speed(iter/s)": 1.633452
    },
    {
      "acc": 0.65944204,
      "epoch": 0.7656012176560122,
      "grad_norm": 5.28125,
      "learning_rate": 7.26552265105646e-06,
      "loss": 1.62280788,
      "memory(GiB)": 107.26,
      "step": 30180,
      "train_speed(iter/s)": 1.633482
    },
    {
      "acc": 0.64229665,
      "epoch": 0.7657280568239473,
      "grad_norm": 5.125,
      "learning_rate": 7.264587797907248e-06,
      "loss": 1.60261612,
      "memory(GiB)": 107.26,
      "step": 30185,
      "train_speed(iter/s)": 1.633512
    },
    {
      "acc": 0.65486569,
      "epoch": 0.7658548959918823,
      "grad_norm": 5.09375,
      "learning_rate": 7.263652845151354e-06,
      "loss": 1.56436367,
      "memory(GiB)": 107.26,
      "step": 30190,
      "train_speed(iter/s)": 1.633542
    },
    {
      "acc": 0.65718236,
      "epoch": 0.7659817351598174,
      "grad_norm": 6.0,
      "learning_rate": 7.262717792829903e-06,
      "loss": 1.58778667,
      "memory(GiB)": 107.26,
      "step": 30195,
      "train_speed(iter/s)": 1.633573
    },
    {
      "acc": 0.660671,
      "epoch": 0.7661085743277524,
      "grad_norm": 7.125,
      "learning_rate": 7.261782640984021e-06,
      "loss": 1.57647133,
      "memory(GiB)": 107.26,
      "step": 30200,
      "train_speed(iter/s)": 1.633603
    },
    {
      "acc": 0.63908391,
      "epoch": 0.7662354134956875,
      "grad_norm": 5.875,
      "learning_rate": 7.26084738965484e-06,
      "loss": 1.64984779,
      "memory(GiB)": 107.26,
      "step": 30205,
      "train_speed(iter/s)": 1.633633
    },
    {
      "acc": 0.65461035,
      "epoch": 0.7663622526636226,
      "grad_norm": 6.34375,
      "learning_rate": 7.2599120388834964e-06,
      "loss": 1.65573215,
      "memory(GiB)": 107.26,
      "step": 30210,
      "train_speed(iter/s)": 1.633663
    },
    {
      "acc": 0.66593251,
      "epoch": 0.7664890918315576,
      "grad_norm": 5.84375,
      "learning_rate": 7.258976588711133e-06,
      "loss": 1.59546089,
      "memory(GiB)": 107.26,
      "step": 30215,
      "train_speed(iter/s)": 1.633695
    },
    {
      "acc": 0.66262951,
      "epoch": 0.7666159309994927,
      "grad_norm": 6.28125,
      "learning_rate": 7.258041039178891e-06,
      "loss": 1.56573458,
      "memory(GiB)": 107.26,
      "step": 30220,
      "train_speed(iter/s)": 1.633725
    },
    {
      "acc": 0.64453897,
      "epoch": 0.7667427701674278,
      "grad_norm": 6.71875,
      "learning_rate": 7.257105390327925e-06,
      "loss": 1.64220371,
      "memory(GiB)": 107.26,
      "step": 30225,
      "train_speed(iter/s)": 1.633754
    },
    {
      "acc": 0.65888615,
      "epoch": 0.7668696093353627,
      "grad_norm": 5.46875,
      "learning_rate": 7.256169642199386e-06,
      "loss": 1.56288233,
      "memory(GiB)": 107.26,
      "step": 30230,
      "train_speed(iter/s)": 1.633784
    },
    {
      "acc": 0.649687,
      "epoch": 0.7669964485032978,
      "grad_norm": 5.625,
      "learning_rate": 7.255233794834432e-06,
      "loss": 1.61188564,
      "memory(GiB)": 107.26,
      "step": 30235,
      "train_speed(iter/s)": 1.633815
    },
    {
      "acc": 0.67573872,
      "epoch": 0.7671232876712328,
      "grad_norm": 6.0625,
      "learning_rate": 7.254297848274229e-06,
      "loss": 1.52309999,
      "memory(GiB)": 107.26,
      "step": 30240,
      "train_speed(iter/s)": 1.633845
    },
    {
      "acc": 0.6575633,
      "epoch": 0.7672501268391679,
      "grad_norm": 6.90625,
      "learning_rate": 7.25336180255994e-06,
      "loss": 1.60211353,
      "memory(GiB)": 107.26,
      "step": 30245,
      "train_speed(iter/s)": 1.633875
    },
    {
      "acc": 0.65879192,
      "epoch": 0.767376966007103,
      "grad_norm": 4.875,
      "learning_rate": 7.25242565773274e-06,
      "loss": 1.66904621,
      "memory(GiB)": 107.26,
      "step": 30250,
      "train_speed(iter/s)": 1.633902
    },
    {
      "acc": 0.65146317,
      "epoch": 0.767503805175038,
      "grad_norm": 5.09375,
      "learning_rate": 7.251489413833801e-06,
      "loss": 1.62308445,
      "memory(GiB)": 107.26,
      "step": 30255,
      "train_speed(iter/s)": 1.633932
    },
    {
      "acc": 0.65045757,
      "epoch": 0.7676306443429731,
      "grad_norm": 5.875,
      "learning_rate": 7.250553070904307e-06,
      "loss": 1.65252571,
      "memory(GiB)": 107.26,
      "step": 30260,
      "train_speed(iter/s)": 1.633962
    },
    {
      "acc": 0.66111755,
      "epoch": 0.7677574835109082,
      "grad_norm": 5.9375,
      "learning_rate": 7.2496166289854404e-06,
      "loss": 1.57442474,
      "memory(GiB)": 107.26,
      "step": 30265,
      "train_speed(iter/s)": 1.633992
    },
    {
      "acc": 0.65822687,
      "epoch": 0.7678843226788432,
      "grad_norm": 7.4375,
      "learning_rate": 7.24868008811839e-06,
      "loss": 1.60991058,
      "memory(GiB)": 107.26,
      "step": 30270,
      "train_speed(iter/s)": 1.634024
    },
    {
      "acc": 0.66016645,
      "epoch": 0.7680111618467783,
      "grad_norm": 5.78125,
      "learning_rate": 7.247743448344351e-06,
      "loss": 1.61061821,
      "memory(GiB)": 107.26,
      "step": 30275,
      "train_speed(iter/s)": 1.634055
    },
    {
      "acc": 0.64137654,
      "epoch": 0.7681380010147133,
      "grad_norm": 4.96875,
      "learning_rate": 7.246806709704519e-06,
      "loss": 1.6832943,
      "memory(GiB)": 107.26,
      "step": 30280,
      "train_speed(iter/s)": 1.634083
    },
    {
      "acc": 0.64787626,
      "epoch": 0.7682648401826484,
      "grad_norm": 5.1875,
      "learning_rate": 7.245869872240098e-06,
      "loss": 1.6844635,
      "memory(GiB)": 107.26,
      "step": 30285,
      "train_speed(iter/s)": 1.63411
    },
    {
      "acc": 0.66646914,
      "epoch": 0.7683916793505835,
      "grad_norm": 5.625,
      "learning_rate": 7.244932935992292e-06,
      "loss": 1.63848801,
      "memory(GiB)": 107.26,
      "step": 30290,
      "train_speed(iter/s)": 1.63414
    },
    {
      "acc": 0.65764675,
      "epoch": 0.7685185185185185,
      "grad_norm": 7.0,
      "learning_rate": 7.243995901002312e-06,
      "loss": 1.66301079,
      "memory(GiB)": 107.26,
      "step": 30295,
      "train_speed(iter/s)": 1.63417
    },
    {
      "acc": 0.65270033,
      "epoch": 0.7686453576864536,
      "grad_norm": 5.84375,
      "learning_rate": 7.243058767311374e-06,
      "loss": 1.60568981,
      "memory(GiB)": 107.26,
      "step": 30300,
      "train_speed(iter/s)": 1.634199
    },
    {
      "acc": 0.67055054,
      "epoch": 0.7687721968543887,
      "grad_norm": 6.78125,
      "learning_rate": 7.2421215349606955e-06,
      "loss": 1.54382925,
      "memory(GiB)": 107.26,
      "step": 30305,
      "train_speed(iter/s)": 1.63423
    },
    {
      "acc": 0.66608658,
      "epoch": 0.7688990360223237,
      "grad_norm": 5.9375,
      "learning_rate": 7.241184203991505e-06,
      "loss": 1.51537476,
      "memory(GiB)": 107.26,
      "step": 30310,
      "train_speed(iter/s)": 1.634258
    },
    {
      "acc": 0.64886823,
      "epoch": 0.7690258751902588,
      "grad_norm": 5.9375,
      "learning_rate": 7.240246774445024e-06,
      "loss": 1.59635525,
      "memory(GiB)": 107.26,
      "step": 30315,
      "train_speed(iter/s)": 1.634288
    },
    {
      "acc": 0.65674829,
      "epoch": 0.7691527143581938,
      "grad_norm": 5.90625,
      "learning_rate": 7.23930924636249e-06,
      "loss": 1.62847023,
      "memory(GiB)": 107.26,
      "step": 30320,
      "train_speed(iter/s)": 1.634316
    },
    {
      "acc": 0.64854236,
      "epoch": 0.7692795535261289,
      "grad_norm": 5.8125,
      "learning_rate": 7.238371619785134e-06,
      "loss": 1.58781557,
      "memory(GiB)": 107.26,
      "step": 30325,
      "train_speed(iter/s)": 1.634346
    },
    {
      "acc": 0.65082769,
      "epoch": 0.769406392694064,
      "grad_norm": 5.46875,
      "learning_rate": 7.237433894754205e-06,
      "loss": 1.62256737,
      "memory(GiB)": 107.26,
      "step": 30330,
      "train_speed(iter/s)": 1.634373
    },
    {
      "acc": 0.65163741,
      "epoch": 0.769533231861999,
      "grad_norm": 5.8125,
      "learning_rate": 7.23649607131094e-06,
      "loss": 1.64911957,
      "memory(GiB)": 107.26,
      "step": 30335,
      "train_speed(iter/s)": 1.634402
    },
    {
      "acc": 0.65260997,
      "epoch": 0.7696600710299341,
      "grad_norm": 5.5,
      "learning_rate": 7.235558149496595e-06,
      "loss": 1.59176273,
      "memory(GiB)": 107.26,
      "step": 30340,
      "train_speed(iter/s)": 1.634433
    },
    {
      "acc": 0.66915121,
      "epoch": 0.7697869101978692,
      "grad_norm": 5.8125,
      "learning_rate": 7.23462012935242e-06,
      "loss": 1.58822021,
      "memory(GiB)": 107.26,
      "step": 30345,
      "train_speed(iter/s)": 1.634462
    },
    {
      "acc": 0.64483595,
      "epoch": 0.7699137493658041,
      "grad_norm": 5.5625,
      "learning_rate": 7.233682010919676e-06,
      "loss": 1.62909527,
      "memory(GiB)": 107.26,
      "step": 30350,
      "train_speed(iter/s)": 1.634492
    },
    {
      "acc": 0.66873941,
      "epoch": 0.7700405885337392,
      "grad_norm": 4.9375,
      "learning_rate": 7.2327437942396236e-06,
      "loss": 1.5787117,
      "memory(GiB)": 107.26,
      "step": 30355,
      "train_speed(iter/s)": 1.634521
    },
    {
      "acc": 0.66961937,
      "epoch": 0.7701674277016742,
      "grad_norm": 6.0625,
      "learning_rate": 7.231805479353532e-06,
      "loss": 1.54168282,
      "memory(GiB)": 107.26,
      "step": 30360,
      "train_speed(iter/s)": 1.634548
    },
    {
      "acc": 0.66384664,
      "epoch": 0.7702942668696093,
      "grad_norm": 5.0625,
      "learning_rate": 7.2308670663026705e-06,
      "loss": 1.639851,
      "memory(GiB)": 107.26,
      "step": 30365,
      "train_speed(iter/s)": 1.634579
    },
    {
      "acc": 0.64510469,
      "epoch": 0.7704211060375444,
      "grad_norm": 4.84375,
      "learning_rate": 7.229928555128315e-06,
      "loss": 1.65079098,
      "memory(GiB)": 107.26,
      "step": 30370,
      "train_speed(iter/s)": 1.634608
    },
    {
      "acc": 0.66312037,
      "epoch": 0.7705479452054794,
      "grad_norm": 6.625,
      "learning_rate": 7.228989945871745e-06,
      "loss": 1.63608627,
      "memory(GiB)": 107.26,
      "step": 30375,
      "train_speed(iter/s)": 1.634636
    },
    {
      "acc": 0.65444736,
      "epoch": 0.7706747843734145,
      "grad_norm": 5.09375,
      "learning_rate": 7.2280512385742475e-06,
      "loss": 1.5800992,
      "memory(GiB)": 107.26,
      "step": 30380,
      "train_speed(iter/s)": 1.634665
    },
    {
      "acc": 0.65825605,
      "epoch": 0.7708016235413496,
      "grad_norm": 6.34375,
      "learning_rate": 7.227112433277107e-06,
      "loss": 1.58305798,
      "memory(GiB)": 107.26,
      "step": 30385,
      "train_speed(iter/s)": 1.634694
    },
    {
      "acc": 0.65663486,
      "epoch": 0.7709284627092846,
      "grad_norm": 6.59375,
      "learning_rate": 7.2261735300216195e-06,
      "loss": 1.63428841,
      "memory(GiB)": 107.26,
      "step": 30390,
      "train_speed(iter/s)": 1.634721
    },
    {
      "acc": 0.65226974,
      "epoch": 0.7710553018772197,
      "grad_norm": 6.75,
      "learning_rate": 7.22523452884908e-06,
      "loss": 1.65994682,
      "memory(GiB)": 107.26,
      "step": 30395,
      "train_speed(iter/s)": 1.634751
    },
    {
      "acc": 0.63917208,
      "epoch": 0.7711821410451547,
      "grad_norm": 6.0625,
      "learning_rate": 7.224295429800792e-06,
      "loss": 1.68284969,
      "memory(GiB)": 107.26,
      "step": 30400,
      "train_speed(iter/s)": 1.634782
    },
    {
      "acc": 0.65070257,
      "epoch": 0.7713089802130898,
      "grad_norm": 5.46875,
      "learning_rate": 7.22335623291806e-06,
      "loss": 1.64408379,
      "memory(GiB)": 107.26,
      "step": 30405,
      "train_speed(iter/s)": 1.634812
    },
    {
      "acc": 0.65407734,
      "epoch": 0.7714358193810249,
      "grad_norm": 5.15625,
      "learning_rate": 7.222416938242194e-06,
      "loss": 1.64246502,
      "memory(GiB)": 107.26,
      "step": 30410,
      "train_speed(iter/s)": 1.634843
    },
    {
      "acc": 0.67182989,
      "epoch": 0.7715626585489599,
      "grad_norm": 6.40625,
      "learning_rate": 7.221477545814509e-06,
      "loss": 1.4954545,
      "memory(GiB)": 107.26,
      "step": 30415,
      "train_speed(iter/s)": 1.634872
    },
    {
      "acc": 0.65843563,
      "epoch": 0.771689497716895,
      "grad_norm": 5.0,
      "learning_rate": 7.220538055676323e-06,
      "loss": 1.60254974,
      "memory(GiB)": 107.26,
      "step": 30420,
      "train_speed(iter/s)": 1.634901
    },
    {
      "acc": 0.64874716,
      "epoch": 0.7718163368848301,
      "grad_norm": 9.5625,
      "learning_rate": 7.21959846786896e-06,
      "loss": 1.61789932,
      "memory(GiB)": 107.26,
      "step": 30425,
      "train_speed(iter/s)": 1.634928
    },
    {
      "acc": 0.6658041,
      "epoch": 0.7719431760527651,
      "grad_norm": 6.84375,
      "learning_rate": 7.218658782433746e-06,
      "loss": 1.57872143,
      "memory(GiB)": 107.26,
      "step": 30430,
      "train_speed(iter/s)": 1.634957
    },
    {
      "acc": 0.66571145,
      "epoch": 0.7720700152207002,
      "grad_norm": 7.6875,
      "learning_rate": 7.217718999412013e-06,
      "loss": 1.54681292,
      "memory(GiB)": 107.26,
      "step": 30435,
      "train_speed(iter/s)": 1.634987
    },
    {
      "acc": 0.65551558,
      "epoch": 0.7721968543886352,
      "grad_norm": 9.9375,
      "learning_rate": 7.216779118845097e-06,
      "loss": 1.61225777,
      "memory(GiB)": 107.26,
      "step": 30440,
      "train_speed(iter/s)": 1.635016
    },
    {
      "acc": 0.64738917,
      "epoch": 0.7723236935565703,
      "grad_norm": 6.21875,
      "learning_rate": 7.215839140774339e-06,
      "loss": 1.60115471,
      "memory(GiB)": 107.26,
      "step": 30445,
      "train_speed(iter/s)": 1.635046
    },
    {
      "acc": 0.65567551,
      "epoch": 0.7724505327245054,
      "grad_norm": 5.375,
      "learning_rate": 7.214899065241082e-06,
      "loss": 1.63994865,
      "memory(GiB)": 107.26,
      "step": 30450,
      "train_speed(iter/s)": 1.635074
    },
    {
      "acc": 0.66446095,
      "epoch": 0.7725773718924404,
      "grad_norm": 6.6875,
      "learning_rate": 7.213958892286674e-06,
      "loss": 1.59551678,
      "memory(GiB)": 107.26,
      "step": 30455,
      "train_speed(iter/s)": 1.635103
    },
    {
      "acc": 0.65487013,
      "epoch": 0.7727042110603755,
      "grad_norm": 5.40625,
      "learning_rate": 7.213018621952472e-06,
      "loss": 1.57060051,
      "memory(GiB)": 107.26,
      "step": 30460,
      "train_speed(iter/s)": 1.635134
    },
    {
      "acc": 0.66386309,
      "epoch": 0.7728310502283106,
      "grad_norm": 6.09375,
      "learning_rate": 7.212078254279828e-06,
      "loss": 1.54142752,
      "memory(GiB)": 107.26,
      "step": 30465,
      "train_speed(iter/s)": 1.635164
    },
    {
      "acc": 0.64772062,
      "epoch": 0.7729578893962455,
      "grad_norm": 5.25,
      "learning_rate": 7.211137789310109e-06,
      "loss": 1.66402931,
      "memory(GiB)": 107.26,
      "step": 30470,
      "train_speed(iter/s)": 1.635193
    },
    {
      "acc": 0.67215691,
      "epoch": 0.7730847285641806,
      "grad_norm": 5.34375,
      "learning_rate": 7.2101972270846756e-06,
      "loss": 1.52053585,
      "memory(GiB)": 107.26,
      "step": 30475,
      "train_speed(iter/s)": 1.635222
    },
    {
      "acc": 0.67320266,
      "epoch": 0.7732115677321156,
      "grad_norm": 5.03125,
      "learning_rate": 7.2092565676449e-06,
      "loss": 1.49859238,
      "memory(GiB)": 107.26,
      "step": 30480,
      "train_speed(iter/s)": 1.635251
    },
    {
      "acc": 0.67555428,
      "epoch": 0.7733384069000507,
      "grad_norm": 6.5625,
      "learning_rate": 7.208315811032158e-06,
      "loss": 1.48835354,
      "memory(GiB)": 107.26,
      "step": 30485,
      "train_speed(iter/s)": 1.63528
    },
    {
      "acc": 0.66074233,
      "epoch": 0.7734652460679858,
      "grad_norm": 6.3125,
      "learning_rate": 7.207374957287828e-06,
      "loss": 1.58826437,
      "memory(GiB)": 107.26,
      "step": 30490,
      "train_speed(iter/s)": 1.635312
    },
    {
      "acc": 0.65030069,
      "epoch": 0.7735920852359208,
      "grad_norm": 5.84375,
      "learning_rate": 7.2064340064532914e-06,
      "loss": 1.62204819,
      "memory(GiB)": 107.26,
      "step": 30495,
      "train_speed(iter/s)": 1.63534
    },
    {
      "acc": 0.62768269,
      "epoch": 0.7737189244038559,
      "grad_norm": 5.84375,
      "learning_rate": 7.205492958569936e-06,
      "loss": 1.67540817,
      "memory(GiB)": 107.26,
      "step": 30500,
      "train_speed(iter/s)": 1.635371
    },
    {
      "acc": 0.66002293,
      "epoch": 0.773845763571791,
      "grad_norm": 5.3125,
      "learning_rate": 7.204551813679154e-06,
      "loss": 1.6727169,
      "memory(GiB)": 107.26,
      "step": 30505,
      "train_speed(iter/s)": 1.6354
    },
    {
      "acc": 0.64538474,
      "epoch": 0.773972602739726,
      "grad_norm": 5.96875,
      "learning_rate": 7.2036105718223405e-06,
      "loss": 1.65025215,
      "memory(GiB)": 107.26,
      "step": 30510,
      "train_speed(iter/s)": 1.63543
    },
    {
      "acc": 0.6518836,
      "epoch": 0.7740994419076611,
      "grad_norm": 5.875,
      "learning_rate": 7.202669233040896e-06,
      "loss": 1.63001099,
      "memory(GiB)": 107.26,
      "step": 30515,
      "train_speed(iter/s)": 1.635461
    },
    {
      "acc": 0.65905566,
      "epoch": 0.7742262810755961,
      "grad_norm": 6.21875,
      "learning_rate": 7.201727797376223e-06,
      "loss": 1.65753632,
      "memory(GiB)": 107.26,
      "step": 30520,
      "train_speed(iter/s)": 1.635488
    },
    {
      "acc": 0.6614409,
      "epoch": 0.7743531202435312,
      "grad_norm": 5.4375,
      "learning_rate": 7.200786264869732e-06,
      "loss": 1.61937027,
      "memory(GiB)": 107.26,
      "step": 30525,
      "train_speed(iter/s)": 1.635517
    },
    {
      "acc": 0.64703631,
      "epoch": 0.7744799594114663,
      "grad_norm": 6.75,
      "learning_rate": 7.199844635562836e-06,
      "loss": 1.6106411,
      "memory(GiB)": 107.26,
      "step": 30530,
      "train_speed(iter/s)": 1.635545
    },
    {
      "acc": 0.64952555,
      "epoch": 0.7746067985794013,
      "grad_norm": 4.90625,
      "learning_rate": 7.19890290949695e-06,
      "loss": 1.64010735,
      "memory(GiB)": 107.26,
      "step": 30535,
      "train_speed(iter/s)": 1.635574
    },
    {
      "acc": 0.63504782,
      "epoch": 0.7747336377473364,
      "grad_norm": 5.6875,
      "learning_rate": 7.197961086713498e-06,
      "loss": 1.64006424,
      "memory(GiB)": 107.26,
      "step": 30540,
      "train_speed(iter/s)": 1.635604
    },
    {
      "acc": 0.64450493,
      "epoch": 0.7748604769152715,
      "grad_norm": 6.90625,
      "learning_rate": 7.197019167253904e-06,
      "loss": 1.61944942,
      "memory(GiB)": 107.26,
      "step": 30545,
      "train_speed(iter/s)": 1.635636
    },
    {
      "acc": 0.66522994,
      "epoch": 0.7749873160832065,
      "grad_norm": 4.25,
      "learning_rate": 7.196077151159597e-06,
      "loss": 1.60584488,
      "memory(GiB)": 107.26,
      "step": 30550,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.66381049,
      "epoch": 0.7751141552511416,
      "grad_norm": 5.75,
      "learning_rate": 7.195135038472013e-06,
      "loss": 1.5891283,
      "memory(GiB)": 107.26,
      "step": 30555,
      "train_speed(iter/s)": 1.635697
    },
    {
      "acc": 0.68021297,
      "epoch": 0.7752409944190766,
      "grad_norm": 5.3125,
      "learning_rate": 7.194192829232589e-06,
      "loss": 1.58056335,
      "memory(GiB)": 107.26,
      "step": 30560,
      "train_speed(iter/s)": 1.635725
    },
    {
      "acc": 0.65287666,
      "epoch": 0.7753678335870117,
      "grad_norm": 6.875,
      "learning_rate": 7.1932505234827686e-06,
      "loss": 1.60172901,
      "memory(GiB)": 107.26,
      "step": 30565,
      "train_speed(iter/s)": 1.635756
    },
    {
      "acc": 0.65607567,
      "epoch": 0.7754946727549468,
      "grad_norm": 5.8125,
      "learning_rate": 7.192308121263998e-06,
      "loss": 1.65015259,
      "memory(GiB)": 107.26,
      "step": 30570,
      "train_speed(iter/s)": 1.635786
    },
    {
      "acc": 0.6658083,
      "epoch": 0.7756215119228818,
      "grad_norm": 5.90625,
      "learning_rate": 7.191365622617728e-06,
      "loss": 1.59011879,
      "memory(GiB)": 107.26,
      "step": 30575,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.64922142,
      "epoch": 0.7757483510908169,
      "grad_norm": 5.09375,
      "learning_rate": 7.190423027585414e-06,
      "loss": 1.65585232,
      "memory(GiB)": 107.26,
      "step": 30580,
      "train_speed(iter/s)": 1.635846
    },
    {
      "acc": 0.65652857,
      "epoch": 0.775875190258752,
      "grad_norm": 8.125,
      "learning_rate": 7.189480336208516e-06,
      "loss": 1.60404053,
      "memory(GiB)": 107.26,
      "step": 30585,
      "train_speed(iter/s)": 1.635878
    },
    {
      "acc": 0.65228233,
      "epoch": 0.776002029426687,
      "grad_norm": 6.5625,
      "learning_rate": 7.188537548528498e-06,
      "loss": 1.59489765,
      "memory(GiB)": 107.26,
      "step": 30590,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.64943886,
      "epoch": 0.776128868594622,
      "grad_norm": 5.53125,
      "learning_rate": 7.187594664586826e-06,
      "loss": 1.61623878,
      "memory(GiB)": 107.26,
      "step": 30595,
      "train_speed(iter/s)": 1.635936
    },
    {
      "acc": 0.65434179,
      "epoch": 0.776255707762557,
      "grad_norm": 5.46875,
      "learning_rate": 7.186651684424975e-06,
      "loss": 1.66895905,
      "memory(GiB)": 107.26,
      "step": 30600,
      "train_speed(iter/s)": 1.635966
    },
    {
      "acc": 0.64830589,
      "epoch": 0.7763825469304921,
      "grad_norm": 7.90625,
      "learning_rate": 7.185708608084418e-06,
      "loss": 1.6182003,
      "memory(GiB)": 107.26,
      "step": 30605,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.66292844,
      "epoch": 0.7765093860984272,
      "grad_norm": 4.96875,
      "learning_rate": 7.184765435606642e-06,
      "loss": 1.57399521,
      "memory(GiB)": 107.26,
      "step": 30610,
      "train_speed(iter/s)": 1.636024
    },
    {
      "acc": 0.66935806,
      "epoch": 0.7766362252663622,
      "grad_norm": 6.21875,
      "learning_rate": 7.183822167033124e-06,
      "loss": 1.61123848,
      "memory(GiB)": 107.26,
      "step": 30615,
      "train_speed(iter/s)": 1.636053
    },
    {
      "acc": 0.65486174,
      "epoch": 0.7767630644342973,
      "grad_norm": 6.03125,
      "learning_rate": 7.18287880240536e-06,
      "loss": 1.60520649,
      "memory(GiB)": 107.26,
      "step": 30620,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.67707996,
      "epoch": 0.7768899036022324,
      "grad_norm": 6.96875,
      "learning_rate": 7.1819353417648386e-06,
      "loss": 1.55840244,
      "memory(GiB)": 107.26,
      "step": 30625,
      "train_speed(iter/s)": 1.636113
    },
    {
      "acc": 0.65891223,
      "epoch": 0.7770167427701674,
      "grad_norm": 6.0,
      "learning_rate": 7.180991785153059e-06,
      "loss": 1.62628822,
      "memory(GiB)": 107.26,
      "step": 30630,
      "train_speed(iter/s)": 1.636141
    },
    {
      "acc": 0.65944872,
      "epoch": 0.7771435819381025,
      "grad_norm": 5.46875,
      "learning_rate": 7.180048132611524e-06,
      "loss": 1.5419241,
      "memory(GiB)": 107.26,
      "step": 30635,
      "train_speed(iter/s)": 1.636171
    },
    {
      "acc": 0.65319638,
      "epoch": 0.7772704211060375,
      "grad_norm": 10.25,
      "learning_rate": 7.17910438418174e-06,
      "loss": 1.58357925,
      "memory(GiB)": 107.26,
      "step": 30640,
      "train_speed(iter/s)": 1.636199
    },
    {
      "acc": 0.64738016,
      "epoch": 0.7773972602739726,
      "grad_norm": 5.875,
      "learning_rate": 7.178160539905214e-06,
      "loss": 1.65631466,
      "memory(GiB)": 107.26,
      "step": 30645,
      "train_speed(iter/s)": 1.63623
    },
    {
      "acc": 0.66691637,
      "epoch": 0.7775240994419077,
      "grad_norm": 6.1875,
      "learning_rate": 7.1772165998234645e-06,
      "loss": 1.55498533,
      "memory(GiB)": 107.26,
      "step": 30650,
      "train_speed(iter/s)": 1.636258
    },
    {
      "acc": 0.64904814,
      "epoch": 0.7776509386098427,
      "grad_norm": 4.875,
      "learning_rate": 7.176272563978007e-06,
      "loss": 1.64443932,
      "memory(GiB)": 107.26,
      "step": 30655,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.65022783,
      "epoch": 0.7777777777777778,
      "grad_norm": 6.125,
      "learning_rate": 7.175328432410367e-06,
      "loss": 1.61195831,
      "memory(GiB)": 107.26,
      "step": 30660,
      "train_speed(iter/s)": 1.636316
    },
    {
      "acc": 0.64395561,
      "epoch": 0.7779046169457129,
      "grad_norm": 6.3125,
      "learning_rate": 7.17438420516207e-06,
      "loss": 1.66042862,
      "memory(GiB)": 107.26,
      "step": 30665,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.6630724,
      "epoch": 0.7780314561136479,
      "grad_norm": 5.84375,
      "learning_rate": 7.173439882274647e-06,
      "loss": 1.53771601,
      "memory(GiB)": 107.26,
      "step": 30670,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.65414743,
      "epoch": 0.778158295281583,
      "grad_norm": 5.6875,
      "learning_rate": 7.172495463789635e-06,
      "loss": 1.63615723,
      "memory(GiB)": 107.26,
      "step": 30675,
      "train_speed(iter/s)": 1.636402
    },
    {
      "acc": 0.65784302,
      "epoch": 0.778285134449518,
      "grad_norm": 5.6875,
      "learning_rate": 7.171550949748574e-06,
      "loss": 1.61946259,
      "memory(GiB)": 107.26,
      "step": 30680,
      "train_speed(iter/s)": 1.636431
    },
    {
      "acc": 0.65488625,
      "epoch": 0.7784119736174531,
      "grad_norm": 5.0,
      "learning_rate": 7.170606340193003e-06,
      "loss": 1.59765997,
      "memory(GiB)": 107.26,
      "step": 30685,
      "train_speed(iter/s)": 1.636461
    },
    {
      "acc": 0.65165424,
      "epoch": 0.7785388127853882,
      "grad_norm": 5.84375,
      "learning_rate": 7.1696616351644786e-06,
      "loss": 1.67578697,
      "memory(GiB)": 107.26,
      "step": 30690,
      "train_speed(iter/s)": 1.636489
    },
    {
      "acc": 0.6690865,
      "epoch": 0.7786656519533232,
      "grad_norm": 5.875,
      "learning_rate": 7.168716834704546e-06,
      "loss": 1.60980949,
      "memory(GiB)": 107.26,
      "step": 30695,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.65484409,
      "epoch": 0.7787924911212583,
      "grad_norm": 5.34375,
      "learning_rate": 7.167771938854766e-06,
      "loss": 1.62836246,
      "memory(GiB)": 107.26,
      "step": 30700,
      "train_speed(iter/s)": 1.636551
    },
    {
      "acc": 0.65498524,
      "epoch": 0.7789193302891934,
      "grad_norm": 7.9375,
      "learning_rate": 7.166826947656696e-06,
      "loss": 1.69484024,
      "memory(GiB)": 107.26,
      "step": 30705,
      "train_speed(iter/s)": 1.636582
    },
    {
      "acc": 0.65259657,
      "epoch": 0.7790461694571283,
      "grad_norm": 4.75,
      "learning_rate": 7.165881861151904e-06,
      "loss": 1.60934887,
      "memory(GiB)": 107.26,
      "step": 30710,
      "train_speed(iter/s)": 1.63661
    },
    {
      "acc": 0.6601182,
      "epoch": 0.7791730086250634,
      "grad_norm": 6.09375,
      "learning_rate": 7.164936679381957e-06,
      "loss": 1.59868069,
      "memory(GiB)": 107.26,
      "step": 30715,
      "train_speed(iter/s)": 1.636638
    },
    {
      "acc": 0.65729599,
      "epoch": 0.7792998477929984,
      "grad_norm": 5.53125,
      "learning_rate": 7.16399140238843e-06,
      "loss": 1.65911503,
      "memory(GiB)": 107.26,
      "step": 30720,
      "train_speed(iter/s)": 1.636666
    },
    {
      "acc": 0.65079694,
      "epoch": 0.7794266869609335,
      "grad_norm": 4.9375,
      "learning_rate": 7.163046030212899e-06,
      "loss": 1.62213116,
      "memory(GiB)": 107.26,
      "step": 30725,
      "train_speed(iter/s)": 1.636696
    },
    {
      "acc": 0.66206989,
      "epoch": 0.7795535261288686,
      "grad_norm": 5.53125,
      "learning_rate": 7.1621005628969475e-06,
      "loss": 1.5307085,
      "memory(GiB)": 107.26,
      "step": 30730,
      "train_speed(iter/s)": 1.636726
    },
    {
      "acc": 0.64623404,
      "epoch": 0.7796803652968036,
      "grad_norm": 6.34375,
      "learning_rate": 7.161155000482159e-06,
      "loss": 1.6867939,
      "memory(GiB)": 107.26,
      "step": 30735,
      "train_speed(iter/s)": 1.636755
    },
    {
      "acc": 0.63943319,
      "epoch": 0.7798072044647387,
      "grad_norm": 5.53125,
      "learning_rate": 7.160209343010125e-06,
      "loss": 1.63440323,
      "memory(GiB)": 107.26,
      "step": 30740,
      "train_speed(iter/s)": 1.636783
    },
    {
      "acc": 0.65245943,
      "epoch": 0.7799340436326738,
      "grad_norm": 5.4375,
      "learning_rate": 7.1592635905224386e-06,
      "loss": 1.67462692,
      "memory(GiB)": 107.26,
      "step": 30745,
      "train_speed(iter/s)": 1.636812
    },
    {
      "acc": 0.64611311,
      "epoch": 0.7800608828006088,
      "grad_norm": 6.34375,
      "learning_rate": 7.1583177430606995e-06,
      "loss": 1.67396774,
      "memory(GiB)": 107.26,
      "step": 30750,
      "train_speed(iter/s)": 1.63684
    },
    {
      "acc": 0.64364958,
      "epoch": 0.7801877219685439,
      "grad_norm": 5.6875,
      "learning_rate": 7.1573718006665095e-06,
      "loss": 1.6576107,
      "memory(GiB)": 107.26,
      "step": 30755,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.64075098,
      "epoch": 0.7803145611364789,
      "grad_norm": 7.03125,
      "learning_rate": 7.156425763381477e-06,
      "loss": 1.66395149,
      "memory(GiB)": 107.26,
      "step": 30760,
      "train_speed(iter/s)": 1.636899
    },
    {
      "acc": 0.66395569,
      "epoch": 0.780441400304414,
      "grad_norm": 5.6875,
      "learning_rate": 7.155479631247211e-06,
      "loss": 1.57881393,
      "memory(GiB)": 107.26,
      "step": 30765,
      "train_speed(iter/s)": 1.636928
    },
    {
      "acc": 0.64132605,
      "epoch": 0.7805682394723491,
      "grad_norm": 5.28125,
      "learning_rate": 7.154533404305327e-06,
      "loss": 1.65135784,
      "memory(GiB)": 107.26,
      "step": 30770,
      "train_speed(iter/s)": 1.636956
    },
    {
      "acc": 0.66526337,
      "epoch": 0.7806950786402841,
      "grad_norm": 5.875,
      "learning_rate": 7.153587082597445e-06,
      "loss": 1.47362204,
      "memory(GiB)": 107.26,
      "step": 30775,
      "train_speed(iter/s)": 1.636983
    },
    {
      "acc": 0.66942844,
      "epoch": 0.7808219178082192,
      "grad_norm": 5.3125,
      "learning_rate": 7.152640666165187e-06,
      "loss": 1.50902376,
      "memory(GiB)": 107.26,
      "step": 30780,
      "train_speed(iter/s)": 1.637014
    },
    {
      "acc": 0.66340733,
      "epoch": 0.7809487569761543,
      "grad_norm": 5.4375,
      "learning_rate": 7.151694155050184e-06,
      "loss": 1.55822525,
      "memory(GiB)": 107.26,
      "step": 30785,
      "train_speed(iter/s)": 1.637042
    },
    {
      "acc": 0.65109696,
      "epoch": 0.7810755961440893,
      "grad_norm": 5.0625,
      "learning_rate": 7.150747549294064e-06,
      "loss": 1.66960526,
      "memory(GiB)": 107.26,
      "step": 30790,
      "train_speed(iter/s)": 1.637071
    },
    {
      "acc": 0.65702529,
      "epoch": 0.7812024353120244,
      "grad_norm": 6.625,
      "learning_rate": 7.149800848938464e-06,
      "loss": 1.57980947,
      "memory(GiB)": 107.26,
      "step": 30795,
      "train_speed(iter/s)": 1.6371
    },
    {
      "acc": 0.63773665,
      "epoch": 0.7813292744799594,
      "grad_norm": 5.90625,
      "learning_rate": 7.1488540540250254e-06,
      "loss": 1.65127792,
      "memory(GiB)": 107.26,
      "step": 30800,
      "train_speed(iter/s)": 1.637129
    },
    {
      "acc": 0.65122681,
      "epoch": 0.7814561136478945,
      "grad_norm": 5.0625,
      "learning_rate": 7.14790716459539e-06,
      "loss": 1.60646725,
      "memory(GiB)": 107.26,
      "step": 30805,
      "train_speed(iter/s)": 1.637156
    },
    {
      "acc": 0.66043463,
      "epoch": 0.7815829528158296,
      "grad_norm": 5.15625,
      "learning_rate": 7.146960180691209e-06,
      "loss": 1.55198698,
      "memory(GiB)": 107.26,
      "step": 30810,
      "train_speed(iter/s)": 1.637185
    },
    {
      "acc": 0.64816837,
      "epoch": 0.7817097919837646,
      "grad_norm": 5.84375,
      "learning_rate": 7.146013102354133e-06,
      "loss": 1.60648384,
      "memory(GiB)": 107.26,
      "step": 30815,
      "train_speed(iter/s)": 1.637215
    },
    {
      "acc": 0.66276803,
      "epoch": 0.7818366311516997,
      "grad_norm": 5.5,
      "learning_rate": 7.145065929625821e-06,
      "loss": 1.54901552,
      "memory(GiB)": 107.26,
      "step": 30820,
      "train_speed(iter/s)": 1.637245
    },
    {
      "acc": 0.65730925,
      "epoch": 0.7819634703196348,
      "grad_norm": 5.4375,
      "learning_rate": 7.1441186625479304e-06,
      "loss": 1.63273067,
      "memory(GiB)": 107.26,
      "step": 30825,
      "train_speed(iter/s)": 1.637272
    },
    {
      "acc": 0.65262175,
      "epoch": 0.7820903094875697,
      "grad_norm": 7.25,
      "learning_rate": 7.143171301162131e-06,
      "loss": 1.64197006,
      "memory(GiB)": 107.26,
      "step": 30830,
      "train_speed(iter/s)": 1.637302
    },
    {
      "acc": 0.63268476,
      "epoch": 0.7822171486555048,
      "grad_norm": 6.09375,
      "learning_rate": 7.142223845510086e-06,
      "loss": 1.71115379,
      "memory(GiB)": 107.26,
      "step": 30835,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.64756174,
      "epoch": 0.7823439878234398,
      "grad_norm": 6.3125,
      "learning_rate": 7.1412762956334746e-06,
      "loss": 1.63337746,
      "memory(GiB)": 107.26,
      "step": 30840,
      "train_speed(iter/s)": 1.637363
    },
    {
      "acc": 0.65478659,
      "epoch": 0.7824708269913749,
      "grad_norm": 5.03125,
      "learning_rate": 7.140328651573969e-06,
      "loss": 1.58976927,
      "memory(GiB)": 107.26,
      "step": 30845,
      "train_speed(iter/s)": 1.637391
    },
    {
      "acc": 0.65226812,
      "epoch": 0.78259766615931,
      "grad_norm": 5.71875,
      "learning_rate": 7.139380913373255e-06,
      "loss": 1.61686172,
      "memory(GiB)": 107.26,
      "step": 30850,
      "train_speed(iter/s)": 1.63742
    },
    {
      "acc": 0.650634,
      "epoch": 0.782724505327245,
      "grad_norm": 5.03125,
      "learning_rate": 7.138433081073017e-06,
      "loss": 1.64412384,
      "memory(GiB)": 107.26,
      "step": 30855,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.66604409,
      "epoch": 0.7828513444951801,
      "grad_norm": 5.53125,
      "learning_rate": 7.137485154714945e-06,
      "loss": 1.5839673,
      "memory(GiB)": 107.26,
      "step": 30860,
      "train_speed(iter/s)": 1.637478
    },
    {
      "acc": 0.66617074,
      "epoch": 0.7829781836631152,
      "grad_norm": 5.59375,
      "learning_rate": 7.1365371343407304e-06,
      "loss": 1.59314184,
      "memory(GiB)": 107.26,
      "step": 30865,
      "train_speed(iter/s)": 1.637506
    },
    {
      "acc": 0.67707481,
      "epoch": 0.7831050228310502,
      "grad_norm": 6.125,
      "learning_rate": 7.135589019992076e-06,
      "loss": 1.59035444,
      "memory(GiB)": 107.26,
      "step": 30870,
      "train_speed(iter/s)": 1.637535
    },
    {
      "acc": 0.66602597,
      "epoch": 0.7832318619989853,
      "grad_norm": 6.1875,
      "learning_rate": 7.134640811710681e-06,
      "loss": 1.58947344,
      "memory(GiB)": 107.26,
      "step": 30875,
      "train_speed(iter/s)": 1.637563
    },
    {
      "acc": 0.64053383,
      "epoch": 0.7833587011669203,
      "grad_norm": 6.4375,
      "learning_rate": 7.133692509538253e-06,
      "loss": 1.60733871,
      "memory(GiB)": 107.26,
      "step": 30880,
      "train_speed(iter/s)": 1.637591
    },
    {
      "acc": 0.65207176,
      "epoch": 0.7834855403348554,
      "grad_norm": 5.65625,
      "learning_rate": 7.132744113516502e-06,
      "loss": 1.69174347,
      "memory(GiB)": 107.26,
      "step": 30885,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.67224426,
      "epoch": 0.7836123795027905,
      "grad_norm": 5.59375,
      "learning_rate": 7.1317956236871436e-06,
      "loss": 1.58318987,
      "memory(GiB)": 107.26,
      "step": 30890,
      "train_speed(iter/s)": 1.63765
    },
    {
      "acc": 0.66608834,
      "epoch": 0.7837392186707255,
      "grad_norm": 5.59375,
      "learning_rate": 7.130847040091893e-06,
      "loss": 1.58331203,
      "memory(GiB)": 107.26,
      "step": 30895,
      "train_speed(iter/s)": 1.637678
    },
    {
      "acc": 0.65211349,
      "epoch": 0.7838660578386606,
      "grad_norm": 5.53125,
      "learning_rate": 7.1298983627724795e-06,
      "loss": 1.6122673,
      "memory(GiB)": 107.26,
      "step": 30900,
      "train_speed(iter/s)": 1.637704
    },
    {
      "acc": 0.63837581,
      "epoch": 0.7839928970065957,
      "grad_norm": 6.5625,
      "learning_rate": 7.128949591770624e-06,
      "loss": 1.68804188,
      "memory(GiB)": 107.26,
      "step": 30905,
      "train_speed(iter/s)": 1.637736
    },
    {
      "acc": 0.66685181,
      "epoch": 0.7841197361745307,
      "grad_norm": 6.34375,
      "learning_rate": 7.128000727128063e-06,
      "loss": 1.63648643,
      "memory(GiB)": 107.26,
      "step": 30910,
      "train_speed(iter/s)": 1.637765
    },
    {
      "acc": 0.64562726,
      "epoch": 0.7842465753424658,
      "grad_norm": 6.09375,
      "learning_rate": 7.127051768886527e-06,
      "loss": 1.63818398,
      "memory(GiB)": 107.26,
      "step": 30915,
      "train_speed(iter/s)": 1.637794
    },
    {
      "acc": 0.67620625,
      "epoch": 0.7843734145104008,
      "grad_norm": 6.59375,
      "learning_rate": 7.126102717087758e-06,
      "loss": 1.55689049,
      "memory(GiB)": 107.26,
      "step": 30920,
      "train_speed(iter/s)": 1.637821
    },
    {
      "acc": 0.64565697,
      "epoch": 0.7845002536783359,
      "grad_norm": 5.09375,
      "learning_rate": 7.1251535717735e-06,
      "loss": 1.65126534,
      "memory(GiB)": 107.26,
      "step": 30925,
      "train_speed(iter/s)": 1.63785
    },
    {
      "acc": 0.64443326,
      "epoch": 0.784627092846271,
      "grad_norm": 7.03125,
      "learning_rate": 7.1242043329854995e-06,
      "loss": 1.69032688,
      "memory(GiB)": 107.26,
      "step": 30930,
      "train_speed(iter/s)": 1.637878
    },
    {
      "acc": 0.65155067,
      "epoch": 0.784753932014206,
      "grad_norm": 5.875,
      "learning_rate": 7.123255000765508e-06,
      "loss": 1.59071312,
      "memory(GiB)": 107.26,
      "step": 30935,
      "train_speed(iter/s)": 1.637908
    },
    {
      "acc": 0.65654821,
      "epoch": 0.7848807711821411,
      "grad_norm": 6.46875,
      "learning_rate": 7.122305575155283e-06,
      "loss": 1.60332165,
      "memory(GiB)": 107.26,
      "step": 30940,
      "train_speed(iter/s)": 1.637938
    },
    {
      "acc": 0.65992761,
      "epoch": 0.7850076103500762,
      "grad_norm": 5.5625,
      "learning_rate": 7.121356056196582e-06,
      "loss": 1.63548241,
      "memory(GiB)": 107.26,
      "step": 30945,
      "train_speed(iter/s)": 1.637966
    },
    {
      "acc": 0.67780595,
      "epoch": 0.7851344495180111,
      "grad_norm": 5.0625,
      "learning_rate": 7.1204064439311715e-06,
      "loss": 1.56389694,
      "memory(GiB)": 107.26,
      "step": 30950,
      "train_speed(iter/s)": 1.637994
    },
    {
      "acc": 0.66311407,
      "epoch": 0.7852612886859462,
      "grad_norm": 6.375,
      "learning_rate": 7.119456738400818e-06,
      "loss": 1.56350679,
      "memory(GiB)": 107.26,
      "step": 30955,
      "train_speed(iter/s)": 1.638023
    },
    {
      "acc": 0.66002016,
      "epoch": 0.7853881278538812,
      "grad_norm": 5.34375,
      "learning_rate": 7.118506939647295e-06,
      "loss": 1.49937019,
      "memory(GiB)": 107.26,
      "step": 30960,
      "train_speed(iter/s)": 1.638052
    },
    {
      "acc": 0.65607243,
      "epoch": 0.7855149670218163,
      "grad_norm": 5.90625,
      "learning_rate": 7.1175570477123776e-06,
      "loss": 1.62767658,
      "memory(GiB)": 107.26,
      "step": 30965,
      "train_speed(iter/s)": 1.638079
    },
    {
      "acc": 0.65050111,
      "epoch": 0.7856418061897514,
      "grad_norm": 5.9375,
      "learning_rate": 7.116607062637848e-06,
      "loss": 1.62889214,
      "memory(GiB)": 107.26,
      "step": 30970,
      "train_speed(iter/s)": 1.638106
    },
    {
      "acc": 0.65895529,
      "epoch": 0.7857686453576864,
      "grad_norm": 6.40625,
      "learning_rate": 7.115656984465489e-06,
      "loss": 1.57907181,
      "memory(GiB)": 107.26,
      "step": 30975,
      "train_speed(iter/s)": 1.638132
    },
    {
      "acc": 0.64549918,
      "epoch": 0.7858954845256215,
      "grad_norm": 6.03125,
      "learning_rate": 7.114706813237091e-06,
      "loss": 1.68856697,
      "memory(GiB)": 107.26,
      "step": 30980,
      "train_speed(iter/s)": 1.63816
    },
    {
      "acc": 0.65708437,
      "epoch": 0.7860223236935566,
      "grad_norm": 4.8125,
      "learning_rate": 7.1137565489944445e-06,
      "loss": 1.62159348,
      "memory(GiB)": 107.26,
      "step": 30985,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.66555815,
      "epoch": 0.7861491628614916,
      "grad_norm": 5.40625,
      "learning_rate": 7.112806191779349e-06,
      "loss": 1.56197338,
      "memory(GiB)": 107.26,
      "step": 30990,
      "train_speed(iter/s)": 1.638215
    },
    {
      "acc": 0.64561439,
      "epoch": 0.7862760020294267,
      "grad_norm": 5.21875,
      "learning_rate": 7.111855741633603e-06,
      "loss": 1.64821491,
      "memory(GiB)": 107.26,
      "step": 30995,
      "train_speed(iter/s)": 1.638242
    },
    {
      "acc": 0.65975032,
      "epoch": 0.7864028411973617,
      "grad_norm": 6.90625,
      "learning_rate": 7.1109051985990145e-06,
      "loss": 1.62765789,
      "memory(GiB)": 107.26,
      "step": 31000,
      "train_speed(iter/s)": 1.638271
    },
    {
      "epoch": 0.7864028411973617,
      "eval_acc": 0.6455726866266965,
      "eval_loss": 1.5767943859100342,
      "eval_runtime": 58.2768,
      "eval_samples_per_second": 109.306,
      "eval_steps_per_second": 27.335,
      "step": 31000
    },
    {
      "acc": 0.64757233,
      "epoch": 0.7865296803652968,
      "grad_norm": 4.875,
      "learning_rate": 7.109954562717389e-06,
      "loss": 1.63938522,
      "memory(GiB)": 107.26,
      "step": 31005,
      "train_speed(iter/s)": 1.632909
    },
    {
      "acc": 0.66574173,
      "epoch": 0.7866565195332319,
      "grad_norm": 9.3125,
      "learning_rate": 7.109003834030543e-06,
      "loss": 1.57466583,
      "memory(GiB)": 107.26,
      "step": 31010,
      "train_speed(iter/s)": 1.632937
    },
    {
      "acc": 0.65865822,
      "epoch": 0.7867833587011669,
      "grad_norm": 7.0,
      "learning_rate": 7.108053012580291e-06,
      "loss": 1.58323221,
      "memory(GiB)": 107.26,
      "step": 31015,
      "train_speed(iter/s)": 1.632965
    },
    {
      "acc": 0.65308952,
      "epoch": 0.786910197869102,
      "grad_norm": 5.1875,
      "learning_rate": 7.107102098408457e-06,
      "loss": 1.57040215,
      "memory(GiB)": 107.26,
      "step": 31020,
      "train_speed(iter/s)": 1.632993
    },
    {
      "acc": 0.65615973,
      "epoch": 0.7870370370370371,
      "grad_norm": 5.46875,
      "learning_rate": 7.106151091556865e-06,
      "loss": 1.6058672,
      "memory(GiB)": 107.26,
      "step": 31025,
      "train_speed(iter/s)": 1.633022
    },
    {
      "acc": 0.65740523,
      "epoch": 0.7871638762049721,
      "grad_norm": 5.28125,
      "learning_rate": 7.105199992067344e-06,
      "loss": 1.60298538,
      "memory(GiB)": 107.26,
      "step": 31030,
      "train_speed(iter/s)": 1.633052
    },
    {
      "acc": 0.65304718,
      "epoch": 0.7872907153729072,
      "grad_norm": 5.6875,
      "learning_rate": 7.1042487999817275e-06,
      "loss": 1.59884739,
      "memory(GiB)": 107.26,
      "step": 31035,
      "train_speed(iter/s)": 1.633075
    },
    {
      "acc": 0.65780239,
      "epoch": 0.7874175545408422,
      "grad_norm": 6.6875,
      "learning_rate": 7.103297515341857e-06,
      "loss": 1.59267845,
      "memory(GiB)": 107.26,
      "step": 31040,
      "train_speed(iter/s)": 1.633104
    },
    {
      "acc": 0.65838223,
      "epoch": 0.7875443937087773,
      "grad_norm": 6.90625,
      "learning_rate": 7.1023461381895685e-06,
      "loss": 1.57966022,
      "memory(GiB)": 107.26,
      "step": 31045,
      "train_speed(iter/s)": 1.633132
    },
    {
      "acc": 0.65857053,
      "epoch": 0.7876712328767124,
      "grad_norm": 5.5,
      "learning_rate": 7.1013946685667125e-06,
      "loss": 1.62191448,
      "memory(GiB)": 107.26,
      "step": 31050,
      "train_speed(iter/s)": 1.633162
    },
    {
      "acc": 0.66186371,
      "epoch": 0.7877980720446474,
      "grad_norm": 6.28125,
      "learning_rate": 7.100443106515135e-06,
      "loss": 1.5466877,
      "memory(GiB)": 107.26,
      "step": 31055,
      "train_speed(iter/s)": 1.633189
    },
    {
      "acc": 0.67042389,
      "epoch": 0.7879249112125825,
      "grad_norm": 6.71875,
      "learning_rate": 7.099491452076693e-06,
      "loss": 1.57204342,
      "memory(GiB)": 107.26,
      "step": 31060,
      "train_speed(iter/s)": 1.633219
    },
    {
      "acc": 0.65227017,
      "epoch": 0.7880517503805176,
      "grad_norm": 8.5625,
      "learning_rate": 7.098539705293242e-06,
      "loss": 1.62192554,
      "memory(GiB)": 107.26,
      "step": 31065,
      "train_speed(iter/s)": 1.633249
    },
    {
      "acc": 0.64737072,
      "epoch": 0.7881785895484525,
      "grad_norm": 6.8125,
      "learning_rate": 7.097587866206647e-06,
      "loss": 1.60865879,
      "memory(GiB)": 107.26,
      "step": 31070,
      "train_speed(iter/s)": 1.633279
    },
    {
      "acc": 0.66815062,
      "epoch": 0.7883054287163876,
      "grad_norm": 6.59375,
      "learning_rate": 7.096635934858772e-06,
      "loss": 1.54231491,
      "memory(GiB)": 107.26,
      "step": 31075,
      "train_speed(iter/s)": 1.633309
    },
    {
      "acc": 0.65892611,
      "epoch": 0.7884322678843226,
      "grad_norm": 5.84375,
      "learning_rate": 7.095683911291488e-06,
      "loss": 1.62408829,
      "memory(GiB)": 107.26,
      "step": 31080,
      "train_speed(iter/s)": 1.633338
    },
    {
      "acc": 0.65147743,
      "epoch": 0.7885591070522577,
      "grad_norm": 6.65625,
      "learning_rate": 7.0947317955466686e-06,
      "loss": 1.70138359,
      "memory(GiB)": 107.26,
      "step": 31085,
      "train_speed(iter/s)": 1.633368
    },
    {
      "acc": 0.65220976,
      "epoch": 0.7886859462201928,
      "grad_norm": 5.5625,
      "learning_rate": 7.093779587666193e-06,
      "loss": 1.62732544,
      "memory(GiB)": 107.26,
      "step": 31090,
      "train_speed(iter/s)": 1.633398
    },
    {
      "acc": 0.6654089,
      "epoch": 0.7888127853881278,
      "grad_norm": 5.34375,
      "learning_rate": 7.092827287691943e-06,
      "loss": 1.55268145,
      "memory(GiB)": 107.26,
      "step": 31095,
      "train_speed(iter/s)": 1.633425
    },
    {
      "acc": 0.66052661,
      "epoch": 0.7889396245560629,
      "grad_norm": 5.59375,
      "learning_rate": 7.091874895665806e-06,
      "loss": 1.58308468,
      "memory(GiB)": 107.26,
      "step": 31100,
      "train_speed(iter/s)": 1.633454
    },
    {
      "acc": 0.65798273,
      "epoch": 0.789066463723998,
      "grad_norm": 7.34375,
      "learning_rate": 7.09092241162967e-06,
      "loss": 1.58995972,
      "memory(GiB)": 107.26,
      "step": 31105,
      "train_speed(iter/s)": 1.633483
    },
    {
      "acc": 0.66508694,
      "epoch": 0.789193302891933,
      "grad_norm": 5.6875,
      "learning_rate": 7.089969835625432e-06,
      "loss": 1.60591736,
      "memory(GiB)": 107.26,
      "step": 31110,
      "train_speed(iter/s)": 1.633512
    },
    {
      "acc": 0.64632173,
      "epoch": 0.7893201420598681,
      "grad_norm": 5.8125,
      "learning_rate": 7.089017167694988e-06,
      "loss": 1.65225315,
      "memory(GiB)": 107.26,
      "step": 31115,
      "train_speed(iter/s)": 1.633542
    },
    {
      "acc": 0.64636788,
      "epoch": 0.7894469812278031,
      "grad_norm": 6.125,
      "learning_rate": 7.088064407880244e-06,
      "loss": 1.68166046,
      "memory(GiB)": 107.26,
      "step": 31120,
      "train_speed(iter/s)": 1.633571
    },
    {
      "acc": 0.65687761,
      "epoch": 0.7895738203957382,
      "grad_norm": 5.6875,
      "learning_rate": 7.087111556223103e-06,
      "loss": 1.60457573,
      "memory(GiB)": 107.26,
      "step": 31125,
      "train_speed(iter/s)": 1.6336
    },
    {
      "acc": 0.65077486,
      "epoch": 0.7897006595636733,
      "grad_norm": 6.15625,
      "learning_rate": 7.08615861276548e-06,
      "loss": 1.65730438,
      "memory(GiB)": 107.26,
      "step": 31130,
      "train_speed(iter/s)": 1.63363
    },
    {
      "acc": 0.64588904,
      "epoch": 0.7898274987316083,
      "grad_norm": 5.6875,
      "learning_rate": 7.085205577549285e-06,
      "loss": 1.61276989,
      "memory(GiB)": 107.26,
      "step": 31135,
      "train_speed(iter/s)": 1.633658
    },
    {
      "acc": 0.6609386,
      "epoch": 0.7899543378995434,
      "grad_norm": 4.875,
      "learning_rate": 7.08425245061644e-06,
      "loss": 1.57487993,
      "memory(GiB)": 107.26,
      "step": 31140,
      "train_speed(iter/s)": 1.633686
    },
    {
      "acc": 0.64660015,
      "epoch": 0.7900811770674785,
      "grad_norm": 5.40625,
      "learning_rate": 7.083299232008867e-06,
      "loss": 1.64494247,
      "memory(GiB)": 107.26,
      "step": 31145,
      "train_speed(iter/s)": 1.633716
    },
    {
      "acc": 0.65915112,
      "epoch": 0.7902080162354135,
      "grad_norm": 5.71875,
      "learning_rate": 7.082345921768492e-06,
      "loss": 1.55005655,
      "memory(GiB)": 107.26,
      "step": 31150,
      "train_speed(iter/s)": 1.633745
    },
    {
      "acc": 0.65915904,
      "epoch": 0.7903348554033486,
      "grad_norm": 5.875,
      "learning_rate": 7.0813925199372455e-06,
      "loss": 1.60656395,
      "memory(GiB)": 107.26,
      "step": 31155,
      "train_speed(iter/s)": 1.633774
    },
    {
      "acc": 0.63018637,
      "epoch": 0.7904616945712836,
      "grad_norm": 5.1875,
      "learning_rate": 7.080439026557065e-06,
      "loss": 1.61498108,
      "memory(GiB)": 107.26,
      "step": 31160,
      "train_speed(iter/s)": 1.633801
    },
    {
      "acc": 0.67653117,
      "epoch": 0.7905885337392187,
      "grad_norm": 5.875,
      "learning_rate": 7.079485441669887e-06,
      "loss": 1.4972065,
      "memory(GiB)": 107.26,
      "step": 31165,
      "train_speed(iter/s)": 1.633829
    },
    {
      "acc": 0.63481498,
      "epoch": 0.7907153729071538,
      "grad_norm": 6.03125,
      "learning_rate": 7.0785317653176534e-06,
      "loss": 1.63518505,
      "memory(GiB)": 107.26,
      "step": 31170,
      "train_speed(iter/s)": 1.633857
    },
    {
      "acc": 0.63802643,
      "epoch": 0.7908422120750888,
      "grad_norm": 6.0,
      "learning_rate": 7.077577997542316e-06,
      "loss": 1.67583656,
      "memory(GiB)": 107.26,
      "step": 31175,
      "train_speed(iter/s)": 1.633885
    },
    {
      "acc": 0.64503856,
      "epoch": 0.7909690512430239,
      "grad_norm": 6.09375,
      "learning_rate": 7.0766241383858195e-06,
      "loss": 1.7330164,
      "memory(GiB)": 107.26,
      "step": 31180,
      "train_speed(iter/s)": 1.633913
    },
    {
      "acc": 0.65054131,
      "epoch": 0.791095890410959,
      "grad_norm": 5.625,
      "learning_rate": 7.075670187890123e-06,
      "loss": 1.61693344,
      "memory(GiB)": 107.26,
      "step": 31185,
      "train_speed(iter/s)": 1.633942
    },
    {
      "acc": 0.66031713,
      "epoch": 0.791222729578894,
      "grad_norm": 5.5,
      "learning_rate": 7.0747161460971845e-06,
      "loss": 1.59147406,
      "memory(GiB)": 107.26,
      "step": 31190,
      "train_speed(iter/s)": 1.633973
    },
    {
      "acc": 0.64606352,
      "epoch": 0.791349568746829,
      "grad_norm": 6.25,
      "learning_rate": 7.073762013048966e-06,
      "loss": 1.67986145,
      "memory(GiB)": 107.26,
      "step": 31195,
      "train_speed(iter/s)": 1.634002
    },
    {
      "acc": 0.65435033,
      "epoch": 0.791476407914764,
      "grad_norm": 5.40625,
      "learning_rate": 7.072807788787437e-06,
      "loss": 1.58141699,
      "memory(GiB)": 107.26,
      "step": 31200,
      "train_speed(iter/s)": 1.634032
    },
    {
      "acc": 0.65349216,
      "epoch": 0.7916032470826991,
      "grad_norm": 6.5625,
      "learning_rate": 7.071853473354566e-06,
      "loss": 1.60701485,
      "memory(GiB)": 107.26,
      "step": 31205,
      "train_speed(iter/s)": 1.63406
    },
    {
      "acc": 0.66085987,
      "epoch": 0.7917300862506342,
      "grad_norm": 5.0,
      "learning_rate": 7.070899066792329e-06,
      "loss": 1.58331842,
      "memory(GiB)": 107.26,
      "step": 31210,
      "train_speed(iter/s)": 1.634088
    },
    {
      "acc": 0.66153445,
      "epoch": 0.7918569254185692,
      "grad_norm": 5.75,
      "learning_rate": 7.069944569142706e-06,
      "loss": 1.57464418,
      "memory(GiB)": 107.26,
      "step": 31215,
      "train_speed(iter/s)": 1.634118
    },
    {
      "acc": 0.6492681,
      "epoch": 0.7919837645865043,
      "grad_norm": 8.4375,
      "learning_rate": 7.068989980447679e-06,
      "loss": 1.66913986,
      "memory(GiB)": 107.26,
      "step": 31220,
      "train_speed(iter/s)": 1.634145
    },
    {
      "acc": 0.65748329,
      "epoch": 0.7921106037544394,
      "grad_norm": 6.25,
      "learning_rate": 7.068035300749237e-06,
      "loss": 1.60141544,
      "memory(GiB)": 107.26,
      "step": 31225,
      "train_speed(iter/s)": 1.634176
    },
    {
      "acc": 0.64853754,
      "epoch": 0.7922374429223744,
      "grad_norm": 4.96875,
      "learning_rate": 7.067080530089366e-06,
      "loss": 1.68210335,
      "memory(GiB)": 107.26,
      "step": 31230,
      "train_speed(iter/s)": 1.634206
    },
    {
      "acc": 0.64236946,
      "epoch": 0.7923642820903095,
      "grad_norm": 5.75,
      "learning_rate": 7.066125668510067e-06,
      "loss": 1.66154804,
      "memory(GiB)": 107.26,
      "step": 31235,
      "train_speed(iter/s)": 1.634235
    },
    {
      "acc": 0.64561367,
      "epoch": 0.7924911212582445,
      "grad_norm": 5.5625,
      "learning_rate": 7.065170716053336e-06,
      "loss": 1.63204155,
      "memory(GiB)": 107.26,
      "step": 31240,
      "train_speed(iter/s)": 1.634264
    },
    {
      "acc": 0.64884477,
      "epoch": 0.7926179604261796,
      "grad_norm": 6.46875,
      "learning_rate": 7.064215672761175e-06,
      "loss": 1.63680115,
      "memory(GiB)": 107.26,
      "step": 31245,
      "train_speed(iter/s)": 1.634288
    },
    {
      "acc": 0.66306729,
      "epoch": 0.7927447995941147,
      "grad_norm": 5.0625,
      "learning_rate": 7.063260538675594e-06,
      "loss": 1.56321697,
      "memory(GiB)": 107.26,
      "step": 31250,
      "train_speed(iter/s)": 1.634316
    },
    {
      "acc": 0.65950441,
      "epoch": 0.7928716387620497,
      "grad_norm": 7.0,
      "learning_rate": 7.062305313838601e-06,
      "loss": 1.5964489,
      "memory(GiB)": 107.26,
      "step": 31255,
      "train_speed(iter/s)": 1.634342
    },
    {
      "acc": 0.65043149,
      "epoch": 0.7929984779299848,
      "grad_norm": 6.03125,
      "learning_rate": 7.061349998292215e-06,
      "loss": 1.68801231,
      "memory(GiB)": 107.26,
      "step": 31260,
      "train_speed(iter/s)": 1.63437
    },
    {
      "acc": 0.65283613,
      "epoch": 0.7931253170979199,
      "grad_norm": 5.84375,
      "learning_rate": 7.060394592078452e-06,
      "loss": 1.61269951,
      "memory(GiB)": 107.26,
      "step": 31265,
      "train_speed(iter/s)": 1.634398
    },
    {
      "acc": 0.6575778,
      "epoch": 0.7932521562658549,
      "grad_norm": 5.78125,
      "learning_rate": 7.0594390952393365e-06,
      "loss": 1.61107903,
      "memory(GiB)": 107.26,
      "step": 31270,
      "train_speed(iter/s)": 1.634426
    },
    {
      "acc": 0.64295101,
      "epoch": 0.79337899543379,
      "grad_norm": 5.53125,
      "learning_rate": 7.058483507816894e-06,
      "loss": 1.64365768,
      "memory(GiB)": 107.26,
      "step": 31275,
      "train_speed(iter/s)": 1.634456
    },
    {
      "acc": 0.66307268,
      "epoch": 0.793505834601725,
      "grad_norm": 4.71875,
      "learning_rate": 7.057527829853157e-06,
      "loss": 1.5413166,
      "memory(GiB)": 107.26,
      "step": 31280,
      "train_speed(iter/s)": 1.634486
    },
    {
      "acc": 0.64996595,
      "epoch": 0.7936326737696601,
      "grad_norm": 7.5,
      "learning_rate": 7.056572061390159e-06,
      "loss": 1.64518414,
      "memory(GiB)": 107.26,
      "step": 31285,
      "train_speed(iter/s)": 1.634515
    },
    {
      "acc": 0.65264473,
      "epoch": 0.7937595129375952,
      "grad_norm": 4.90625,
      "learning_rate": 7.055616202469939e-06,
      "loss": 1.64386635,
      "memory(GiB)": 107.26,
      "step": 31290,
      "train_speed(iter/s)": 1.634544
    },
    {
      "acc": 0.65310621,
      "epoch": 0.7938863521055302,
      "grad_norm": 6.5625,
      "learning_rate": 7.054660253134543e-06,
      "loss": 1.60843353,
      "memory(GiB)": 107.26,
      "step": 31295,
      "train_speed(iter/s)": 1.634572
    },
    {
      "acc": 0.6521873,
      "epoch": 0.7940131912734653,
      "grad_norm": 5.59375,
      "learning_rate": 7.053704213426015e-06,
      "loss": 1.69099503,
      "memory(GiB)": 107.26,
      "step": 31300,
      "train_speed(iter/s)": 1.634603
    },
    {
      "acc": 0.65218287,
      "epoch": 0.7941400304414004,
      "grad_norm": 5.53125,
      "learning_rate": 7.052748083386406e-06,
      "loss": 1.65677357,
      "memory(GiB)": 107.26,
      "step": 31305,
      "train_speed(iter/s)": 1.634632
    },
    {
      "acc": 0.6685442,
      "epoch": 0.7942668696093353,
      "grad_norm": 6.625,
      "learning_rate": 7.051791863057772e-06,
      "loss": 1.59142437,
      "memory(GiB)": 107.26,
      "step": 31310,
      "train_speed(iter/s)": 1.63466
    },
    {
      "acc": 0.64603367,
      "epoch": 0.7943937087772704,
      "grad_norm": 6.59375,
      "learning_rate": 7.050835552482171e-06,
      "loss": 1.61281013,
      "memory(GiB)": 107.26,
      "step": 31315,
      "train_speed(iter/s)": 1.634687
    },
    {
      "acc": 0.66085606,
      "epoch": 0.7945205479452054,
      "grad_norm": 7.75,
      "learning_rate": 7.049879151701666e-06,
      "loss": 1.60690403,
      "memory(GiB)": 107.26,
      "step": 31320,
      "train_speed(iter/s)": 1.634717
    },
    {
      "acc": 0.64276886,
      "epoch": 0.7946473871131405,
      "grad_norm": 6.03125,
      "learning_rate": 7.048922660758324e-06,
      "loss": 1.60473785,
      "memory(GiB)": 107.26,
      "step": 31325,
      "train_speed(iter/s)": 1.634746
    },
    {
      "acc": 0.66217289,
      "epoch": 0.7947742262810756,
      "grad_norm": 5.71875,
      "learning_rate": 7.047966079694215e-06,
      "loss": 1.54961376,
      "memory(GiB)": 107.26,
      "step": 31330,
      "train_speed(iter/s)": 1.634774
    },
    {
      "acc": 0.6608645,
      "epoch": 0.7949010654490106,
      "grad_norm": 7.96875,
      "learning_rate": 7.047009408551414e-06,
      "loss": 1.55356884,
      "memory(GiB)": 107.26,
      "step": 31335,
      "train_speed(iter/s)": 1.634803
    },
    {
      "acc": 0.64311752,
      "epoch": 0.7950279046169457,
      "grad_norm": 5.5625,
      "learning_rate": 7.046052647372002e-06,
      "loss": 1.63000526,
      "memory(GiB)": 107.26,
      "step": 31340,
      "train_speed(iter/s)": 1.63483
    },
    {
      "acc": 0.65527148,
      "epoch": 0.7951547437848808,
      "grad_norm": 6.4375,
      "learning_rate": 7.045095796198057e-06,
      "loss": 1.63339863,
      "memory(GiB)": 107.26,
      "step": 31345,
      "train_speed(iter/s)": 1.634858
    },
    {
      "acc": 0.64365945,
      "epoch": 0.7952815829528158,
      "grad_norm": 5.625,
      "learning_rate": 7.044138855071671e-06,
      "loss": 1.75067177,
      "memory(GiB)": 107.26,
      "step": 31350,
      "train_speed(iter/s)": 1.634888
    },
    {
      "acc": 0.65477886,
      "epoch": 0.7954084221207509,
      "grad_norm": 6.1875,
      "learning_rate": 7.043181824034929e-06,
      "loss": 1.61393414,
      "memory(GiB)": 107.26,
      "step": 31355,
      "train_speed(iter/s)": 1.634917
    },
    {
      "acc": 0.66993504,
      "epoch": 0.7955352612886859,
      "grad_norm": 7.1875,
      "learning_rate": 7.042224703129929e-06,
      "loss": 1.56680183,
      "memory(GiB)": 107.26,
      "step": 31360,
      "train_speed(iter/s)": 1.634948
    },
    {
      "acc": 0.65087652,
      "epoch": 0.795662100456621,
      "grad_norm": 5.96875,
      "learning_rate": 7.0412674923987705e-06,
      "loss": 1.66979389,
      "memory(GiB)": 107.26,
      "step": 31365,
      "train_speed(iter/s)": 1.634978
    },
    {
      "acc": 0.66384845,
      "epoch": 0.7957889396245561,
      "grad_norm": 7.34375,
      "learning_rate": 7.040310191883552e-06,
      "loss": 1.55617056,
      "memory(GiB)": 107.26,
      "step": 31370,
      "train_speed(iter/s)": 1.635008
    },
    {
      "acc": 0.65551004,
      "epoch": 0.7959157787924911,
      "grad_norm": 5.40625,
      "learning_rate": 7.039352801626383e-06,
      "loss": 1.59915304,
      "memory(GiB)": 107.26,
      "step": 31375,
      "train_speed(iter/s)": 1.63504
    },
    {
      "acc": 0.67593594,
      "epoch": 0.7960426179604262,
      "grad_norm": 6.78125,
      "learning_rate": 7.0383953216693725e-06,
      "loss": 1.45270824,
      "memory(GiB)": 107.26,
      "step": 31380,
      "train_speed(iter/s)": 1.635069
    },
    {
      "acc": 0.66488671,
      "epoch": 0.7961694571283613,
      "grad_norm": 6.03125,
      "learning_rate": 7.037437752054635e-06,
      "loss": 1.60349197,
      "memory(GiB)": 107.26,
      "step": 31385,
      "train_speed(iter/s)": 1.635098
    },
    {
      "acc": 0.6614655,
      "epoch": 0.7962962962962963,
      "grad_norm": 6.40625,
      "learning_rate": 7.036480092824288e-06,
      "loss": 1.58103199,
      "memory(GiB)": 107.26,
      "step": 31390,
      "train_speed(iter/s)": 1.635128
    },
    {
      "acc": 0.65328074,
      "epoch": 0.7964231354642314,
      "grad_norm": 5.875,
      "learning_rate": 7.035522344020455e-06,
      "loss": 1.61199131,
      "memory(GiB)": 107.26,
      "step": 31395,
      "train_speed(iter/s)": 1.635158
    },
    {
      "acc": 0.64309921,
      "epoch": 0.7965499746321664,
      "grad_norm": 6.03125,
      "learning_rate": 7.034564505685262e-06,
      "loss": 1.59272738,
      "memory(GiB)": 107.26,
      "step": 31400,
      "train_speed(iter/s)": 1.635187
    },
    {
      "acc": 0.64977236,
      "epoch": 0.7966768138001015,
      "grad_norm": 6.3125,
      "learning_rate": 7.0336065778608365e-06,
      "loss": 1.60957565,
      "memory(GiB)": 107.26,
      "step": 31405,
      "train_speed(iter/s)": 1.635217
    },
    {
      "acc": 0.66490154,
      "epoch": 0.7968036529680366,
      "grad_norm": 4.6875,
      "learning_rate": 7.032648560589316e-06,
      "loss": 1.54726028,
      "memory(GiB)": 107.26,
      "step": 31410,
      "train_speed(iter/s)": 1.635248
    },
    {
      "acc": 0.66051846,
      "epoch": 0.7969304921359716,
      "grad_norm": 6.0,
      "learning_rate": 7.031690453912835e-06,
      "loss": 1.5356205,
      "memory(GiB)": 107.26,
      "step": 31415,
      "train_speed(iter/s)": 1.635277
    },
    {
      "acc": 0.65544033,
      "epoch": 0.7970573313039067,
      "grad_norm": 6.75,
      "learning_rate": 7.030732257873539e-06,
      "loss": 1.63505249,
      "memory(GiB)": 107.26,
      "step": 31420,
      "train_speed(iter/s)": 1.635306
    },
    {
      "acc": 0.65876904,
      "epoch": 0.7971841704718418,
      "grad_norm": 5.46875,
      "learning_rate": 7.02977397251357e-06,
      "loss": 1.59826202,
      "memory(GiB)": 107.26,
      "step": 31425,
      "train_speed(iter/s)": 1.635336
    },
    {
      "acc": 0.67138009,
      "epoch": 0.7973110096397767,
      "grad_norm": 4.78125,
      "learning_rate": 7.028815597875081e-06,
      "loss": 1.52501717,
      "memory(GiB)": 107.26,
      "step": 31430,
      "train_speed(iter/s)": 1.635365
    },
    {
      "acc": 0.63913412,
      "epoch": 0.7974378488077118,
      "grad_norm": 5.21875,
      "learning_rate": 7.027857134000223e-06,
      "loss": 1.69021683,
      "memory(GiB)": 107.26,
      "step": 31435,
      "train_speed(iter/s)": 1.635395
    },
    {
      "acc": 0.6709847,
      "epoch": 0.7975646879756468,
      "grad_norm": 6.75,
      "learning_rate": 7.026898580931154e-06,
      "loss": 1.6213829,
      "memory(GiB)": 107.26,
      "step": 31440,
      "train_speed(iter/s)": 1.635424
    },
    {
      "acc": 0.67155638,
      "epoch": 0.7976915271435819,
      "grad_norm": 5.65625,
      "learning_rate": 7.025939938710037e-06,
      "loss": 1.56242714,
      "memory(GiB)": 107.26,
      "step": 31445,
      "train_speed(iter/s)": 1.635454
    },
    {
      "acc": 0.63182421,
      "epoch": 0.797818366311517,
      "grad_norm": 5.5625,
      "learning_rate": 7.024981207379036e-06,
      "loss": 1.67539463,
      "memory(GiB)": 107.26,
      "step": 31450,
      "train_speed(iter/s)": 1.635481
    },
    {
      "acc": 0.63707418,
      "epoch": 0.797945205479452,
      "grad_norm": 6.90625,
      "learning_rate": 7.02402238698032e-06,
      "loss": 1.68257847,
      "memory(GiB)": 107.26,
      "step": 31455,
      "train_speed(iter/s)": 1.635509
    },
    {
      "acc": 0.64785337,
      "epoch": 0.7980720446473871,
      "grad_norm": 6.75,
      "learning_rate": 7.023063477556064e-06,
      "loss": 1.65203876,
      "memory(GiB)": 107.26,
      "step": 31460,
      "train_speed(iter/s)": 1.635537
    },
    {
      "acc": 0.65921106,
      "epoch": 0.7981988838153222,
      "grad_norm": 5.25,
      "learning_rate": 7.0221044791484424e-06,
      "loss": 1.54721508,
      "memory(GiB)": 107.26,
      "step": 31465,
      "train_speed(iter/s)": 1.635566
    },
    {
      "acc": 0.64905682,
      "epoch": 0.7983257229832572,
      "grad_norm": 5.15625,
      "learning_rate": 7.021145391799639e-06,
      "loss": 1.59451771,
      "memory(GiB)": 107.26,
      "step": 31470,
      "train_speed(iter/s)": 1.635596
    },
    {
      "acc": 0.65162015,
      "epoch": 0.7984525621511923,
      "grad_norm": 5.125,
      "learning_rate": 7.020186215551837e-06,
      "loss": 1.61367798,
      "memory(GiB)": 107.26,
      "step": 31475,
      "train_speed(iter/s)": 1.635623
    },
    {
      "acc": 0.65757213,
      "epoch": 0.7985794013191273,
      "grad_norm": 5.78125,
      "learning_rate": 7.019226950447227e-06,
      "loss": 1.59809198,
      "memory(GiB)": 107.26,
      "step": 31480,
      "train_speed(iter/s)": 1.635652
    },
    {
      "acc": 0.64026928,
      "epoch": 0.7987062404870624,
      "grad_norm": 5.84375,
      "learning_rate": 7.018267596527998e-06,
      "loss": 1.66018524,
      "memory(GiB)": 107.26,
      "step": 31485,
      "train_speed(iter/s)": 1.635682
    },
    {
      "acc": 0.67351179,
      "epoch": 0.7988330796549975,
      "grad_norm": 6.21875,
      "learning_rate": 7.017308153836352e-06,
      "loss": 1.53560753,
      "memory(GiB)": 107.26,
      "step": 31490,
      "train_speed(iter/s)": 1.635713
    },
    {
      "acc": 0.65102959,
      "epoch": 0.7989599188229325,
      "grad_norm": 5.0625,
      "learning_rate": 7.016348622414484e-06,
      "loss": 1.64828358,
      "memory(GiB)": 107.26,
      "step": 31495,
      "train_speed(iter/s)": 1.635742
    },
    {
      "acc": 0.66073585,
      "epoch": 0.7990867579908676,
      "grad_norm": 5.21875,
      "learning_rate": 7.015389002304604e-06,
      "loss": 1.60693893,
      "memory(GiB)": 107.26,
      "step": 31500,
      "train_speed(iter/s)": 1.635771
    },
    {
      "acc": 0.67532949,
      "epoch": 0.7992135971588027,
      "grad_norm": 5.65625,
      "learning_rate": 7.014429293548916e-06,
      "loss": 1.50706377,
      "memory(GiB)": 107.26,
      "step": 31505,
      "train_speed(iter/s)": 1.635801
    },
    {
      "acc": 0.6562469,
      "epoch": 0.7993404363267377,
      "grad_norm": 5.96875,
      "learning_rate": 7.013469496189633e-06,
      "loss": 1.62239532,
      "memory(GiB)": 107.26,
      "step": 31510,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.66055245,
      "epoch": 0.7994672754946728,
      "grad_norm": 5.6875,
      "learning_rate": 7.012509610268974e-06,
      "loss": 1.49975996,
      "memory(GiB)": 107.26,
      "step": 31515,
      "train_speed(iter/s)": 1.635859
    },
    {
      "acc": 0.65195799,
      "epoch": 0.7995941146626078,
      "grad_norm": 6.34375,
      "learning_rate": 7.011549635829156e-06,
      "loss": 1.60614052,
      "memory(GiB)": 107.26,
      "step": 31520,
      "train_speed(iter/s)": 1.635889
    },
    {
      "acc": 0.6493,
      "epoch": 0.7997209538305429,
      "grad_norm": 5.90625,
      "learning_rate": 7.010589572912404e-06,
      "loss": 1.59612722,
      "memory(GiB)": 107.26,
      "step": 31525,
      "train_speed(iter/s)": 1.635919
    },
    {
      "acc": 0.64656768,
      "epoch": 0.799847792998478,
      "grad_norm": 5.21875,
      "learning_rate": 7.009629421560946e-06,
      "loss": 1.63737621,
      "memory(GiB)": 107.26,
      "step": 31530,
      "train_speed(iter/s)": 1.635947
    },
    {
      "acc": 0.65564647,
      "epoch": 0.799974632166413,
      "grad_norm": 5.59375,
      "learning_rate": 7.008669181817015e-06,
      "loss": 1.63288383,
      "memory(GiB)": 107.26,
      "step": 31535,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.65118885,
      "epoch": 0.8001014713343481,
      "grad_norm": 5.34375,
      "learning_rate": 7.007708853722844e-06,
      "loss": 1.59557819,
      "memory(GiB)": 107.26,
      "step": 31540,
      "train_speed(iter/s)": 1.636005
    },
    {
      "acc": 0.64227943,
      "epoch": 0.8002283105022832,
      "grad_norm": 5.9375,
      "learning_rate": 7.006748437320674e-06,
      "loss": 1.74523048,
      "memory(GiB)": 107.26,
      "step": 31545,
      "train_speed(iter/s)": 1.636035
    },
    {
      "acc": 0.65036449,
      "epoch": 0.8003551496702181,
      "grad_norm": 5.0625,
      "learning_rate": 7.005787932652749e-06,
      "loss": 1.57965479,
      "memory(GiB)": 107.26,
      "step": 31550,
      "train_speed(iter/s)": 1.636064
    },
    {
      "acc": 0.64340105,
      "epoch": 0.8004819888381532,
      "grad_norm": 5.5625,
      "learning_rate": 7.0048273397613145e-06,
      "loss": 1.69293709,
      "memory(GiB)": 107.26,
      "step": 31555,
      "train_speed(iter/s)": 1.636092
    },
    {
      "acc": 0.65830016,
      "epoch": 0.8006088280060882,
      "grad_norm": 5.65625,
      "learning_rate": 7.003866658688624e-06,
      "loss": 1.59785042,
      "memory(GiB)": 107.26,
      "step": 31560,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.65423365,
      "epoch": 0.8007356671740233,
      "grad_norm": 6.09375,
      "learning_rate": 7.0029058894769295e-06,
      "loss": 1.65084858,
      "memory(GiB)": 107.26,
      "step": 31565,
      "train_speed(iter/s)": 1.636151
    },
    {
      "acc": 0.64913125,
      "epoch": 0.8008625063419584,
      "grad_norm": 5.5,
      "learning_rate": 7.001945032168493e-06,
      "loss": 1.6263134,
      "memory(GiB)": 107.26,
      "step": 31570,
      "train_speed(iter/s)": 1.636181
    },
    {
      "acc": 0.65347681,
      "epoch": 0.8009893455098934,
      "grad_norm": 6.0,
      "learning_rate": 7.000984086805575e-06,
      "loss": 1.58074932,
      "memory(GiB)": 107.26,
      "step": 31575,
      "train_speed(iter/s)": 1.63621
    },
    {
      "acc": 0.65131097,
      "epoch": 0.8011161846778285,
      "grad_norm": 5.59375,
      "learning_rate": 7.000023053430444e-06,
      "loss": 1.60889568,
      "memory(GiB)": 107.26,
      "step": 31580,
      "train_speed(iter/s)": 1.636238
    },
    {
      "acc": 0.64893665,
      "epoch": 0.8012430238457636,
      "grad_norm": 7.0,
      "learning_rate": 6.999061932085369e-06,
      "loss": 1.6766613,
      "memory(GiB)": 107.26,
      "step": 31585,
      "train_speed(iter/s)": 1.636266
    },
    {
      "acc": 0.65664902,
      "epoch": 0.8013698630136986,
      "grad_norm": 5.40625,
      "learning_rate": 6.9981007228126255e-06,
      "loss": 1.54973431,
      "memory(GiB)": 107.26,
      "step": 31590,
      "train_speed(iter/s)": 1.636295
    },
    {
      "acc": 0.63851833,
      "epoch": 0.8014967021816337,
      "grad_norm": 5.90625,
      "learning_rate": 6.997139425654491e-06,
      "loss": 1.65576363,
      "memory(GiB)": 107.26,
      "step": 31595,
      "train_speed(iter/s)": 1.636323
    },
    {
      "acc": 0.67258167,
      "epoch": 0.8016235413495687,
      "grad_norm": 6.59375,
      "learning_rate": 6.996178040653248e-06,
      "loss": 1.50720177,
      "memory(GiB)": 107.26,
      "step": 31600,
      "train_speed(iter/s)": 1.63635
    },
    {
      "acc": 0.67264385,
      "epoch": 0.8017503805175038,
      "grad_norm": 5.875,
      "learning_rate": 6.995216567851183e-06,
      "loss": 1.55890961,
      "memory(GiB)": 107.26,
      "step": 31605,
      "train_speed(iter/s)": 1.636377
    },
    {
      "acc": 0.63763704,
      "epoch": 0.8018772196854389,
      "grad_norm": 6.6875,
      "learning_rate": 6.994255007290585e-06,
      "loss": 1.64323063,
      "memory(GiB)": 107.26,
      "step": 31610,
      "train_speed(iter/s)": 1.636404
    },
    {
      "acc": 0.66405573,
      "epoch": 0.8020040588533739,
      "grad_norm": 5.28125,
      "learning_rate": 6.993293359013747e-06,
      "loss": 1.6096077,
      "memory(GiB)": 107.26,
      "step": 31615,
      "train_speed(iter/s)": 1.636432
    },
    {
      "acc": 0.63168311,
      "epoch": 0.802130898021309,
      "grad_norm": 5.1875,
      "learning_rate": 6.992331623062969e-06,
      "loss": 1.67222633,
      "memory(GiB)": 107.26,
      "step": 31620,
      "train_speed(iter/s)": 1.636458
    },
    {
      "acc": 0.6563798,
      "epoch": 0.8022577371892441,
      "grad_norm": 5.5625,
      "learning_rate": 6.9913697994805505e-06,
      "loss": 1.61397095,
      "memory(GiB)": 107.26,
      "step": 31625,
      "train_speed(iter/s)": 1.636486
    },
    {
      "acc": 0.65471926,
      "epoch": 0.8023845763571791,
      "grad_norm": 6.625,
      "learning_rate": 6.990407888308799e-06,
      "loss": 1.57690697,
      "memory(GiB)": 107.26,
      "step": 31630,
      "train_speed(iter/s)": 1.636512
    },
    {
      "acc": 0.6477561,
      "epoch": 0.8025114155251142,
      "grad_norm": 4.78125,
      "learning_rate": 6.98944588959002e-06,
      "loss": 1.6375412,
      "memory(GiB)": 107.26,
      "step": 31635,
      "train_speed(iter/s)": 1.636539
    },
    {
      "acc": 0.6483861,
      "epoch": 0.8026382546930492,
      "grad_norm": 6.25,
      "learning_rate": 6.9884838033665305e-06,
      "loss": 1.5951931,
      "memory(GiB)": 107.26,
      "step": 31640,
      "train_speed(iter/s)": 1.636566
    },
    {
      "acc": 0.63996024,
      "epoch": 0.8027650938609843,
      "grad_norm": 5.0,
      "learning_rate": 6.987521629680643e-06,
      "loss": 1.66841087,
      "memory(GiB)": 107.26,
      "step": 31645,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.64413018,
      "epoch": 0.8028919330289194,
      "grad_norm": 5.21875,
      "learning_rate": 6.9865593685746815e-06,
      "loss": 1.6708643,
      "memory(GiB)": 107.26,
      "step": 31650,
      "train_speed(iter/s)": 1.63662
    },
    {
      "acc": 0.65133648,
      "epoch": 0.8030187721968544,
      "grad_norm": 5.71875,
      "learning_rate": 6.98559702009097e-06,
      "loss": 1.70497379,
      "memory(GiB)": 107.26,
      "step": 31655,
      "train_speed(iter/s)": 1.636648
    },
    {
      "acc": 0.65791612,
      "epoch": 0.8031456113647895,
      "grad_norm": 7.625,
      "learning_rate": 6.984634584271836e-06,
      "loss": 1.64649849,
      "memory(GiB)": 107.26,
      "step": 31660,
      "train_speed(iter/s)": 1.636677
    },
    {
      "acc": 0.65219336,
      "epoch": 0.8032724505327246,
      "grad_norm": 5.40625,
      "learning_rate": 6.983672061159612e-06,
      "loss": 1.60834274,
      "memory(GiB)": 107.26,
      "step": 31665,
      "train_speed(iter/s)": 1.636706
    },
    {
      "acc": 0.65112314,
      "epoch": 0.8033992897006595,
      "grad_norm": 6.40625,
      "learning_rate": 6.982709450796636e-06,
      "loss": 1.61001701,
      "memory(GiB)": 107.26,
      "step": 31670,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.63708506,
      "epoch": 0.8035261288685946,
      "grad_norm": 5.84375,
      "learning_rate": 6.981746753225245e-06,
      "loss": 1.70016632,
      "memory(GiB)": 107.26,
      "step": 31675,
      "train_speed(iter/s)": 1.636761
    },
    {
      "acc": 0.65520887,
      "epoch": 0.8036529680365296,
      "grad_norm": 6.53125,
      "learning_rate": 6.980783968487783e-06,
      "loss": 1.54497299,
      "memory(GiB)": 107.26,
      "step": 31680,
      "train_speed(iter/s)": 1.636788
    },
    {
      "acc": 0.63458281,
      "epoch": 0.8037798072044647,
      "grad_norm": 10.6875,
      "learning_rate": 6.9798210966266e-06,
      "loss": 1.68705482,
      "memory(GiB)": 107.26,
      "step": 31685,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.64320369,
      "epoch": 0.8039066463723998,
      "grad_norm": 5.1875,
      "learning_rate": 6.9788581376840455e-06,
      "loss": 1.60026417,
      "memory(GiB)": 107.26,
      "step": 31690,
      "train_speed(iter/s)": 1.63684
    },
    {
      "acc": 0.65845203,
      "epoch": 0.8040334855403348,
      "grad_norm": 6.15625,
      "learning_rate": 6.977895091702474e-06,
      "loss": 1.5848032,
      "memory(GiB)": 107.26,
      "step": 31695,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.66474333,
      "epoch": 0.8041603247082699,
      "grad_norm": 5.40625,
      "learning_rate": 6.976931958724248e-06,
      "loss": 1.53534956,
      "memory(GiB)": 107.26,
      "step": 31700,
      "train_speed(iter/s)": 1.636895
    },
    {
      "acc": 0.64599733,
      "epoch": 0.804287163876205,
      "grad_norm": 5.53125,
      "learning_rate": 6.975968738791726e-06,
      "loss": 1.63694839,
      "memory(GiB)": 107.26,
      "step": 31705,
      "train_speed(iter/s)": 1.636924
    },
    {
      "acc": 0.64153309,
      "epoch": 0.80441400304414,
      "grad_norm": 5.78125,
      "learning_rate": 6.9750054319472785e-06,
      "loss": 1.70781059,
      "memory(GiB)": 107.26,
      "step": 31710,
      "train_speed(iter/s)": 1.63695
    },
    {
      "acc": 0.66971951,
      "epoch": 0.8045408422120751,
      "grad_norm": 5.4375,
      "learning_rate": 6.974042038233272e-06,
      "loss": 1.59744244,
      "memory(GiB)": 107.26,
      "step": 31715,
      "train_speed(iter/s)": 1.636978
    },
    {
      "acc": 0.63684053,
      "epoch": 0.8046676813800101,
      "grad_norm": 5.34375,
      "learning_rate": 6.9730785576920855e-06,
      "loss": 1.73447933,
      "memory(GiB)": 107.26,
      "step": 31720,
      "train_speed(iter/s)": 1.637005
    },
    {
      "acc": 0.66701121,
      "epoch": 0.8047945205479452,
      "grad_norm": 6.625,
      "learning_rate": 6.972114990366094e-06,
      "loss": 1.55795202,
      "memory(GiB)": 107.26,
      "step": 31725,
      "train_speed(iter/s)": 1.637032
    },
    {
      "acc": 0.62969141,
      "epoch": 0.8049213597158803,
      "grad_norm": 6.71875,
      "learning_rate": 6.97115133629768e-06,
      "loss": 1.59132013,
      "memory(GiB)": 107.26,
      "step": 31730,
      "train_speed(iter/s)": 1.63706
    },
    {
      "acc": 0.66779776,
      "epoch": 0.8050481988838153,
      "grad_norm": 5.25,
      "learning_rate": 6.970187595529229e-06,
      "loss": 1.5646533,
      "memory(GiB)": 107.26,
      "step": 31735,
      "train_speed(iter/s)": 1.637086
    },
    {
      "acc": 0.65147762,
      "epoch": 0.8051750380517504,
      "grad_norm": 5.78125,
      "learning_rate": 6.969223768103133e-06,
      "loss": 1.6891964,
      "memory(GiB)": 107.26,
      "step": 31740,
      "train_speed(iter/s)": 1.637113
    },
    {
      "acc": 0.6507627,
      "epoch": 0.8053018772196855,
      "grad_norm": 5.28125,
      "learning_rate": 6.968259854061783e-06,
      "loss": 1.59935303,
      "memory(GiB)": 107.26,
      "step": 31745,
      "train_speed(iter/s)": 1.637139
    },
    {
      "acc": 0.65812893,
      "epoch": 0.8054287163876205,
      "grad_norm": 5.25,
      "learning_rate": 6.967295853447578e-06,
      "loss": 1.60760384,
      "memory(GiB)": 107.26,
      "step": 31750,
      "train_speed(iter/s)": 1.637166
    },
    {
      "acc": 0.65970874,
      "epoch": 0.8055555555555556,
      "grad_norm": 5.90625,
      "learning_rate": 6.966331766302916e-06,
      "loss": 1.63058414,
      "memory(GiB)": 107.26,
      "step": 31755,
      "train_speed(iter/s)": 1.637195
    },
    {
      "acc": 0.65536618,
      "epoch": 0.8056823947234906,
      "grad_norm": 5.3125,
      "learning_rate": 6.965367592670206e-06,
      "loss": 1.56879215,
      "memory(GiB)": 107.26,
      "step": 31760,
      "train_speed(iter/s)": 1.637224
    },
    {
      "acc": 0.66208749,
      "epoch": 0.8058092338914257,
      "grad_norm": 5.9375,
      "learning_rate": 6.964403332591854e-06,
      "loss": 1.68444309,
      "memory(GiB)": 107.26,
      "step": 31765,
      "train_speed(iter/s)": 1.637251
    },
    {
      "acc": 0.65442572,
      "epoch": 0.8059360730593608,
      "grad_norm": 5.90625,
      "learning_rate": 6.963438986110272e-06,
      "loss": 1.60106144,
      "memory(GiB)": 107.26,
      "step": 31770,
      "train_speed(iter/s)": 1.637279
    },
    {
      "acc": 0.67014709,
      "epoch": 0.8060629122272958,
      "grad_norm": 6.09375,
      "learning_rate": 6.962474553267877e-06,
      "loss": 1.60699959,
      "memory(GiB)": 107.26,
      "step": 31775,
      "train_speed(iter/s)": 1.637304
    },
    {
      "acc": 0.63917484,
      "epoch": 0.8061897513952309,
      "grad_norm": 5.53125,
      "learning_rate": 6.96151003410709e-06,
      "loss": 1.67974262,
      "memory(GiB)": 107.26,
      "step": 31780,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.6571743,
      "epoch": 0.806316590563166,
      "grad_norm": 5.0,
      "learning_rate": 6.960545428670333e-06,
      "loss": 1.64713554,
      "memory(GiB)": 107.26,
      "step": 31785,
      "train_speed(iter/s)": 1.637359
    },
    {
      "acc": 0.66791286,
      "epoch": 0.806443429731101,
      "grad_norm": 8.1875,
      "learning_rate": 6.959580737000038e-06,
      "loss": 1.57795753,
      "memory(GiB)": 107.26,
      "step": 31790,
      "train_speed(iter/s)": 1.637388
    },
    {
      "acc": 0.66564775,
      "epoch": 0.806570268899036,
      "grad_norm": 5.09375,
      "learning_rate": 6.95861595913863e-06,
      "loss": 1.52560682,
      "memory(GiB)": 107.26,
      "step": 31795,
      "train_speed(iter/s)": 1.637415
    },
    {
      "acc": 0.6771359,
      "epoch": 0.806697108066971,
      "grad_norm": 4.78125,
      "learning_rate": 6.95765109512855e-06,
      "loss": 1.50610638,
      "memory(GiB)": 107.26,
      "step": 31800,
      "train_speed(iter/s)": 1.637442
    },
    {
      "acc": 0.65043592,
      "epoch": 0.8068239472349061,
      "grad_norm": 5.78125,
      "learning_rate": 6.956686145012233e-06,
      "loss": 1.58894205,
      "memory(GiB)": 107.26,
      "step": 31805,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.64786196,
      "epoch": 0.8069507864028412,
      "grad_norm": 6.875,
      "learning_rate": 6.955721108832124e-06,
      "loss": 1.61288109,
      "memory(GiB)": 107.26,
      "step": 31810,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.65786214,
      "epoch": 0.8070776255707762,
      "grad_norm": 5.96875,
      "learning_rate": 6.9547559866306695e-06,
      "loss": 1.56943264,
      "memory(GiB)": 107.26,
      "step": 31815,
      "train_speed(iter/s)": 1.637525
    },
    {
      "acc": 0.65914631,
      "epoch": 0.8072044647387113,
      "grad_norm": 6.375,
      "learning_rate": 6.953790778450318e-06,
      "loss": 1.60140972,
      "memory(GiB)": 107.26,
      "step": 31820,
      "train_speed(iter/s)": 1.637551
    },
    {
      "acc": 0.65495863,
      "epoch": 0.8073313039066464,
      "grad_norm": 6.53125,
      "learning_rate": 6.9528254843335254e-06,
      "loss": 1.60486603,
      "memory(GiB)": 107.26,
      "step": 31825,
      "train_speed(iter/s)": 1.63758
    },
    {
      "acc": 0.66042252,
      "epoch": 0.8074581430745814,
      "grad_norm": 5.625,
      "learning_rate": 6.95186010432275e-06,
      "loss": 1.54157944,
      "memory(GiB)": 107.26,
      "step": 31830,
      "train_speed(iter/s)": 1.637607
    },
    {
      "acc": 0.65015907,
      "epoch": 0.8075849822425165,
      "grad_norm": 5.5,
      "learning_rate": 6.950894638460452e-06,
      "loss": 1.63135757,
      "memory(GiB)": 107.26,
      "step": 31835,
      "train_speed(iter/s)": 1.637632
    },
    {
      "acc": 0.65165834,
      "epoch": 0.8077118214104515,
      "grad_norm": 5.25,
      "learning_rate": 6.949929086789098e-06,
      "loss": 1.62540207,
      "memory(GiB)": 107.26,
      "step": 31840,
      "train_speed(iter/s)": 1.637659
    },
    {
      "acc": 0.64335504,
      "epoch": 0.8078386605783866,
      "grad_norm": 5.6875,
      "learning_rate": 6.948963449351156e-06,
      "loss": 1.68301506,
      "memory(GiB)": 107.26,
      "step": 31845,
      "train_speed(iter/s)": 1.637687
    },
    {
      "acc": 0.64293494,
      "epoch": 0.8079654997463217,
      "grad_norm": 6.125,
      "learning_rate": 6.947997726189102e-06,
      "loss": 1.6709095,
      "memory(GiB)": 107.26,
      "step": 31850,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.65304346,
      "epoch": 0.8080923389142567,
      "grad_norm": 7.875,
      "learning_rate": 6.947031917345409e-06,
      "loss": 1.65402298,
      "memory(GiB)": 107.26,
      "step": 31855,
      "train_speed(iter/s)": 1.63774
    },
    {
      "acc": 0.64841862,
      "epoch": 0.8082191780821918,
      "grad_norm": 6.0625,
      "learning_rate": 6.946066022862561e-06,
      "loss": 1.65863914,
      "memory(GiB)": 107.26,
      "step": 31860,
      "train_speed(iter/s)": 1.637767
    },
    {
      "acc": 0.65058589,
      "epoch": 0.8083460172501269,
      "grad_norm": 5.84375,
      "learning_rate": 6.945100042783039e-06,
      "loss": 1.64971886,
      "memory(GiB)": 107.26,
      "step": 31865,
      "train_speed(iter/s)": 1.637791
    },
    {
      "acc": 0.64818745,
      "epoch": 0.8084728564180619,
      "grad_norm": 5.6875,
      "learning_rate": 6.9441339771493345e-06,
      "loss": 1.56851358,
      "memory(GiB)": 107.26,
      "step": 31870,
      "train_speed(iter/s)": 1.637819
    },
    {
      "acc": 0.66528578,
      "epoch": 0.808599695585997,
      "grad_norm": 5.5,
      "learning_rate": 6.943167826003937e-06,
      "loss": 1.52801495,
      "memory(GiB)": 107.26,
      "step": 31875,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.66449409,
      "epoch": 0.808726534753932,
      "grad_norm": 5.09375,
      "learning_rate": 6.942201589389344e-06,
      "loss": 1.590306,
      "memory(GiB)": 107.26,
      "step": 31880,
      "train_speed(iter/s)": 1.637873
    },
    {
      "acc": 0.66747532,
      "epoch": 0.8088533739218671,
      "grad_norm": 4.8125,
      "learning_rate": 6.9412352673480525e-06,
      "loss": 1.57964287,
      "memory(GiB)": 107.26,
      "step": 31885,
      "train_speed(iter/s)": 1.637898
    },
    {
      "acc": 0.63754005,
      "epoch": 0.8089802130898022,
      "grad_norm": 4.78125,
      "learning_rate": 6.940268859922566e-06,
      "loss": 1.66158218,
      "memory(GiB)": 107.26,
      "step": 31890,
      "train_speed(iter/s)": 1.637923
    },
    {
      "acc": 0.6645937,
      "epoch": 0.8091070522577372,
      "grad_norm": 5.71875,
      "learning_rate": 6.939302367155394e-06,
      "loss": 1.58089075,
      "memory(GiB)": 107.26,
      "step": 31895,
      "train_speed(iter/s)": 1.63795
    },
    {
      "acc": 0.63663797,
      "epoch": 0.8092338914256723,
      "grad_norm": 6.78125,
      "learning_rate": 6.9383357890890454e-06,
      "loss": 1.75767899,
      "memory(GiB)": 107.26,
      "step": 31900,
      "train_speed(iter/s)": 1.637979
    },
    {
      "acc": 0.6385026,
      "epoch": 0.8093607305936074,
      "grad_norm": 6.40625,
      "learning_rate": 6.937369125766033e-06,
      "loss": 1.65668526,
      "memory(GiB)": 107.26,
      "step": 31905,
      "train_speed(iter/s)": 1.638007
    },
    {
      "acc": 0.64706106,
      "epoch": 0.8094875697615423,
      "grad_norm": 6.90625,
      "learning_rate": 6.936402377228879e-06,
      "loss": 1.66105251,
      "memory(GiB)": 107.26,
      "step": 31910,
      "train_speed(iter/s)": 1.638034
    },
    {
      "acc": 0.65470076,
      "epoch": 0.8096144089294774,
      "grad_norm": 5.1875,
      "learning_rate": 6.9354355435201015e-06,
      "loss": 1.54109869,
      "memory(GiB)": 107.26,
      "step": 31915,
      "train_speed(iter/s)": 1.638061
    },
    {
      "acc": 0.64888988,
      "epoch": 0.8097412480974124,
      "grad_norm": 5.78125,
      "learning_rate": 6.934468624682229e-06,
      "loss": 1.64780807,
      "memory(GiB)": 107.26,
      "step": 31920,
      "train_speed(iter/s)": 1.638087
    },
    {
      "acc": 0.65892801,
      "epoch": 0.8098680872653475,
      "grad_norm": 6.5,
      "learning_rate": 6.933501620757789e-06,
      "loss": 1.56067877,
      "memory(GiB)": 107.26,
      "step": 31925,
      "train_speed(iter/s)": 1.638114
    },
    {
      "acc": 0.66509089,
      "epoch": 0.8099949264332826,
      "grad_norm": 5.53125,
      "learning_rate": 6.932534531789317e-06,
      "loss": 1.547925,
      "memory(GiB)": 107.26,
      "step": 31930,
      "train_speed(iter/s)": 1.638137
    },
    {
      "acc": 0.65340157,
      "epoch": 0.8101217656012176,
      "grad_norm": 5.46875,
      "learning_rate": 6.931567357819344e-06,
      "loss": 1.59515181,
      "memory(GiB)": 107.26,
      "step": 31935,
      "train_speed(iter/s)": 1.638164
    },
    {
      "acc": 0.63720765,
      "epoch": 0.8102486047691527,
      "grad_norm": 5.1875,
      "learning_rate": 6.930600098890419e-06,
      "loss": 1.61746025,
      "memory(GiB)": 107.26,
      "step": 31940,
      "train_speed(iter/s)": 1.638189
    },
    {
      "acc": 0.64260411,
      "epoch": 0.8103754439370878,
      "grad_norm": 5.28125,
      "learning_rate": 6.929632755045079e-06,
      "loss": 1.63631001,
      "memory(GiB)": 107.26,
      "step": 31945,
      "train_speed(iter/s)": 1.638216
    },
    {
      "acc": 0.66141033,
      "epoch": 0.8105022831050228,
      "grad_norm": 8.3125,
      "learning_rate": 6.9286653263258765e-06,
      "loss": 1.63917561,
      "memory(GiB)": 107.26,
      "step": 31950,
      "train_speed(iter/s)": 1.638245
    },
    {
      "acc": 0.63730392,
      "epoch": 0.8106291222729579,
      "grad_norm": 6.03125,
      "learning_rate": 6.927697812775363e-06,
      "loss": 1.68924656,
      "memory(GiB)": 107.26,
      "step": 31955,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.64088702,
      "epoch": 0.8107559614408929,
      "grad_norm": 6.8125,
      "learning_rate": 6.926730214436091e-06,
      "loss": 1.65115223,
      "memory(GiB)": 107.26,
      "step": 31960,
      "train_speed(iter/s)": 1.638296
    },
    {
      "acc": 0.66395721,
      "epoch": 0.810882800608828,
      "grad_norm": 6.96875,
      "learning_rate": 6.925762531350624e-06,
      "loss": 1.58305988,
      "memory(GiB)": 107.26,
      "step": 31965,
      "train_speed(iter/s)": 1.638323
    },
    {
      "acc": 0.64590797,
      "epoch": 0.8110096397767631,
      "grad_norm": 7.375,
      "learning_rate": 6.924794763561522e-06,
      "loss": 1.65210724,
      "memory(GiB)": 107.26,
      "step": 31970,
      "train_speed(iter/s)": 1.638351
    },
    {
      "acc": 0.65053396,
      "epoch": 0.8111364789446981,
      "grad_norm": 5.65625,
      "learning_rate": 6.923826911111353e-06,
      "loss": 1.60181446,
      "memory(GiB)": 107.26,
      "step": 31975,
      "train_speed(iter/s)": 1.638378
    },
    {
      "acc": 0.65695963,
      "epoch": 0.8112633181126332,
      "grad_norm": 4.875,
      "learning_rate": 6.922858974042688e-06,
      "loss": 1.58870478,
      "memory(GiB)": 107.26,
      "step": 31980,
      "train_speed(iter/s)": 1.638404
    },
    {
      "acc": 0.66785922,
      "epoch": 0.8113901572805683,
      "grad_norm": 6.21875,
      "learning_rate": 6.921890952398098e-06,
      "loss": 1.56708317,
      "memory(GiB)": 107.26,
      "step": 31985,
      "train_speed(iter/s)": 1.63843
    },
    {
      "acc": 0.65374861,
      "epoch": 0.8115169964485033,
      "grad_norm": 7.03125,
      "learning_rate": 6.920922846220166e-06,
      "loss": 1.67089214,
      "memory(GiB)": 107.26,
      "step": 31990,
      "train_speed(iter/s)": 1.638458
    },
    {
      "acc": 0.6325027,
      "epoch": 0.8116438356164384,
      "grad_norm": 6.46875,
      "learning_rate": 6.919954655551469e-06,
      "loss": 1.59934502,
      "memory(GiB)": 107.26,
      "step": 31995,
      "train_speed(iter/s)": 1.638487
    },
    {
      "acc": 0.6746294,
      "epoch": 0.8117706747843734,
      "grad_norm": 5.21875,
      "learning_rate": 6.918986380434594e-06,
      "loss": 1.5499197,
      "memory(GiB)": 107.26,
      "step": 32000,
      "train_speed(iter/s)": 1.638513
    },
    {
      "epoch": 0.8117706747843734,
      "eval_acc": 0.6457560399033532,
      "eval_loss": 1.57610285282135,
      "eval_runtime": 58.6414,
      "eval_samples_per_second": 108.626,
      "eval_steps_per_second": 27.165,
      "step": 32000
    },
    {
      "acc": 0.66393976,
      "epoch": 0.8118975139523085,
      "grad_norm": 6.96875,
      "learning_rate": 6.918018020912132e-06,
      "loss": 1.64369621,
      "memory(GiB)": 107.26,
      "step": 32005,
      "train_speed(iter/s)": 1.633282
    },
    {
      "acc": 0.66803837,
      "epoch": 0.8120243531202436,
      "grad_norm": 6.53125,
      "learning_rate": 6.917049577026673e-06,
      "loss": 1.57339373,
      "memory(GiB)": 107.26,
      "step": 32010,
      "train_speed(iter/s)": 1.633308
    },
    {
      "acc": 0.65935593,
      "epoch": 0.8121511922881786,
      "grad_norm": 4.75,
      "learning_rate": 6.916081048820815e-06,
      "loss": 1.56297636,
      "memory(GiB)": 107.26,
      "step": 32015,
      "train_speed(iter/s)": 1.633336
    },
    {
      "acc": 0.6554276,
      "epoch": 0.8122780314561137,
      "grad_norm": 5.3125,
      "learning_rate": 6.915112436337157e-06,
      "loss": 1.62615929,
      "memory(GiB)": 107.26,
      "step": 32020,
      "train_speed(iter/s)": 1.63336
    },
    {
      "acc": 0.64893494,
      "epoch": 0.8124048706240488,
      "grad_norm": 5.5625,
      "learning_rate": 6.914143739618305e-06,
      "loss": 1.65617294,
      "memory(GiB)": 107.26,
      "step": 32025,
      "train_speed(iter/s)": 1.633389
    },
    {
      "acc": 0.68204007,
      "epoch": 0.8125317097919837,
      "grad_norm": 6.3125,
      "learning_rate": 6.913174958706865e-06,
      "loss": 1.52482357,
      "memory(GiB)": 107.26,
      "step": 32030,
      "train_speed(iter/s)": 1.633414
    },
    {
      "acc": 0.65520697,
      "epoch": 0.8126585489599188,
      "grad_norm": 5.25,
      "learning_rate": 6.912206093645448e-06,
      "loss": 1.63280754,
      "memory(GiB)": 107.26,
      "step": 32035,
      "train_speed(iter/s)": 1.633443
    },
    {
      "acc": 0.66827612,
      "epoch": 0.8127853881278538,
      "grad_norm": 7.96875,
      "learning_rate": 6.91123714447667e-06,
      "loss": 1.58072968,
      "memory(GiB)": 107.26,
      "step": 32040,
      "train_speed(iter/s)": 1.633473
    },
    {
      "acc": 0.64861479,
      "epoch": 0.8129122272957889,
      "grad_norm": 5.1875,
      "learning_rate": 6.910268111243149e-06,
      "loss": 1.58176289,
      "memory(GiB)": 107.26,
      "step": 32045,
      "train_speed(iter/s)": 1.6335
    },
    {
      "acc": 0.63910809,
      "epoch": 0.813039066463724,
      "grad_norm": 6.0625,
      "learning_rate": 6.909298993987508e-06,
      "loss": 1.72066059,
      "memory(GiB)": 107.26,
      "step": 32050,
      "train_speed(iter/s)": 1.633527
    },
    {
      "acc": 0.65280857,
      "epoch": 0.813165905631659,
      "grad_norm": 5.34375,
      "learning_rate": 6.908329792752373e-06,
      "loss": 1.58054399,
      "memory(GiB)": 107.26,
      "step": 32055,
      "train_speed(iter/s)": 1.633554
    },
    {
      "acc": 0.65664806,
      "epoch": 0.8132927447995941,
      "grad_norm": 4.96875,
      "learning_rate": 6.907360507580374e-06,
      "loss": 1.62364864,
      "memory(GiB)": 107.26,
      "step": 32060,
      "train_speed(iter/s)": 1.63358
    },
    {
      "acc": 0.64661937,
      "epoch": 0.8134195839675292,
      "grad_norm": 5.1875,
      "learning_rate": 6.9063911385141425e-06,
      "loss": 1.61508522,
      "memory(GiB)": 107.26,
      "step": 32065,
      "train_speed(iter/s)": 1.633608
    },
    {
      "acc": 0.64370565,
      "epoch": 0.8135464231354642,
      "grad_norm": 5.53125,
      "learning_rate": 6.9054216855963194e-06,
      "loss": 1.64386616,
      "memory(GiB)": 107.26,
      "step": 32070,
      "train_speed(iter/s)": 1.633636
    },
    {
      "acc": 0.64428039,
      "epoch": 0.8136732623033993,
      "grad_norm": 6.5,
      "learning_rate": 6.904452148869541e-06,
      "loss": 1.63389359,
      "memory(GiB)": 107.26,
      "step": 32075,
      "train_speed(iter/s)": 1.633662
    },
    {
      "acc": 0.64763231,
      "epoch": 0.8138001014713343,
      "grad_norm": 5.1875,
      "learning_rate": 6.903482528376457e-06,
      "loss": 1.59145336,
      "memory(GiB)": 107.26,
      "step": 32080,
      "train_speed(iter/s)": 1.633689
    },
    {
      "acc": 0.65249796,
      "epoch": 0.8139269406392694,
      "grad_norm": 6.03125,
      "learning_rate": 6.902512824159711e-06,
      "loss": 1.67030869,
      "memory(GiB)": 107.26,
      "step": 32085,
      "train_speed(iter/s)": 1.633716
    },
    {
      "acc": 0.65229883,
      "epoch": 0.8140537798072045,
      "grad_norm": 5.5625,
      "learning_rate": 6.901543036261957e-06,
      "loss": 1.6234724,
      "memory(GiB)": 107.26,
      "step": 32090,
      "train_speed(iter/s)": 1.633743
    },
    {
      "acc": 0.64428339,
      "epoch": 0.8141806189751395,
      "grad_norm": 7.96875,
      "learning_rate": 6.900573164725852e-06,
      "loss": 1.67717514,
      "memory(GiB)": 107.26,
      "step": 32095,
      "train_speed(iter/s)": 1.633771
    },
    {
      "acc": 0.66641817,
      "epoch": 0.8143074581430746,
      "grad_norm": 5.96875,
      "learning_rate": 6.899603209594052e-06,
      "loss": 1.56629438,
      "memory(GiB)": 107.26,
      "step": 32100,
      "train_speed(iter/s)": 1.633795
    },
    {
      "acc": 0.66047859,
      "epoch": 0.8144342973110097,
      "grad_norm": 5.09375,
      "learning_rate": 6.898633170909224e-06,
      "loss": 1.59941158,
      "memory(GiB)": 107.26,
      "step": 32105,
      "train_speed(iter/s)": 1.633822
    },
    {
      "acc": 0.65399694,
      "epoch": 0.8145611364789447,
      "grad_norm": 6.0625,
      "learning_rate": 6.897663048714031e-06,
      "loss": 1.57719822,
      "memory(GiB)": 107.26,
      "step": 32110,
      "train_speed(iter/s)": 1.633848
    },
    {
      "acc": 0.66222644,
      "epoch": 0.8146879756468798,
      "grad_norm": 5.28125,
      "learning_rate": 6.896692843051145e-06,
      "loss": 1.60690384,
      "memory(GiB)": 107.26,
      "step": 32115,
      "train_speed(iter/s)": 1.633876
    },
    {
      "acc": 0.65599937,
      "epoch": 0.8148148148148148,
      "grad_norm": 5.9375,
      "learning_rate": 6.895722553963239e-06,
      "loss": 1.60285091,
      "memory(GiB)": 107.26,
      "step": 32120,
      "train_speed(iter/s)": 1.633903
    },
    {
      "acc": 0.65920811,
      "epoch": 0.8149416539827499,
      "grad_norm": 6.5625,
      "learning_rate": 6.8947521814929915e-06,
      "loss": 1.65305328,
      "memory(GiB)": 107.26,
      "step": 32125,
      "train_speed(iter/s)": 1.633931
    },
    {
      "acc": 0.63986645,
      "epoch": 0.815068493150685,
      "grad_norm": 7.09375,
      "learning_rate": 6.8937817256830834e-06,
      "loss": 1.73617954,
      "memory(GiB)": 107.26,
      "step": 32130,
      "train_speed(iter/s)": 1.633959
    },
    {
      "acc": 0.65625567,
      "epoch": 0.81519533231862,
      "grad_norm": 6.6875,
      "learning_rate": 6.892811186576199e-06,
      "loss": 1.62067204,
      "memory(GiB)": 107.26,
      "step": 32135,
      "train_speed(iter/s)": 1.633986
    },
    {
      "acc": 0.65273118,
      "epoch": 0.8153221714865551,
      "grad_norm": 5.65625,
      "learning_rate": 6.8918405642150295e-06,
      "loss": 1.61611557,
      "memory(GiB)": 107.26,
      "step": 32140,
      "train_speed(iter/s)": 1.634013
    },
    {
      "acc": 0.65577984,
      "epoch": 0.8154490106544902,
      "grad_norm": 5.0625,
      "learning_rate": 6.890869858642264e-06,
      "loss": 1.60321445,
      "memory(GiB)": 107.26,
      "step": 32145,
      "train_speed(iter/s)": 1.63404
    },
    {
      "acc": 0.65248051,
      "epoch": 0.8155758498224251,
      "grad_norm": 5.375,
      "learning_rate": 6.889899069900603e-06,
      "loss": 1.64560661,
      "memory(GiB)": 107.26,
      "step": 32150,
      "train_speed(iter/s)": 1.634067
    },
    {
      "acc": 0.65291662,
      "epoch": 0.8157026889903602,
      "grad_norm": 6.03125,
      "learning_rate": 6.888928198032741e-06,
      "loss": 1.5980773,
      "memory(GiB)": 107.26,
      "step": 32155,
      "train_speed(iter/s)": 1.634096
    },
    {
      "acc": 0.6604311,
      "epoch": 0.8158295281582952,
      "grad_norm": 6.78125,
      "learning_rate": 6.887957243081384e-06,
      "loss": 1.59901848,
      "memory(GiB)": 107.26,
      "step": 32160,
      "train_speed(iter/s)": 1.634125
    },
    {
      "acc": 0.65565186,
      "epoch": 0.8159563673262303,
      "grad_norm": 5.65625,
      "learning_rate": 6.886986205089237e-06,
      "loss": 1.59496717,
      "memory(GiB)": 107.26,
      "step": 32165,
      "train_speed(iter/s)": 1.634152
    },
    {
      "acc": 0.65986338,
      "epoch": 0.8160832064941654,
      "grad_norm": 6.21875,
      "learning_rate": 6.886015084099011e-06,
      "loss": 1.6038208,
      "memory(GiB)": 107.26,
      "step": 32170,
      "train_speed(iter/s)": 1.634179
    },
    {
      "acc": 0.65892019,
      "epoch": 0.8162100456621004,
      "grad_norm": 5.78125,
      "learning_rate": 6.885043880153424e-06,
      "loss": 1.6073143,
      "memory(GiB)": 107.26,
      "step": 32175,
      "train_speed(iter/s)": 1.634203
    },
    {
      "acc": 0.66693459,
      "epoch": 0.8163368848300355,
      "grad_norm": 6.0625,
      "learning_rate": 6.88407259329519e-06,
      "loss": 1.52714939,
      "memory(GiB)": 107.26,
      "step": 32180,
      "train_speed(iter/s)": 1.634232
    },
    {
      "acc": 0.66009459,
      "epoch": 0.8164637239979706,
      "grad_norm": 5.5625,
      "learning_rate": 6.883101223567031e-06,
      "loss": 1.61822433,
      "memory(GiB)": 107.26,
      "step": 32185,
      "train_speed(iter/s)": 1.634261
    },
    {
      "acc": 0.6580308,
      "epoch": 0.8165905631659056,
      "grad_norm": 4.875,
      "learning_rate": 6.882129771011674e-06,
      "loss": 1.54892139,
      "memory(GiB)": 107.26,
      "step": 32190,
      "train_speed(iter/s)": 1.634286
    },
    {
      "acc": 0.66543446,
      "epoch": 0.8167174023338407,
      "grad_norm": 6.0625,
      "learning_rate": 6.881158235671845e-06,
      "loss": 1.5171236,
      "memory(GiB)": 107.26,
      "step": 32195,
      "train_speed(iter/s)": 1.634312
    },
    {
      "acc": 0.65951328,
      "epoch": 0.8168442415017757,
      "grad_norm": 5.875,
      "learning_rate": 6.8801866175902785e-06,
      "loss": 1.57330246,
      "memory(GiB)": 107.26,
      "step": 32200,
      "train_speed(iter/s)": 1.634338
    },
    {
      "acc": 0.64907103,
      "epoch": 0.8169710806697108,
      "grad_norm": 5.34375,
      "learning_rate": 6.87921491680971e-06,
      "loss": 1.65742016,
      "memory(GiB)": 107.26,
      "step": 32205,
      "train_speed(iter/s)": 1.634364
    },
    {
      "acc": 0.66012859,
      "epoch": 0.8170979198376459,
      "grad_norm": 5.78125,
      "learning_rate": 6.878243133372882e-06,
      "loss": 1.5649374,
      "memory(GiB)": 107.26,
      "step": 32210,
      "train_speed(iter/s)": 1.634392
    },
    {
      "acc": 0.65850859,
      "epoch": 0.8172247590055809,
      "grad_norm": 5.78125,
      "learning_rate": 6.877271267322532e-06,
      "loss": 1.62557621,
      "memory(GiB)": 107.26,
      "step": 32215,
      "train_speed(iter/s)": 1.63442
    },
    {
      "acc": 0.63739662,
      "epoch": 0.817351598173516,
      "grad_norm": 4.59375,
      "learning_rate": 6.876299318701412e-06,
      "loss": 1.66764774,
      "memory(GiB)": 107.26,
      "step": 32220,
      "train_speed(iter/s)": 1.634447
    },
    {
      "acc": 0.65172548,
      "epoch": 0.8174784373414511,
      "grad_norm": 5.8125,
      "learning_rate": 6.875327287552269e-06,
      "loss": 1.61338234,
      "memory(GiB)": 107.26,
      "step": 32225,
      "train_speed(iter/s)": 1.634475
    },
    {
      "acc": 0.6541935,
      "epoch": 0.8176052765093861,
      "grad_norm": 6.0625,
      "learning_rate": 6.8743551739178615e-06,
      "loss": 1.64148178,
      "memory(GiB)": 107.26,
      "step": 32230,
      "train_speed(iter/s)": 1.634502
    },
    {
      "acc": 0.64344125,
      "epoch": 0.8177321156773212,
      "grad_norm": 5.34375,
      "learning_rate": 6.8733829778409425e-06,
      "loss": 1.64630795,
      "memory(GiB)": 107.26,
      "step": 32235,
      "train_speed(iter/s)": 1.634531
    },
    {
      "acc": 0.66356535,
      "epoch": 0.8178589548452562,
      "grad_norm": 5.75,
      "learning_rate": 6.872410699364278e-06,
      "loss": 1.46752462,
      "memory(GiB)": 107.26,
      "step": 32240,
      "train_speed(iter/s)": 1.634559
    },
    {
      "acc": 0.65554724,
      "epoch": 0.8179857940131913,
      "grad_norm": 5.1875,
      "learning_rate": 6.8714383385306305e-06,
      "loss": 1.62728748,
      "memory(GiB)": 107.26,
      "step": 32245,
      "train_speed(iter/s)": 1.634586
    },
    {
      "acc": 0.66039906,
      "epoch": 0.8181126331811264,
      "grad_norm": 5.1875,
      "learning_rate": 6.870465895382769e-06,
      "loss": 1.54872265,
      "memory(GiB)": 107.26,
      "step": 32250,
      "train_speed(iter/s)": 1.634614
    },
    {
      "acc": 0.67040577,
      "epoch": 0.8182394723490614,
      "grad_norm": 5.53125,
      "learning_rate": 6.869493369963468e-06,
      "loss": 1.56728411,
      "memory(GiB)": 107.26,
      "step": 32255,
      "train_speed(iter/s)": 1.634642
    },
    {
      "acc": 0.66036167,
      "epoch": 0.8183663115169965,
      "grad_norm": 6.71875,
      "learning_rate": 6.8685207623155e-06,
      "loss": 1.61439857,
      "memory(GiB)": 107.26,
      "step": 32260,
      "train_speed(iter/s)": 1.63467
    },
    {
      "acc": 0.64325314,
      "epoch": 0.8184931506849316,
      "grad_norm": 5.1875,
      "learning_rate": 6.867548072481649e-06,
      "loss": 1.67407379,
      "memory(GiB)": 107.26,
      "step": 32265,
      "train_speed(iter/s)": 1.634699
    },
    {
      "acc": 0.63347797,
      "epoch": 0.8186199898528665,
      "grad_norm": 5.34375,
      "learning_rate": 6.866575300504695e-06,
      "loss": 1.61705627,
      "memory(GiB)": 107.26,
      "step": 32270,
      "train_speed(iter/s)": 1.634723
    },
    {
      "acc": 0.65691743,
      "epoch": 0.8187468290208016,
      "grad_norm": 7.5,
      "learning_rate": 6.865602446427424e-06,
      "loss": 1.62348709,
      "memory(GiB)": 107.26,
      "step": 32275,
      "train_speed(iter/s)": 1.634752
    },
    {
      "acc": 0.65422549,
      "epoch": 0.8188736681887366,
      "grad_norm": 6.8125,
      "learning_rate": 6.864629510292629e-06,
      "loss": 1.62963963,
      "memory(GiB)": 107.26,
      "step": 32280,
      "train_speed(iter/s)": 1.634779
    },
    {
      "acc": 0.66408701,
      "epoch": 0.8190005073566717,
      "grad_norm": 4.84375,
      "learning_rate": 6.863656492143103e-06,
      "loss": 1.56829777,
      "memory(GiB)": 107.26,
      "step": 32285,
      "train_speed(iter/s)": 1.634803
    },
    {
      "acc": 0.64758854,
      "epoch": 0.8191273465246068,
      "grad_norm": 8.1875,
      "learning_rate": 6.862683392021644e-06,
      "loss": 1.59552746,
      "memory(GiB)": 107.26,
      "step": 32290,
      "train_speed(iter/s)": 1.634831
    },
    {
      "acc": 0.66735768,
      "epoch": 0.8192541856925418,
      "grad_norm": 6.0,
      "learning_rate": 6.861710209971052e-06,
      "loss": 1.47986202,
      "memory(GiB)": 107.26,
      "step": 32295,
      "train_speed(iter/s)": 1.634857
    },
    {
      "acc": 0.64073963,
      "epoch": 0.8193810248604769,
      "grad_norm": 5.1875,
      "learning_rate": 6.860736946034136e-06,
      "loss": 1.66423645,
      "memory(GiB)": 107.26,
      "step": 32300,
      "train_speed(iter/s)": 1.634884
    },
    {
      "acc": 0.6388639,
      "epoch": 0.819507864028412,
      "grad_norm": 6.09375,
      "learning_rate": 6.859763600253698e-06,
      "loss": 1.72613449,
      "memory(GiB)": 107.26,
      "step": 32305,
      "train_speed(iter/s)": 1.634911
    },
    {
      "acc": 0.65564804,
      "epoch": 0.819634703196347,
      "grad_norm": 5.0,
      "learning_rate": 6.858790172672556e-06,
      "loss": 1.62072411,
      "memory(GiB)": 107.26,
      "step": 32310,
      "train_speed(iter/s)": 1.634938
    },
    {
      "acc": 0.66547508,
      "epoch": 0.8197615423642821,
      "grad_norm": 5.5625,
      "learning_rate": 6.857816663333523e-06,
      "loss": 1.57943888,
      "memory(GiB)": 107.26,
      "step": 32315,
      "train_speed(iter/s)": 1.634967
    },
    {
      "acc": 0.66568441,
      "epoch": 0.8198883815322171,
      "grad_norm": 7.625,
      "learning_rate": 6.856843072279418e-06,
      "loss": 1.53723431,
      "memory(GiB)": 107.26,
      "step": 32320,
      "train_speed(iter/s)": 1.634995
    },
    {
      "acc": 0.65385957,
      "epoch": 0.8200152207001522,
      "grad_norm": 5.15625,
      "learning_rate": 6.855869399553065e-06,
      "loss": 1.6625124,
      "memory(GiB)": 107.26,
      "step": 32325,
      "train_speed(iter/s)": 1.635023
    },
    {
      "acc": 0.65744534,
      "epoch": 0.8201420598680873,
      "grad_norm": 7.125,
      "learning_rate": 6.85489564519729e-06,
      "loss": 1.59037447,
      "memory(GiB)": 107.26,
      "step": 32330,
      "train_speed(iter/s)": 1.635049
    },
    {
      "acc": 0.66599569,
      "epoch": 0.8202688990360223,
      "grad_norm": 5.28125,
      "learning_rate": 6.853921809254922e-06,
      "loss": 1.59165907,
      "memory(GiB)": 107.26,
      "step": 32335,
      "train_speed(iter/s)": 1.635076
    },
    {
      "acc": 0.63831067,
      "epoch": 0.8203957382039574,
      "grad_norm": 6.96875,
      "learning_rate": 6.852947891768796e-06,
      "loss": 1.60572338,
      "memory(GiB)": 107.26,
      "step": 32340,
      "train_speed(iter/s)": 1.635103
    },
    {
      "acc": 0.67591271,
      "epoch": 0.8205225773718925,
      "grad_norm": 6.0625,
      "learning_rate": 6.851973892781749e-06,
      "loss": 1.50954609,
      "memory(GiB)": 107.26,
      "step": 32345,
      "train_speed(iter/s)": 1.63513
    },
    {
      "acc": 0.65335698,
      "epoch": 0.8206494165398275,
      "grad_norm": 5.84375,
      "learning_rate": 6.850999812336623e-06,
      "loss": 1.60876675,
      "memory(GiB)": 107.26,
      "step": 32350,
      "train_speed(iter/s)": 1.635153
    },
    {
      "acc": 0.64838223,
      "epoch": 0.8207762557077626,
      "grad_norm": 7.25,
      "learning_rate": 6.850025650476259e-06,
      "loss": 1.62331543,
      "memory(GiB)": 107.26,
      "step": 32355,
      "train_speed(iter/s)": 1.63518
    },
    {
      "acc": 0.6412725,
      "epoch": 0.8209030948756976,
      "grad_norm": 5.0,
      "learning_rate": 6.849051407243509e-06,
      "loss": 1.62683983,
      "memory(GiB)": 107.26,
      "step": 32360,
      "train_speed(iter/s)": 1.635206
    },
    {
      "acc": 0.64832191,
      "epoch": 0.8210299340436327,
      "grad_norm": 6.1875,
      "learning_rate": 6.8480770826812205e-06,
      "loss": 1.61713371,
      "memory(GiB)": 107.26,
      "step": 32365,
      "train_speed(iter/s)": 1.635232
    },
    {
      "acc": 0.64966068,
      "epoch": 0.8211567732115678,
      "grad_norm": 5.53125,
      "learning_rate": 6.847102676832253e-06,
      "loss": 1.64985809,
      "memory(GiB)": 107.26,
      "step": 32370,
      "train_speed(iter/s)": 1.635258
    },
    {
      "acc": 0.6488471,
      "epoch": 0.8212836123795028,
      "grad_norm": 6.09375,
      "learning_rate": 6.8461281897394615e-06,
      "loss": 1.62956848,
      "memory(GiB)": 107.26,
      "step": 32375,
      "train_speed(iter/s)": 1.635287
    },
    {
      "acc": 0.64167128,
      "epoch": 0.8214104515474379,
      "grad_norm": 5.90625,
      "learning_rate": 6.845153621445711e-06,
      "loss": 1.66543274,
      "memory(GiB)": 107.26,
      "step": 32380,
      "train_speed(iter/s)": 1.635315
    },
    {
      "acc": 0.65637875,
      "epoch": 0.821537290715373,
      "grad_norm": 5.6875,
      "learning_rate": 6.844178971993866e-06,
      "loss": 1.66299438,
      "memory(GiB)": 107.26,
      "step": 32385,
      "train_speed(iter/s)": 1.635342
    },
    {
      "acc": 0.66767187,
      "epoch": 0.821664129883308,
      "grad_norm": 6.1875,
      "learning_rate": 6.843204241426797e-06,
      "loss": 1.58016481,
      "memory(GiB)": 107.26,
      "step": 32390,
      "train_speed(iter/s)": 1.635371
    },
    {
      "acc": 0.6691226,
      "epoch": 0.821790969051243,
      "grad_norm": 4.78125,
      "learning_rate": 6.842229429787375e-06,
      "loss": 1.61401806,
      "memory(GiB)": 107.26,
      "step": 32395,
      "train_speed(iter/s)": 1.635398
    },
    {
      "acc": 0.65529542,
      "epoch": 0.821917808219178,
      "grad_norm": 5.21875,
      "learning_rate": 6.841254537118477e-06,
      "loss": 1.53234911,
      "memory(GiB)": 107.26,
      "step": 32400,
      "train_speed(iter/s)": 1.635425
    },
    {
      "acc": 0.64349184,
      "epoch": 0.8220446473871131,
      "grad_norm": 5.78125,
      "learning_rate": 6.840279563462985e-06,
      "loss": 1.69066753,
      "memory(GiB)": 107.26,
      "step": 32405,
      "train_speed(iter/s)": 1.635455
    },
    {
      "acc": 0.65027962,
      "epoch": 0.8221714865550482,
      "grad_norm": 5.6875,
      "learning_rate": 6.839304508863781e-06,
      "loss": 1.61943359,
      "memory(GiB)": 107.26,
      "step": 32410,
      "train_speed(iter/s)": 1.635483
    },
    {
      "acc": 0.66247911,
      "epoch": 0.8222983257229832,
      "grad_norm": 5.03125,
      "learning_rate": 6.838329373363753e-06,
      "loss": 1.63647079,
      "memory(GiB)": 107.26,
      "step": 32415,
      "train_speed(iter/s)": 1.635511
    },
    {
      "acc": 0.65687923,
      "epoch": 0.8224251648909183,
      "grad_norm": 5.15625,
      "learning_rate": 6.8373541570057924e-06,
      "loss": 1.61744423,
      "memory(GiB)": 107.26,
      "step": 32420,
      "train_speed(iter/s)": 1.635539
    },
    {
      "acc": 0.66247163,
      "epoch": 0.8225520040588534,
      "grad_norm": 6.15625,
      "learning_rate": 6.836378859832791e-06,
      "loss": 1.61967697,
      "memory(GiB)": 107.26,
      "step": 32425,
      "train_speed(iter/s)": 1.635568
    },
    {
      "acc": 0.64993491,
      "epoch": 0.8226788432267884,
      "grad_norm": 5.21875,
      "learning_rate": 6.83540348188765e-06,
      "loss": 1.60473518,
      "memory(GiB)": 107.26,
      "step": 32430,
      "train_speed(iter/s)": 1.635597
    },
    {
      "acc": 0.66220131,
      "epoch": 0.8228056823947235,
      "grad_norm": 7.15625,
      "learning_rate": 6.834428023213268e-06,
      "loss": 1.5645031,
      "memory(GiB)": 107.26,
      "step": 32435,
      "train_speed(iter/s)": 1.635624
    },
    {
      "acc": 0.64771023,
      "epoch": 0.8229325215626585,
      "grad_norm": 4.6875,
      "learning_rate": 6.833452483852554e-06,
      "loss": 1.5846118,
      "memory(GiB)": 107.26,
      "step": 32440,
      "train_speed(iter/s)": 1.635654
    },
    {
      "acc": 0.65924816,
      "epoch": 0.8230593607305936,
      "grad_norm": 7.84375,
      "learning_rate": 6.832476863848411e-06,
      "loss": 1.59638138,
      "memory(GiB)": 107.26,
      "step": 32445,
      "train_speed(iter/s)": 1.63568
    },
    {
      "acc": 0.64527903,
      "epoch": 0.8231861998985287,
      "grad_norm": 6.09375,
      "learning_rate": 6.831501163243756e-06,
      "loss": 1.67870884,
      "memory(GiB)": 107.26,
      "step": 32450,
      "train_speed(iter/s)": 1.635706
    },
    {
      "acc": 0.64573102,
      "epoch": 0.8233130390664637,
      "grad_norm": 8.0,
      "learning_rate": 6.830525382081501e-06,
      "loss": 1.62880192,
      "memory(GiB)": 107.26,
      "step": 32455,
      "train_speed(iter/s)": 1.635733
    },
    {
      "acc": 0.65216861,
      "epoch": 0.8234398782343988,
      "grad_norm": 4.96875,
      "learning_rate": 6.829549520404568e-06,
      "loss": 1.57188931,
      "memory(GiB)": 107.26,
      "step": 32460,
      "train_speed(iter/s)": 1.635762
    },
    {
      "acc": 0.66088529,
      "epoch": 0.8235667174023339,
      "grad_norm": 5.40625,
      "learning_rate": 6.828573578255879e-06,
      "loss": 1.55594711,
      "memory(GiB)": 107.26,
      "step": 32465,
      "train_speed(iter/s)": 1.635787
    },
    {
      "acc": 0.65349712,
      "epoch": 0.8236935565702689,
      "grad_norm": 5.25,
      "learning_rate": 6.82759755567836e-06,
      "loss": 1.59866133,
      "memory(GiB)": 107.26,
      "step": 32470,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.6498455,
      "epoch": 0.823820395738204,
      "grad_norm": 5.65625,
      "learning_rate": 6.826621452714941e-06,
      "loss": 1.66149826,
      "memory(GiB)": 107.26,
      "step": 32475,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.64923029,
      "epoch": 0.823947234906139,
      "grad_norm": 5.03125,
      "learning_rate": 6.825645269408556e-06,
      "loss": 1.62181969,
      "memory(GiB)": 107.26,
      "step": 32480,
      "train_speed(iter/s)": 1.635872
    },
    {
      "acc": 0.6503171,
      "epoch": 0.8240740740740741,
      "grad_norm": 6.125,
      "learning_rate": 6.82466900580214e-06,
      "loss": 1.58356457,
      "memory(GiB)": 107.26,
      "step": 32485,
      "train_speed(iter/s)": 1.635899
    },
    {
      "acc": 0.64487762,
      "epoch": 0.8242009132420092,
      "grad_norm": 5.46875,
      "learning_rate": 6.823692661938634e-06,
      "loss": 1.66829224,
      "memory(GiB)": 107.26,
      "step": 32490,
      "train_speed(iter/s)": 1.635927
    },
    {
      "acc": 0.65993452,
      "epoch": 0.8243277524099442,
      "grad_norm": 6.25,
      "learning_rate": 6.822716237860984e-06,
      "loss": 1.62499199,
      "memory(GiB)": 107.26,
      "step": 32495,
      "train_speed(iter/s)": 1.635954
    },
    {
      "acc": 0.66408768,
      "epoch": 0.8244545915778793,
      "grad_norm": 5.09375,
      "learning_rate": 6.821739733612135e-06,
      "loss": 1.56554947,
      "memory(GiB)": 107.26,
      "step": 32500,
      "train_speed(iter/s)": 1.635983
    },
    {
      "acc": 0.65803909,
      "epoch": 0.8245814307458144,
      "grad_norm": 6.21875,
      "learning_rate": 6.820763149235039e-06,
      "loss": 1.68456879,
      "memory(GiB)": 107.26,
      "step": 32505,
      "train_speed(iter/s)": 1.636011
    },
    {
      "acc": 0.65737128,
      "epoch": 0.8247082699137493,
      "grad_norm": 5.4375,
      "learning_rate": 6.819786484772652e-06,
      "loss": 1.58501978,
      "memory(GiB)": 107.26,
      "step": 32510,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.62374258,
      "epoch": 0.8248351090816844,
      "grad_norm": 5.53125,
      "learning_rate": 6.8188097402679275e-06,
      "loss": 1.72287846,
      "memory(GiB)": 107.26,
      "step": 32515,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.6654952,
      "epoch": 0.8249619482496194,
      "grad_norm": 4.9375,
      "learning_rate": 6.817832915763833e-06,
      "loss": 1.54363441,
      "memory(GiB)": 107.26,
      "step": 32520,
      "train_speed(iter/s)": 1.636092
    },
    {
      "acc": 0.64710436,
      "epoch": 0.8250887874175545,
      "grad_norm": 5.53125,
      "learning_rate": 6.81685601130333e-06,
      "loss": 1.63108559,
      "memory(GiB)": 107.26,
      "step": 32525,
      "train_speed(iter/s)": 1.63612
    },
    {
      "acc": 0.66746187,
      "epoch": 0.8252156265854896,
      "grad_norm": 6.71875,
      "learning_rate": 6.8158790269293885e-06,
      "loss": 1.55511694,
      "memory(GiB)": 107.26,
      "step": 32530,
      "train_speed(iter/s)": 1.636149
    },
    {
      "acc": 0.666292,
      "epoch": 0.8253424657534246,
      "grad_norm": 6.5,
      "learning_rate": 6.8149019626849785e-06,
      "loss": 1.53077421,
      "memory(GiB)": 107.26,
      "step": 32535,
      "train_speed(iter/s)": 1.636177
    },
    {
      "acc": 0.65224347,
      "epoch": 0.8254693049213597,
      "grad_norm": 5.28125,
      "learning_rate": 6.813924818613079e-06,
      "loss": 1.62263145,
      "memory(GiB)": 107.26,
      "step": 32540,
      "train_speed(iter/s)": 1.636203
    },
    {
      "acc": 0.66190138,
      "epoch": 0.8255961440892948,
      "grad_norm": 6.1875,
      "learning_rate": 6.812947594756667e-06,
      "loss": 1.62164688,
      "memory(GiB)": 107.26,
      "step": 32545,
      "train_speed(iter/s)": 1.636231
    },
    {
      "acc": 0.67309875,
      "epoch": 0.8257229832572298,
      "grad_norm": 5.59375,
      "learning_rate": 6.811970291158725e-06,
      "loss": 1.58454685,
      "memory(GiB)": 107.26,
      "step": 32550,
      "train_speed(iter/s)": 1.636261
    },
    {
      "acc": 0.65828438,
      "epoch": 0.8258498224251649,
      "grad_norm": 5.53125,
      "learning_rate": 6.810992907862239e-06,
      "loss": 1.60759621,
      "memory(GiB)": 107.26,
      "step": 32555,
      "train_speed(iter/s)": 1.636287
    },
    {
      "acc": 0.63931465,
      "epoch": 0.8259766615930999,
      "grad_norm": 6.5625,
      "learning_rate": 6.810015444910202e-06,
      "loss": 1.60354195,
      "memory(GiB)": 107.26,
      "step": 32560,
      "train_speed(iter/s)": 1.636315
    },
    {
      "acc": 0.66305399,
      "epoch": 0.826103500761035,
      "grad_norm": 6.3125,
      "learning_rate": 6.809037902345603e-06,
      "loss": 1.65512371,
      "memory(GiB)": 107.26,
      "step": 32565,
      "train_speed(iter/s)": 1.636346
    },
    {
      "acc": 0.65062537,
      "epoch": 0.8262303399289701,
      "grad_norm": 6.40625,
      "learning_rate": 6.808060280211439e-06,
      "loss": 1.59509916,
      "memory(GiB)": 107.26,
      "step": 32570,
      "train_speed(iter/s)": 1.636374
    },
    {
      "acc": 0.6655386,
      "epoch": 0.8263571790969051,
      "grad_norm": 5.65625,
      "learning_rate": 6.807082578550713e-06,
      "loss": 1.58064823,
      "memory(GiB)": 107.26,
      "step": 32575,
      "train_speed(iter/s)": 1.636403
    },
    {
      "acc": 0.6549489,
      "epoch": 0.8264840182648402,
      "grad_norm": 5.96875,
      "learning_rate": 6.806104797406428e-06,
      "loss": 1.60469551,
      "memory(GiB)": 107.26,
      "step": 32580,
      "train_speed(iter/s)": 1.63643
    },
    {
      "acc": 0.65230603,
      "epoch": 0.8266108574327753,
      "grad_norm": 5.90625,
      "learning_rate": 6.805126936821588e-06,
      "loss": 1.62425346,
      "memory(GiB)": 107.26,
      "step": 32585,
      "train_speed(iter/s)": 1.636458
    },
    {
      "acc": 0.65537229,
      "epoch": 0.8267376966007103,
      "grad_norm": 5.5625,
      "learning_rate": 6.804148996839208e-06,
      "loss": 1.61069221,
      "memory(GiB)": 107.26,
      "step": 32590,
      "train_speed(iter/s)": 1.636486
    },
    {
      "acc": 0.65936303,
      "epoch": 0.8268645357686454,
      "grad_norm": 5.59375,
      "learning_rate": 6.803170977502298e-06,
      "loss": 1.62829285,
      "memory(GiB)": 107.26,
      "step": 32595,
      "train_speed(iter/s)": 1.636512
    },
    {
      "acc": 0.67077489,
      "epoch": 0.8269913749365804,
      "grad_norm": 9.0,
      "learning_rate": 6.802192878853879e-06,
      "loss": 1.60500183,
      "memory(GiB)": 107.26,
      "step": 32600,
      "train_speed(iter/s)": 1.636539
    },
    {
      "acc": 0.65993624,
      "epoch": 0.8271182141045155,
      "grad_norm": 6.0,
      "learning_rate": 6.801214700936972e-06,
      "loss": 1.5348999,
      "memory(GiB)": 107.26,
      "step": 32605,
      "train_speed(iter/s)": 1.636567
    },
    {
      "acc": 0.65883837,
      "epoch": 0.8272450532724506,
      "grad_norm": 5.46875,
      "learning_rate": 6.8002364437946e-06,
      "loss": 1.58786488,
      "memory(GiB)": 107.26,
      "step": 32610,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.65503912,
      "epoch": 0.8273718924403856,
      "grad_norm": 6.0,
      "learning_rate": 6.799258107469792e-06,
      "loss": 1.64665565,
      "memory(GiB)": 107.26,
      "step": 32615,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.65461464,
      "epoch": 0.8274987316083207,
      "grad_norm": 4.40625,
      "learning_rate": 6.798279692005578e-06,
      "loss": 1.56424026,
      "memory(GiB)": 107.26,
      "step": 32620,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.66185098,
      "epoch": 0.8276255707762558,
      "grad_norm": 6.125,
      "learning_rate": 6.7973011974449965e-06,
      "loss": 1.55325537,
      "memory(GiB)": 107.26,
      "step": 32625,
      "train_speed(iter/s)": 1.636678
    },
    {
      "acc": 0.66082478,
      "epoch": 0.8277524099441907,
      "grad_norm": 8.5,
      "learning_rate": 6.796322623831082e-06,
      "loss": 1.55895348,
      "memory(GiB)": 107.26,
      "step": 32630,
      "train_speed(iter/s)": 1.636708
    },
    {
      "acc": 0.65573144,
      "epoch": 0.8278792491121258,
      "grad_norm": 6.75,
      "learning_rate": 6.795343971206879e-06,
      "loss": 1.55839272,
      "memory(GiB)": 107.26,
      "step": 32635,
      "train_speed(iter/s)": 1.636738
    },
    {
      "acc": 0.65737138,
      "epoch": 0.8280060882800608,
      "grad_norm": 11.75,
      "learning_rate": 6.794365239615433e-06,
      "loss": 1.57717695,
      "memory(GiB)": 107.26,
      "step": 32640,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.65624828,
      "epoch": 0.8281329274479959,
      "grad_norm": 4.75,
      "learning_rate": 6.793386429099792e-06,
      "loss": 1.60059948,
      "memory(GiB)": 107.26,
      "step": 32645,
      "train_speed(iter/s)": 1.636792
    },
    {
      "acc": 0.66085892,
      "epoch": 0.828259766615931,
      "grad_norm": 5.0,
      "learning_rate": 6.79240753970301e-06,
      "loss": 1.5855855,
      "memory(GiB)": 107.26,
      "step": 32650,
      "train_speed(iter/s)": 1.636819
    },
    {
      "acc": 0.64934101,
      "epoch": 0.828386605783866,
      "grad_norm": 5.125,
      "learning_rate": 6.791428571468139e-06,
      "loss": 1.64846573,
      "memory(GiB)": 107.26,
      "step": 32655,
      "train_speed(iter/s)": 1.636847
    },
    {
      "acc": 0.64298506,
      "epoch": 0.8285134449518011,
      "grad_norm": 6.6875,
      "learning_rate": 6.7904495244382454e-06,
      "loss": 1.65990829,
      "memory(GiB)": 107.26,
      "step": 32660,
      "train_speed(iter/s)": 1.636875
    },
    {
      "acc": 0.6483386,
      "epoch": 0.8286402841197362,
      "grad_norm": 5.5,
      "learning_rate": 6.789470398656385e-06,
      "loss": 1.67112026,
      "memory(GiB)": 107.26,
      "step": 32665,
      "train_speed(iter/s)": 1.636903
    },
    {
      "acc": 0.66024408,
      "epoch": 0.8287671232876712,
      "grad_norm": 5.84375,
      "learning_rate": 6.788491194165629e-06,
      "loss": 1.5963726,
      "memory(GiB)": 107.26,
      "step": 32670,
      "train_speed(iter/s)": 1.636931
    },
    {
      "acc": 0.67239919,
      "epoch": 0.8288939624556063,
      "grad_norm": 5.65625,
      "learning_rate": 6.787511911009044e-06,
      "loss": 1.51363029,
      "memory(GiB)": 107.26,
      "step": 32675,
      "train_speed(iter/s)": 1.636959
    },
    {
      "acc": 0.64193835,
      "epoch": 0.8290208016235413,
      "grad_norm": 5.34375,
      "learning_rate": 6.786532549229704e-06,
      "loss": 1.63559685,
      "memory(GiB)": 107.26,
      "step": 32680,
      "train_speed(iter/s)": 1.636986
    },
    {
      "acc": 0.66636362,
      "epoch": 0.8291476407914764,
      "grad_norm": 5.25,
      "learning_rate": 6.785553108870686e-06,
      "loss": 1.54751329,
      "memory(GiB)": 107.26,
      "step": 32685,
      "train_speed(iter/s)": 1.637013
    },
    {
      "acc": 0.64006414,
      "epoch": 0.8292744799594115,
      "grad_norm": 5.375,
      "learning_rate": 6.784573589975072e-06,
      "loss": 1.646315,
      "memory(GiB)": 107.26,
      "step": 32690,
      "train_speed(iter/s)": 1.637041
    },
    {
      "acc": 0.63784962,
      "epoch": 0.8294013191273465,
      "grad_norm": 6.34375,
      "learning_rate": 6.783593992585943e-06,
      "loss": 1.67931118,
      "memory(GiB)": 107.26,
      "step": 32695,
      "train_speed(iter/s)": 1.637069
    },
    {
      "acc": 0.65170984,
      "epoch": 0.8295281582952816,
      "grad_norm": 5.53125,
      "learning_rate": 6.7826143167463876e-06,
      "loss": 1.64780731,
      "memory(GiB)": 107.26,
      "step": 32700,
      "train_speed(iter/s)": 1.637096
    },
    {
      "acc": 0.65854573,
      "epoch": 0.8296549974632167,
      "grad_norm": 5.8125,
      "learning_rate": 6.781634562499495e-06,
      "loss": 1.64535923,
      "memory(GiB)": 107.26,
      "step": 32705,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.66159353,
      "epoch": 0.8297818366311517,
      "grad_norm": 6.03125,
      "learning_rate": 6.780654729888361e-06,
      "loss": 1.60204048,
      "memory(GiB)": 107.26,
      "step": 32710,
      "train_speed(iter/s)": 1.637149
    },
    {
      "acc": 0.65879107,
      "epoch": 0.8299086757990868,
      "grad_norm": 10.6875,
      "learning_rate": 6.779674818956081e-06,
      "loss": 1.62156944,
      "memory(GiB)": 107.26,
      "step": 32715,
      "train_speed(iter/s)": 1.637177
    },
    {
      "acc": 0.63922291,
      "epoch": 0.8300355149670218,
      "grad_norm": 6.71875,
      "learning_rate": 6.778694829745756e-06,
      "loss": 1.632094,
      "memory(GiB)": 107.26,
      "step": 32720,
      "train_speed(iter/s)": 1.637204
    },
    {
      "acc": 0.63975558,
      "epoch": 0.8301623541349569,
      "grad_norm": 5.28125,
      "learning_rate": 6.777714762300492e-06,
      "loss": 1.60052319,
      "memory(GiB)": 107.26,
      "step": 32725,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.66300774,
      "epoch": 0.830289193302892,
      "grad_norm": 5.5,
      "learning_rate": 6.776734616663397e-06,
      "loss": 1.56456165,
      "memory(GiB)": 107.26,
      "step": 32730,
      "train_speed(iter/s)": 1.637256
    },
    {
      "acc": 0.67374697,
      "epoch": 0.830416032470827,
      "grad_norm": 5.46875,
      "learning_rate": 6.77575439287758e-06,
      "loss": 1.47687073,
      "memory(GiB)": 107.26,
      "step": 32735,
      "train_speed(iter/s)": 1.637282
    },
    {
      "acc": 0.65440354,
      "epoch": 0.8305428716387621,
      "grad_norm": 5.875,
      "learning_rate": 6.774774090986157e-06,
      "loss": 1.6119854,
      "memory(GiB)": 107.26,
      "step": 32740,
      "train_speed(iter/s)": 1.637309
    },
    {
      "acc": 0.67413797,
      "epoch": 0.8306697108066972,
      "grad_norm": 5.21875,
      "learning_rate": 6.773793711032244e-06,
      "loss": 1.58809319,
      "memory(GiB)": 107.26,
      "step": 32745,
      "train_speed(iter/s)": 1.637336
    },
    {
      "acc": 0.6478025,
      "epoch": 0.8307965499746321,
      "grad_norm": 5.90625,
      "learning_rate": 6.772813253058965e-06,
      "loss": 1.63967094,
      "memory(GiB)": 107.26,
      "step": 32750,
      "train_speed(iter/s)": 1.637362
    },
    {
      "acc": 0.65996633,
      "epoch": 0.8309233891425672,
      "grad_norm": 6.78125,
      "learning_rate": 6.771832717109444e-06,
      "loss": 1.58591442,
      "memory(GiB)": 107.26,
      "step": 32755,
      "train_speed(iter/s)": 1.63739
    },
    {
      "acc": 0.65029159,
      "epoch": 0.8310502283105022,
      "grad_norm": 4.96875,
      "learning_rate": 6.77085210322681e-06,
      "loss": 1.64019451,
      "memory(GiB)": 107.26,
      "step": 32760,
      "train_speed(iter/s)": 1.637418
    },
    {
      "acc": 0.64071603,
      "epoch": 0.8311770674784373,
      "grad_norm": 5.5,
      "learning_rate": 6.769871411454195e-06,
      "loss": 1.63422089,
      "memory(GiB)": 107.26,
      "step": 32765,
      "train_speed(iter/s)": 1.637445
    },
    {
      "acc": 0.6322793,
      "epoch": 0.8313039066463724,
      "grad_norm": 5.4375,
      "learning_rate": 6.768890641834732e-06,
      "loss": 1.68215828,
      "memory(GiB)": 107.26,
      "step": 32770,
      "train_speed(iter/s)": 1.637473
    },
    {
      "acc": 0.64893618,
      "epoch": 0.8314307458143074,
      "grad_norm": 4.6875,
      "learning_rate": 6.767909794411562e-06,
      "loss": 1.61356106,
      "memory(GiB)": 107.26,
      "step": 32775,
      "train_speed(iter/s)": 1.637499
    },
    {
      "acc": 0.66691628,
      "epoch": 0.8315575849822425,
      "grad_norm": 8.0625,
      "learning_rate": 6.7669288692278256e-06,
      "loss": 1.6025034,
      "memory(GiB)": 107.26,
      "step": 32780,
      "train_speed(iter/s)": 1.637526
    },
    {
      "acc": 0.65464559,
      "epoch": 0.8316844241501776,
      "grad_norm": 5.96875,
      "learning_rate": 6.76594786632667e-06,
      "loss": 1.53782272,
      "memory(GiB)": 107.26,
      "step": 32785,
      "train_speed(iter/s)": 1.637552
    },
    {
      "acc": 0.65324879,
      "epoch": 0.8318112633181126,
      "grad_norm": 5.65625,
      "learning_rate": 6.764966785751242e-06,
      "loss": 1.6249649,
      "memory(GiB)": 107.26,
      "step": 32790,
      "train_speed(iter/s)": 1.637579
    },
    {
      "acc": 0.65566382,
      "epoch": 0.8319381024860477,
      "grad_norm": 5.15625,
      "learning_rate": 6.763985627544693e-06,
      "loss": 1.62783623,
      "memory(GiB)": 107.26,
      "step": 32795,
      "train_speed(iter/s)": 1.637605
    },
    {
      "acc": 0.64118962,
      "epoch": 0.8320649416539827,
      "grad_norm": 5.3125,
      "learning_rate": 6.763004391750183e-06,
      "loss": 1.63354149,
      "memory(GiB)": 107.26,
      "step": 32800,
      "train_speed(iter/s)": 1.637632
    },
    {
      "acc": 0.65647268,
      "epoch": 0.8321917808219178,
      "grad_norm": 7.28125,
      "learning_rate": 6.762023078410867e-06,
      "loss": 1.62805367,
      "memory(GiB)": 107.26,
      "step": 32805,
      "train_speed(iter/s)": 1.637659
    },
    {
      "acc": 0.65218029,
      "epoch": 0.8323186199898529,
      "grad_norm": 5.0625,
      "learning_rate": 6.7610416875699095e-06,
      "loss": 1.58924828,
      "memory(GiB)": 107.26,
      "step": 32810,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.6623064,
      "epoch": 0.8324454591577879,
      "grad_norm": 5.625,
      "learning_rate": 6.760060219270476e-06,
      "loss": 1.57786417,
      "memory(GiB)": 107.26,
      "step": 32815,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.66748066,
      "epoch": 0.832572298325723,
      "grad_norm": 5.53125,
      "learning_rate": 6.759078673555736e-06,
      "loss": 1.5939127,
      "memory(GiB)": 107.26,
      "step": 32820,
      "train_speed(iter/s)": 1.637739
    },
    {
      "acc": 0.65092163,
      "epoch": 0.8326991374936581,
      "grad_norm": 6.21875,
      "learning_rate": 6.758097050468862e-06,
      "loss": 1.61537991,
      "memory(GiB)": 107.26,
      "step": 32825,
      "train_speed(iter/s)": 1.637767
    },
    {
      "acc": 0.66326427,
      "epoch": 0.8328259766615931,
      "grad_norm": 5.625,
      "learning_rate": 6.757115350053032e-06,
      "loss": 1.63619499,
      "memory(GiB)": 107.26,
      "step": 32830,
      "train_speed(iter/s)": 1.637793
    },
    {
      "acc": 0.64975243,
      "epoch": 0.8329528158295282,
      "grad_norm": 4.875,
      "learning_rate": 6.756133572351422e-06,
      "loss": 1.60693474,
      "memory(GiB)": 107.26,
      "step": 32835,
      "train_speed(iter/s)": 1.63782
    },
    {
      "acc": 0.65787697,
      "epoch": 0.8330796549974632,
      "grad_norm": 5.75,
      "learning_rate": 6.755151717407218e-06,
      "loss": 1.574506,
      "memory(GiB)": 107.26,
      "step": 32840,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.64539862,
      "epoch": 0.8332064941653983,
      "grad_norm": 4.9375,
      "learning_rate": 6.754169785263605e-06,
      "loss": 1.67978916,
      "memory(GiB)": 107.26,
      "step": 32845,
      "train_speed(iter/s)": 1.637875
    },
    {
      "acc": 0.65806422,
      "epoch": 0.8333333333333334,
      "grad_norm": 6.34375,
      "learning_rate": 6.753187775963773e-06,
      "loss": 1.61088943,
      "memory(GiB)": 107.26,
      "step": 32850,
      "train_speed(iter/s)": 1.637902
    },
    {
      "acc": 0.66494823,
      "epoch": 0.8334601725012684,
      "grad_norm": 5.9375,
      "learning_rate": 6.752205689550915e-06,
      "loss": 1.53929348,
      "memory(GiB)": 107.26,
      "step": 32855,
      "train_speed(iter/s)": 1.63793
    },
    {
      "acc": 0.65732417,
      "epoch": 0.8335870116692035,
      "grad_norm": 5.15625,
      "learning_rate": 6.751223526068228e-06,
      "loss": 1.609342,
      "memory(GiB)": 107.26,
      "step": 32860,
      "train_speed(iter/s)": 1.637957
    },
    {
      "acc": 0.64417095,
      "epoch": 0.8337138508371386,
      "grad_norm": 5.28125,
      "learning_rate": 6.75024128555891e-06,
      "loss": 1.6455349,
      "memory(GiB)": 107.26,
      "step": 32865,
      "train_speed(iter/s)": 1.637986
    },
    {
      "acc": 0.65553541,
      "epoch": 0.8338406900050735,
      "grad_norm": 5.09375,
      "learning_rate": 6.7492589680661695e-06,
      "loss": 1.61569824,
      "memory(GiB)": 107.26,
      "step": 32870,
      "train_speed(iter/s)": 1.638013
    },
    {
      "acc": 0.64900298,
      "epoch": 0.8339675291730086,
      "grad_norm": 5.9375,
      "learning_rate": 6.748276573633207e-06,
      "loss": 1.66464977,
      "memory(GiB)": 107.26,
      "step": 32875,
      "train_speed(iter/s)": 1.638041
    },
    {
      "acc": 0.65366497,
      "epoch": 0.8340943683409436,
      "grad_norm": 5.34375,
      "learning_rate": 6.747294102303237e-06,
      "loss": 1.6072998,
      "memory(GiB)": 107.26,
      "step": 32880,
      "train_speed(iter/s)": 1.638069
    },
    {
      "acc": 0.66518197,
      "epoch": 0.8342212075088787,
      "grad_norm": 5.625,
      "learning_rate": 6.746311554119469e-06,
      "loss": 1.56165924,
      "memory(GiB)": 107.26,
      "step": 32885,
      "train_speed(iter/s)": 1.638097
    },
    {
      "acc": 0.66784906,
      "epoch": 0.8343480466768138,
      "grad_norm": 6.03125,
      "learning_rate": 6.745328929125125e-06,
      "loss": 1.54831572,
      "memory(GiB)": 107.26,
      "step": 32890,
      "train_speed(iter/s)": 1.638124
    },
    {
      "acc": 0.65774364,
      "epoch": 0.8344748858447488,
      "grad_norm": 6.21875,
      "learning_rate": 6.7443462273634195e-06,
      "loss": 1.54018421,
      "memory(GiB)": 107.26,
      "step": 32895,
      "train_speed(iter/s)": 1.63815
    },
    {
      "acc": 0.65409651,
      "epoch": 0.8346017250126839,
      "grad_norm": 6.09375,
      "learning_rate": 6.74336344887758e-06,
      "loss": 1.56877642,
      "memory(GiB)": 107.26,
      "step": 32900,
      "train_speed(iter/s)": 1.638178
    },
    {
      "acc": 0.65434484,
      "epoch": 0.834728564180619,
      "grad_norm": 5.90625,
      "learning_rate": 6.742380593710834e-06,
      "loss": 1.59168415,
      "memory(GiB)": 107.26,
      "step": 32905,
      "train_speed(iter/s)": 1.638205
    },
    {
      "acc": 0.64572687,
      "epoch": 0.834855403348554,
      "grad_norm": 5.96875,
      "learning_rate": 6.7413976619064085e-06,
      "loss": 1.64755211,
      "memory(GiB)": 107.26,
      "step": 32910,
      "train_speed(iter/s)": 1.638232
    },
    {
      "acc": 0.65875878,
      "epoch": 0.8349822425164891,
      "grad_norm": 5.71875,
      "learning_rate": 6.74041465350754e-06,
      "loss": 1.5391428,
      "memory(GiB)": 107.26,
      "step": 32915,
      "train_speed(iter/s)": 1.638258
    },
    {
      "acc": 0.66857843,
      "epoch": 0.8351090816844241,
      "grad_norm": 5.375,
      "learning_rate": 6.739431568557464e-06,
      "loss": 1.49730883,
      "memory(GiB)": 107.26,
      "step": 32920,
      "train_speed(iter/s)": 1.638284
    },
    {
      "acc": 0.6353159,
      "epoch": 0.8352359208523592,
      "grad_norm": 5.1875,
      "learning_rate": 6.738448407099423e-06,
      "loss": 1.69268703,
      "memory(GiB)": 107.26,
      "step": 32925,
      "train_speed(iter/s)": 1.638312
    },
    {
      "acc": 0.65779152,
      "epoch": 0.8353627600202943,
      "grad_norm": 6.1875,
      "learning_rate": 6.737465169176658e-06,
      "loss": 1.5484416,
      "memory(GiB)": 107.26,
      "step": 32930,
      "train_speed(iter/s)": 1.638339
    },
    {
      "acc": 0.65046997,
      "epoch": 0.8354895991882293,
      "grad_norm": 8.5625,
      "learning_rate": 6.736481854832418e-06,
      "loss": 1.63785591,
      "memory(GiB)": 107.26,
      "step": 32935,
      "train_speed(iter/s)": 1.638365
    },
    {
      "acc": 0.65530529,
      "epoch": 0.8356164383561644,
      "grad_norm": 6.5625,
      "learning_rate": 6.735498464109953e-06,
      "loss": 1.5823575,
      "memory(GiB)": 107.26,
      "step": 32940,
      "train_speed(iter/s)": 1.638391
    },
    {
      "acc": 0.65970244,
      "epoch": 0.8357432775240995,
      "grad_norm": 6.5625,
      "learning_rate": 6.734514997052517e-06,
      "loss": 1.54790878,
      "memory(GiB)": 107.26,
      "step": 32945,
      "train_speed(iter/s)": 1.638418
    },
    {
      "acc": 0.66891809,
      "epoch": 0.8358701166920345,
      "grad_norm": 8.25,
      "learning_rate": 6.733531453703368e-06,
      "loss": 1.60948639,
      "memory(GiB)": 107.26,
      "step": 32950,
      "train_speed(iter/s)": 1.638444
    },
    {
      "acc": 0.65748167,
      "epoch": 0.8359969558599696,
      "grad_norm": 7.125,
      "learning_rate": 6.732547834105765e-06,
      "loss": 1.53375082,
      "memory(GiB)": 107.26,
      "step": 32955,
      "train_speed(iter/s)": 1.638471
    },
    {
      "acc": 0.66838403,
      "epoch": 0.8361237950279046,
      "grad_norm": 5.96875,
      "learning_rate": 6.731564138302975e-06,
      "loss": 1.58501377,
      "memory(GiB)": 107.26,
      "step": 32960,
      "train_speed(iter/s)": 1.638498
    },
    {
      "acc": 0.66609545,
      "epoch": 0.8362506341958397,
      "grad_norm": 5.0,
      "learning_rate": 6.730580366338261e-06,
      "loss": 1.60643272,
      "memory(GiB)": 107.26,
      "step": 32965,
      "train_speed(iter/s)": 1.638526
    },
    {
      "acc": 0.66789684,
      "epoch": 0.8363774733637748,
      "grad_norm": 5.09375,
      "learning_rate": 6.729596518254897e-06,
      "loss": 1.55170212,
      "memory(GiB)": 107.26,
      "step": 32970,
      "train_speed(iter/s)": 1.63855
    },
    {
      "acc": 0.64494123,
      "epoch": 0.8365043125317098,
      "grad_norm": 7.1875,
      "learning_rate": 6.728612594096155e-06,
      "loss": 1.61461601,
      "memory(GiB)": 107.26,
      "step": 32975,
      "train_speed(iter/s)": 1.638578
    },
    {
      "acc": 0.63801928,
      "epoch": 0.8366311516996449,
      "grad_norm": 5.0,
      "learning_rate": 6.727628593905315e-06,
      "loss": 1.653965,
      "memory(GiB)": 107.26,
      "step": 32980,
      "train_speed(iter/s)": 1.638604
    },
    {
      "acc": 0.66122475,
      "epoch": 0.83675799086758,
      "grad_norm": 6.21875,
      "learning_rate": 6.726644517725655e-06,
      "loss": 1.60932159,
      "memory(GiB)": 107.26,
      "step": 32985,
      "train_speed(iter/s)": 1.638631
    },
    {
      "acc": 0.63274193,
      "epoch": 0.836884830035515,
      "grad_norm": 6.59375,
      "learning_rate": 6.725660365600462e-06,
      "loss": 1.68729858,
      "memory(GiB)": 107.26,
      "step": 32990,
      "train_speed(iter/s)": 1.638658
    },
    {
      "acc": 0.63724275,
      "epoch": 0.83701166920345,
      "grad_norm": 6.21875,
      "learning_rate": 6.724676137573021e-06,
      "loss": 1.64082508,
      "memory(GiB)": 107.26,
      "step": 32995,
      "train_speed(iter/s)": 1.638684
    },
    {
      "acc": 0.63562245,
      "epoch": 0.837138508371385,
      "grad_norm": 8.3125,
      "learning_rate": 6.723691833686622e-06,
      "loss": 1.63438835,
      "memory(GiB)": 107.26,
      "step": 33000,
      "train_speed(iter/s)": 1.638711
    },
    {
      "epoch": 0.837138508371385,
      "eval_acc": 0.6457886174787045,
      "eval_loss": 1.575567364692688,
      "eval_runtime": 58.089,
      "eval_samples_per_second": 109.659,
      "eval_steps_per_second": 27.423,
      "step": 33000
    },
    {
      "acc": 0.65270977,
      "epoch": 0.8372653475393201,
      "grad_norm": 7.25,
      "learning_rate": 6.722707453984561e-06,
      "loss": 1.63201828,
      "memory(GiB)": 107.26,
      "step": 33005,
      "train_speed(iter/s)": 1.633681
    },
    {
      "acc": 0.67310781,
      "epoch": 0.8373921867072552,
      "grad_norm": 6.90625,
      "learning_rate": 6.721722998510135e-06,
      "loss": 1.63963547,
      "memory(GiB)": 107.26,
      "step": 33010,
      "train_speed(iter/s)": 1.633708
    },
    {
      "acc": 0.6590044,
      "epoch": 0.8375190258751902,
      "grad_norm": 5.1875,
      "learning_rate": 6.720738467306644e-06,
      "loss": 1.58788128,
      "memory(GiB)": 107.26,
      "step": 33015,
      "train_speed(iter/s)": 1.633734
    },
    {
      "acc": 0.66271214,
      "epoch": 0.8376458650431253,
      "grad_norm": 5.1875,
      "learning_rate": 6.719753860417394e-06,
      "loss": 1.63545761,
      "memory(GiB)": 107.26,
      "step": 33020,
      "train_speed(iter/s)": 1.633759
    },
    {
      "acc": 0.64273181,
      "epoch": 0.8377727042110604,
      "grad_norm": 6.71875,
      "learning_rate": 6.718769177885689e-06,
      "loss": 1.64766083,
      "memory(GiB)": 107.26,
      "step": 33025,
      "train_speed(iter/s)": 1.633785
    },
    {
      "acc": 0.67503633,
      "epoch": 0.8378995433789954,
      "grad_norm": 5.34375,
      "learning_rate": 6.717784419754845e-06,
      "loss": 1.50163164,
      "memory(GiB)": 107.26,
      "step": 33030,
      "train_speed(iter/s)": 1.633812
    },
    {
      "acc": 0.66373549,
      "epoch": 0.8380263825469305,
      "grad_norm": 5.78125,
      "learning_rate": 6.71679958606817e-06,
      "loss": 1.56779099,
      "memory(GiB)": 107.26,
      "step": 33035,
      "train_speed(iter/s)": 1.63384
    },
    {
      "acc": 0.63731375,
      "epoch": 0.8381532217148655,
      "grad_norm": 5.90625,
      "learning_rate": 6.715814676868985e-06,
      "loss": 1.66424999,
      "memory(GiB)": 107.26,
      "step": 33040,
      "train_speed(iter/s)": 1.633867
    },
    {
      "acc": 0.64177208,
      "epoch": 0.8382800608828006,
      "grad_norm": 5.09375,
      "learning_rate": 6.714829692200611e-06,
      "loss": 1.68111534,
      "memory(GiB)": 107.26,
      "step": 33045,
      "train_speed(iter/s)": 1.633894
    },
    {
      "acc": 0.64890814,
      "epoch": 0.8384069000507357,
      "grad_norm": 7.6875,
      "learning_rate": 6.71384463210637e-06,
      "loss": 1.67067566,
      "memory(GiB)": 107.26,
      "step": 33050,
      "train_speed(iter/s)": 1.633919
    },
    {
      "acc": 0.65133052,
      "epoch": 0.8385337392186707,
      "grad_norm": 5.25,
      "learning_rate": 6.7128594966295904e-06,
      "loss": 1.58246441,
      "memory(GiB)": 107.26,
      "step": 33055,
      "train_speed(iter/s)": 1.63394
    },
    {
      "acc": 0.66042347,
      "epoch": 0.8386605783866058,
      "grad_norm": 5.84375,
      "learning_rate": 6.711874285813602e-06,
      "loss": 1.56692257,
      "memory(GiB)": 107.26,
      "step": 33060,
      "train_speed(iter/s)": 1.633965
    },
    {
      "acc": 0.66584649,
      "epoch": 0.8387874175545409,
      "grad_norm": 4.96875,
      "learning_rate": 6.710888999701741e-06,
      "loss": 1.60463047,
      "memory(GiB)": 107.26,
      "step": 33065,
      "train_speed(iter/s)": 1.633991
    },
    {
      "acc": 0.66672449,
      "epoch": 0.8389142567224759,
      "grad_norm": 5.8125,
      "learning_rate": 6.7099036383373425e-06,
      "loss": 1.53181839,
      "memory(GiB)": 107.26,
      "step": 33070,
      "train_speed(iter/s)": 1.634018
    },
    {
      "acc": 0.64782906,
      "epoch": 0.839041095890411,
      "grad_norm": 5.8125,
      "learning_rate": 6.708918201763748e-06,
      "loss": 1.63721275,
      "memory(GiB)": 107.26,
      "step": 33075,
      "train_speed(iter/s)": 1.634044
    },
    {
      "acc": 0.6564826,
      "epoch": 0.839167935058346,
      "grad_norm": 5.09375,
      "learning_rate": 6.707932690024302e-06,
      "loss": 1.63065109,
      "memory(GiB)": 107.26,
      "step": 33080,
      "train_speed(iter/s)": 1.634071
    },
    {
      "acc": 0.64856415,
      "epoch": 0.8392947742262811,
      "grad_norm": 6.28125,
      "learning_rate": 6.706947103162348e-06,
      "loss": 1.57642059,
      "memory(GiB)": 107.26,
      "step": 33085,
      "train_speed(iter/s)": 1.634099
    },
    {
      "acc": 0.65623808,
      "epoch": 0.8394216133942162,
      "grad_norm": 6.09375,
      "learning_rate": 6.7059614412212425e-06,
      "loss": 1.64540348,
      "memory(GiB)": 107.26,
      "step": 33090,
      "train_speed(iter/s)": 1.634127
    },
    {
      "acc": 0.65889034,
      "epoch": 0.8395484525621512,
      "grad_norm": 5.3125,
      "learning_rate": 6.704975704244334e-06,
      "loss": 1.60552216,
      "memory(GiB)": 107.26,
      "step": 33095,
      "train_speed(iter/s)": 1.634154
    },
    {
      "acc": 0.65190573,
      "epoch": 0.8396752917300863,
      "grad_norm": 7.25,
      "learning_rate": 6.703989892274985e-06,
      "loss": 1.65951271,
      "memory(GiB)": 107.26,
      "step": 33100,
      "train_speed(iter/s)": 1.634182
    },
    {
      "acc": 0.65876555,
      "epoch": 0.8398021308980214,
      "grad_norm": 8.0625,
      "learning_rate": 6.703004005356549e-06,
      "loss": 1.55150394,
      "memory(GiB)": 107.26,
      "step": 33105,
      "train_speed(iter/s)": 1.634209
    },
    {
      "acc": 0.63346443,
      "epoch": 0.8399289700659563,
      "grad_norm": 7.78125,
      "learning_rate": 6.7020180435323965e-06,
      "loss": 1.65833244,
      "memory(GiB)": 107.26,
      "step": 33110,
      "train_speed(iter/s)": 1.634239
    },
    {
      "acc": 0.64021816,
      "epoch": 0.8400558092338914,
      "grad_norm": 5.9375,
      "learning_rate": 6.701032006845889e-06,
      "loss": 1.61048431,
      "memory(GiB)": 107.26,
      "step": 33115,
      "train_speed(iter/s)": 1.634266
    },
    {
      "acc": 0.66582642,
      "epoch": 0.8401826484018264,
      "grad_norm": 5.875,
      "learning_rate": 6.700045895340401e-06,
      "loss": 1.53527222,
      "memory(GiB)": 107.26,
      "step": 33120,
      "train_speed(iter/s)": 1.634295
    },
    {
      "acc": 0.65648694,
      "epoch": 0.8403094875697615,
      "grad_norm": 7.90625,
      "learning_rate": 6.699059709059304e-06,
      "loss": 1.64785976,
      "memory(GiB)": 107.26,
      "step": 33125,
      "train_speed(iter/s)": 1.634324
    },
    {
      "acc": 0.67328687,
      "epoch": 0.8404363267376966,
      "grad_norm": 5.34375,
      "learning_rate": 6.698073448045975e-06,
      "loss": 1.58423519,
      "memory(GiB)": 107.26,
      "step": 33130,
      "train_speed(iter/s)": 1.634351
    },
    {
      "acc": 0.6677012,
      "epoch": 0.8405631659056316,
      "grad_norm": 5.21875,
      "learning_rate": 6.697087112343795e-06,
      "loss": 1.5326704,
      "memory(GiB)": 107.26,
      "step": 33135,
      "train_speed(iter/s)": 1.634379
    },
    {
      "acc": 0.64167943,
      "epoch": 0.8406900050735667,
      "grad_norm": 5.9375,
      "learning_rate": 6.696100701996146e-06,
      "loss": 1.68892384,
      "memory(GiB)": 107.26,
      "step": 33140,
      "train_speed(iter/s)": 1.634408
    },
    {
      "acc": 0.66409349,
      "epoch": 0.8408168442415018,
      "grad_norm": 6.25,
      "learning_rate": 6.6951142170464164e-06,
      "loss": 1.55952644,
      "memory(GiB)": 107.26,
      "step": 33145,
      "train_speed(iter/s)": 1.634436
    },
    {
      "acc": 0.65505791,
      "epoch": 0.8409436834094368,
      "grad_norm": 5.65625,
      "learning_rate": 6.694127657537995e-06,
      "loss": 1.6306942,
      "memory(GiB)": 107.26,
      "step": 33150,
      "train_speed(iter/s)": 1.634465
    },
    {
      "acc": 0.64726124,
      "epoch": 0.8410705225773719,
      "grad_norm": 5.5,
      "learning_rate": 6.693141023514276e-06,
      "loss": 1.60471764,
      "memory(GiB)": 107.26,
      "step": 33155,
      "train_speed(iter/s)": 1.634493
    },
    {
      "acc": 0.64906836,
      "epoch": 0.8411973617453069,
      "grad_norm": 5.53125,
      "learning_rate": 6.6921543150186555e-06,
      "loss": 1.63876686,
      "memory(GiB)": 107.26,
      "step": 33160,
      "train_speed(iter/s)": 1.63452
    },
    {
      "acc": 0.66530533,
      "epoch": 0.841324200913242,
      "grad_norm": 6.8125,
      "learning_rate": 6.691167532094531e-06,
      "loss": 1.58795719,
      "memory(GiB)": 107.26,
      "step": 33165,
      "train_speed(iter/s)": 1.634548
    },
    {
      "acc": 0.66380329,
      "epoch": 0.8414510400811771,
      "grad_norm": 6.96875,
      "learning_rate": 6.690180674785311e-06,
      "loss": 1.66640186,
      "memory(GiB)": 107.26,
      "step": 33170,
      "train_speed(iter/s)": 1.634577
    },
    {
      "acc": 0.64290233,
      "epoch": 0.8415778792491121,
      "grad_norm": 5.4375,
      "learning_rate": 6.689193743134397e-06,
      "loss": 1.64339123,
      "memory(GiB)": 107.26,
      "step": 33175,
      "train_speed(iter/s)": 1.634607
    },
    {
      "acc": 0.65074568,
      "epoch": 0.8417047184170472,
      "grad_norm": 6.65625,
      "learning_rate": 6.688206737185201e-06,
      "loss": 1.61374569,
      "memory(GiB)": 107.26,
      "step": 33180,
      "train_speed(iter/s)": 1.634636
    },
    {
      "acc": 0.64724813,
      "epoch": 0.8418315575849823,
      "grad_norm": 6.21875,
      "learning_rate": 6.687219656981135e-06,
      "loss": 1.59716864,
      "memory(GiB)": 107.26,
      "step": 33185,
      "train_speed(iter/s)": 1.634662
    },
    {
      "acc": 0.67918143,
      "epoch": 0.8419583967529173,
      "grad_norm": 5.75,
      "learning_rate": 6.686232502565616e-06,
      "loss": 1.57895031,
      "memory(GiB)": 107.26,
      "step": 33190,
      "train_speed(iter/s)": 1.634691
    },
    {
      "acc": 0.65584588,
      "epoch": 0.8420852359208524,
      "grad_norm": 5.5625,
      "learning_rate": 6.685245273982063e-06,
      "loss": 1.64173012,
      "memory(GiB)": 107.26,
      "step": 33195,
      "train_speed(iter/s)": 1.63472
    },
    {
      "acc": 0.63902512,
      "epoch": 0.8422120750887874,
      "grad_norm": 5.875,
      "learning_rate": 6.684257971273899e-06,
      "loss": 1.72377224,
      "memory(GiB)": 107.26,
      "step": 33200,
      "train_speed(iter/s)": 1.634748
    },
    {
      "acc": 0.66817207,
      "epoch": 0.8423389142567225,
      "grad_norm": 5.1875,
      "learning_rate": 6.68327059448455e-06,
      "loss": 1.57995291,
      "memory(GiB)": 107.26,
      "step": 33205,
      "train_speed(iter/s)": 1.634776
    },
    {
      "acc": 0.66396513,
      "epoch": 0.8424657534246576,
      "grad_norm": 7.1875,
      "learning_rate": 6.682283143657444e-06,
      "loss": 1.54377804,
      "memory(GiB)": 107.26,
      "step": 33210,
      "train_speed(iter/s)": 1.634806
    },
    {
      "acc": 0.65219955,
      "epoch": 0.8425925925925926,
      "grad_norm": 7.78125,
      "learning_rate": 6.681295618836015e-06,
      "loss": 1.67003822,
      "memory(GiB)": 107.26,
      "step": 33215,
      "train_speed(iter/s)": 1.634836
    },
    {
      "acc": 0.64577007,
      "epoch": 0.8427194317605277,
      "grad_norm": 5.28125,
      "learning_rate": 6.680308020063699e-06,
      "loss": 1.61047211,
      "memory(GiB)": 107.26,
      "step": 33220,
      "train_speed(iter/s)": 1.634864
    },
    {
      "acc": 0.65421352,
      "epoch": 0.8428462709284628,
      "grad_norm": 5.9375,
      "learning_rate": 6.679320347383933e-06,
      "loss": 1.63300457,
      "memory(GiB)": 107.26,
      "step": 33225,
      "train_speed(iter/s)": 1.634891
    },
    {
      "acc": 0.6567976,
      "epoch": 0.8429731100963977,
      "grad_norm": 7.1875,
      "learning_rate": 6.678332600840161e-06,
      "loss": 1.63503456,
      "memory(GiB)": 107.26,
      "step": 33230,
      "train_speed(iter/s)": 1.634919
    },
    {
      "acc": 0.67380571,
      "epoch": 0.8430999492643328,
      "grad_norm": 5.625,
      "learning_rate": 6.677344780475827e-06,
      "loss": 1.53659,
      "memory(GiB)": 107.26,
      "step": 33235,
      "train_speed(iter/s)": 1.634949
    },
    {
      "acc": 0.66985335,
      "epoch": 0.8432267884322678,
      "grad_norm": 5.09375,
      "learning_rate": 6.676356886334383e-06,
      "loss": 1.56984911,
      "memory(GiB)": 107.26,
      "step": 33240,
      "train_speed(iter/s)": 1.634976
    },
    {
      "acc": 0.6458457,
      "epoch": 0.8433536276002029,
      "grad_norm": 5.90625,
      "learning_rate": 6.675368918459276e-06,
      "loss": 1.65674934,
      "memory(GiB)": 107.26,
      "step": 33245,
      "train_speed(iter/s)": 1.635005
    },
    {
      "acc": 0.67723303,
      "epoch": 0.843480466768138,
      "grad_norm": 5.09375,
      "learning_rate": 6.674380876893967e-06,
      "loss": 1.52234802,
      "memory(GiB)": 107.26,
      "step": 33250,
      "train_speed(iter/s)": 1.635035
    },
    {
      "acc": 0.65861788,
      "epoch": 0.843607305936073,
      "grad_norm": 7.34375,
      "learning_rate": 6.673392761681908e-06,
      "loss": 1.56752672,
      "memory(GiB)": 107.26,
      "step": 33255,
      "train_speed(iter/s)": 1.635063
    },
    {
      "acc": 0.66966267,
      "epoch": 0.8437341451040081,
      "grad_norm": 5.34375,
      "learning_rate": 6.672404572866566e-06,
      "loss": 1.55277596,
      "memory(GiB)": 107.26,
      "step": 33260,
      "train_speed(iter/s)": 1.635091
    },
    {
      "acc": 0.66005096,
      "epoch": 0.8438609842719432,
      "grad_norm": 5.28125,
      "learning_rate": 6.671416310491406e-06,
      "loss": 1.53709698,
      "memory(GiB)": 107.26,
      "step": 33265,
      "train_speed(iter/s)": 1.635115
    },
    {
      "acc": 0.64196749,
      "epoch": 0.8439878234398782,
      "grad_norm": 5.96875,
      "learning_rate": 6.670427974599891e-06,
      "loss": 1.69622765,
      "memory(GiB)": 107.26,
      "step": 33270,
      "train_speed(iter/s)": 1.635143
    },
    {
      "acc": 0.65786552,
      "epoch": 0.8441146626078133,
      "grad_norm": 7.8125,
      "learning_rate": 6.669439565235498e-06,
      "loss": 1.64209785,
      "memory(GiB)": 107.26,
      "step": 33275,
      "train_speed(iter/s)": 1.635171
    },
    {
      "acc": 0.6553998,
      "epoch": 0.8442415017757483,
      "grad_norm": 5.9375,
      "learning_rate": 6.668451082441698e-06,
      "loss": 1.61554947,
      "memory(GiB)": 107.26,
      "step": 33280,
      "train_speed(iter/s)": 1.6352
    },
    {
      "acc": 0.63697472,
      "epoch": 0.8443683409436834,
      "grad_norm": 5.875,
      "learning_rate": 6.667462526261972e-06,
      "loss": 1.66026955,
      "memory(GiB)": 107.26,
      "step": 33285,
      "train_speed(iter/s)": 1.635228
    },
    {
      "acc": 0.66530867,
      "epoch": 0.8444951801116185,
      "grad_norm": 4.59375,
      "learning_rate": 6.666473896739798e-06,
      "loss": 1.57641735,
      "memory(GiB)": 107.26,
      "step": 33290,
      "train_speed(iter/s)": 1.635256
    },
    {
      "acc": 0.65954456,
      "epoch": 0.8446220192795535,
      "grad_norm": 6.09375,
      "learning_rate": 6.665485193918663e-06,
      "loss": 1.59219036,
      "memory(GiB)": 107.26,
      "step": 33295,
      "train_speed(iter/s)": 1.635284
    },
    {
      "acc": 0.64143925,
      "epoch": 0.8447488584474886,
      "grad_norm": 6.0625,
      "learning_rate": 6.664496417842053e-06,
      "loss": 1.61944141,
      "memory(GiB)": 107.26,
      "step": 33300,
      "train_speed(iter/s)": 1.635314
    },
    {
      "acc": 0.6585248,
      "epoch": 0.8448756976154237,
      "grad_norm": 4.6875,
      "learning_rate": 6.6635075685534566e-06,
      "loss": 1.56756182,
      "memory(GiB)": 107.26,
      "step": 33305,
      "train_speed(iter/s)": 1.635342
    },
    {
      "acc": 0.64922209,
      "epoch": 0.8450025367833587,
      "grad_norm": 7.25,
      "learning_rate": 6.662518646096374e-06,
      "loss": 1.63727036,
      "memory(GiB)": 107.26,
      "step": 33310,
      "train_speed(iter/s)": 1.635371
    },
    {
      "acc": 0.65412908,
      "epoch": 0.8451293759512938,
      "grad_norm": 4.8125,
      "learning_rate": 6.661529650514296e-06,
      "loss": 1.61124172,
      "memory(GiB)": 107.26,
      "step": 33315,
      "train_speed(iter/s)": 1.635397
    },
    {
      "acc": 0.64485292,
      "epoch": 0.8452562151192288,
      "grad_norm": 5.40625,
      "learning_rate": 6.6605405818507274e-06,
      "loss": 1.64717197,
      "memory(GiB)": 107.26,
      "step": 33320,
      "train_speed(iter/s)": 1.635425
    },
    {
      "acc": 0.64504681,
      "epoch": 0.8453830542871639,
      "grad_norm": 6.0,
      "learning_rate": 6.659551440149169e-06,
      "loss": 1.59644623,
      "memory(GiB)": 107.26,
      "step": 33325,
      "train_speed(iter/s)": 1.635451
    },
    {
      "acc": 0.66047468,
      "epoch": 0.845509893455099,
      "grad_norm": 6.125,
      "learning_rate": 6.65856222545313e-06,
      "loss": 1.58540964,
      "memory(GiB)": 107.26,
      "step": 33330,
      "train_speed(iter/s)": 1.635478
    },
    {
      "acc": 0.65453253,
      "epoch": 0.845636732623034,
      "grad_norm": 4.65625,
      "learning_rate": 6.657572937806118e-06,
      "loss": 1.6203083,
      "memory(GiB)": 107.26,
      "step": 33335,
      "train_speed(iter/s)": 1.635504
    },
    {
      "acc": 0.66870689,
      "epoch": 0.8457635717909691,
      "grad_norm": 5.0,
      "learning_rate": 6.656583577251649e-06,
      "loss": 1.50935307,
      "memory(GiB)": 107.26,
      "step": 33340,
      "train_speed(iter/s)": 1.63553
    },
    {
      "acc": 0.63879795,
      "epoch": 0.8458904109589042,
      "grad_norm": 4.9375,
      "learning_rate": 6.655594143833237e-06,
      "loss": 1.66118393,
      "memory(GiB)": 107.26,
      "step": 33345,
      "train_speed(iter/s)": 1.635559
    },
    {
      "acc": 0.65120659,
      "epoch": 0.8460172501268391,
      "grad_norm": 5.65625,
      "learning_rate": 6.654604637594404e-06,
      "loss": 1.60661087,
      "memory(GiB)": 107.26,
      "step": 33350,
      "train_speed(iter/s)": 1.635587
    },
    {
      "acc": 0.64487476,
      "epoch": 0.8461440892947742,
      "grad_norm": 5.75,
      "learning_rate": 6.653615058578672e-06,
      "loss": 1.6541954,
      "memory(GiB)": 107.26,
      "step": 33355,
      "train_speed(iter/s)": 1.635615
    },
    {
      "acc": 0.67058682,
      "epoch": 0.8462709284627092,
      "grad_norm": 5.59375,
      "learning_rate": 6.652625406829566e-06,
      "loss": 1.56716003,
      "memory(GiB)": 107.26,
      "step": 33360,
      "train_speed(iter/s)": 1.635644
    },
    {
      "acc": 0.65444999,
      "epoch": 0.8463977676306443,
      "grad_norm": 7.125,
      "learning_rate": 6.651635682390616e-06,
      "loss": 1.56987514,
      "memory(GiB)": 107.26,
      "step": 33365,
      "train_speed(iter/s)": 1.63567
    },
    {
      "acc": 0.66086345,
      "epoch": 0.8465246067985794,
      "grad_norm": 6.03125,
      "learning_rate": 6.650645885305356e-06,
      "loss": 1.57297497,
      "memory(GiB)": 107.26,
      "step": 33370,
      "train_speed(iter/s)": 1.635697
    },
    {
      "acc": 0.64320002,
      "epoch": 0.8466514459665144,
      "grad_norm": 8.625,
      "learning_rate": 6.649656015617319e-06,
      "loss": 1.68561897,
      "memory(GiB)": 107.26,
      "step": 33375,
      "train_speed(iter/s)": 1.635724
    },
    {
      "acc": 0.65072069,
      "epoch": 0.8467782851344495,
      "grad_norm": 5.03125,
      "learning_rate": 6.648666073370046e-06,
      "loss": 1.58533573,
      "memory(GiB)": 107.26,
      "step": 33380,
      "train_speed(iter/s)": 1.635749
    },
    {
      "acc": 0.65497293,
      "epoch": 0.8469051243023846,
      "grad_norm": 5.25,
      "learning_rate": 6.647676058607076e-06,
      "loss": 1.62276821,
      "memory(GiB)": 107.26,
      "step": 33385,
      "train_speed(iter/s)": 1.635775
    },
    {
      "acc": 0.65208297,
      "epoch": 0.8470319634703196,
      "grad_norm": 5.84375,
      "learning_rate": 6.64668597137196e-06,
      "loss": 1.6559433,
      "memory(GiB)": 107.26,
      "step": 33390,
      "train_speed(iter/s)": 1.635803
    },
    {
      "acc": 0.66810999,
      "epoch": 0.8471588026382547,
      "grad_norm": 6.625,
      "learning_rate": 6.645695811708241e-06,
      "loss": 1.59134617,
      "memory(GiB)": 107.26,
      "step": 33395,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.66600256,
      "epoch": 0.8472856418061897,
      "grad_norm": 5.25,
      "learning_rate": 6.644705579659474e-06,
      "loss": 1.60210209,
      "memory(GiB)": 107.26,
      "step": 33400,
      "train_speed(iter/s)": 1.635857
    },
    {
      "acc": 0.65742593,
      "epoch": 0.8474124809741248,
      "grad_norm": 6.53125,
      "learning_rate": 6.643715275269212e-06,
      "loss": 1.62654228,
      "memory(GiB)": 107.26,
      "step": 33405,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.65765524,
      "epoch": 0.8475393201420599,
      "grad_norm": 6.3125,
      "learning_rate": 6.642724898581013e-06,
      "loss": 1.62223473,
      "memory(GiB)": 107.26,
      "step": 33410,
      "train_speed(iter/s)": 1.635912
    },
    {
      "acc": 0.66433859,
      "epoch": 0.8476661593099949,
      "grad_norm": 5.90625,
      "learning_rate": 6.6417344496384394e-06,
      "loss": 1.57372742,
      "memory(GiB)": 107.26,
      "step": 33415,
      "train_speed(iter/s)": 1.635938
    },
    {
      "acc": 0.62334013,
      "epoch": 0.84779299847793,
      "grad_norm": 4.53125,
      "learning_rate": 6.640743928485054e-06,
      "loss": 1.68893623,
      "memory(GiB)": 107.26,
      "step": 33420,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.65648451,
      "epoch": 0.8479198376458651,
      "grad_norm": 8.1875,
      "learning_rate": 6.639753335164426e-06,
      "loss": 1.66761837,
      "memory(GiB)": 107.26,
      "step": 33425,
      "train_speed(iter/s)": 1.635992
    },
    {
      "acc": 0.65545902,
      "epoch": 0.8480466768138001,
      "grad_norm": 6.1875,
      "learning_rate": 6.638762669720126e-06,
      "loss": 1.61317215,
      "memory(GiB)": 107.26,
      "step": 33430,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.67150221,
      "epoch": 0.8481735159817352,
      "grad_norm": 6.34375,
      "learning_rate": 6.637771932195726e-06,
      "loss": 1.53342361,
      "memory(GiB)": 107.26,
      "step": 33435,
      "train_speed(iter/s)": 1.636047
    },
    {
      "acc": 0.67308769,
      "epoch": 0.8483003551496702,
      "grad_norm": 5.21875,
      "learning_rate": 6.636781122634804e-06,
      "loss": 1.57959232,
      "memory(GiB)": 107.26,
      "step": 33440,
      "train_speed(iter/s)": 1.636074
    },
    {
      "acc": 0.67521672,
      "epoch": 0.8484271943176053,
      "grad_norm": 4.90625,
      "learning_rate": 6.635790241080941e-06,
      "loss": 1.44894133,
      "memory(GiB)": 107.26,
      "step": 33445,
      "train_speed(iter/s)": 1.636102
    },
    {
      "acc": 0.65384202,
      "epoch": 0.8485540334855404,
      "grad_norm": 5.875,
      "learning_rate": 6.634799287577721e-06,
      "loss": 1.59544468,
      "memory(GiB)": 107.26,
      "step": 33450,
      "train_speed(iter/s)": 1.636128
    },
    {
      "acc": 0.66716232,
      "epoch": 0.8486808726534754,
      "grad_norm": 6.25,
      "learning_rate": 6.6338082621687286e-06,
      "loss": 1.53411179,
      "memory(GiB)": 107.26,
      "step": 33455,
      "train_speed(iter/s)": 1.636156
    },
    {
      "acc": 0.64660311,
      "epoch": 0.8488077118214105,
      "grad_norm": 8.75,
      "learning_rate": 6.6328171648975545e-06,
      "loss": 1.68349113,
      "memory(GiB)": 107.26,
      "step": 33460,
      "train_speed(iter/s)": 1.636183
    },
    {
      "acc": 0.66802964,
      "epoch": 0.8489345509893456,
      "grad_norm": 6.25,
      "learning_rate": 6.63182599580779e-06,
      "loss": 1.53537884,
      "memory(GiB)": 107.26,
      "step": 33465,
      "train_speed(iter/s)": 1.63621
    },
    {
      "acc": 0.66363921,
      "epoch": 0.8490613901572805,
      "grad_norm": 5.21875,
      "learning_rate": 6.630834754943036e-06,
      "loss": 1.58770275,
      "memory(GiB)": 107.26,
      "step": 33470,
      "train_speed(iter/s)": 1.636238
    },
    {
      "acc": 0.65645103,
      "epoch": 0.8491882293252156,
      "grad_norm": 6.5,
      "learning_rate": 6.629843442346886e-06,
      "loss": 1.6131958,
      "memory(GiB)": 107.26,
      "step": 33475,
      "train_speed(iter/s)": 1.636266
    },
    {
      "acc": 0.66149659,
      "epoch": 0.8493150684931506,
      "grad_norm": 6.59375,
      "learning_rate": 6.628852058062944e-06,
      "loss": 1.6268137,
      "memory(GiB)": 107.26,
      "step": 33480,
      "train_speed(iter/s)": 1.636292
    },
    {
      "acc": 0.63187261,
      "epoch": 0.8494419076610857,
      "grad_norm": 5.59375,
      "learning_rate": 6.627860602134818e-06,
      "loss": 1.74686012,
      "memory(GiB)": 107.26,
      "step": 33485,
      "train_speed(iter/s)": 1.636318
    },
    {
      "acc": 0.64392219,
      "epoch": 0.8495687468290208,
      "grad_norm": 4.90625,
      "learning_rate": 6.626869074606113e-06,
      "loss": 1.64276581,
      "memory(GiB)": 107.26,
      "step": 33490,
      "train_speed(iter/s)": 1.636346
    },
    {
      "acc": 0.63419681,
      "epoch": 0.8496955859969558,
      "grad_norm": 5.40625,
      "learning_rate": 6.625877475520445e-06,
      "loss": 1.67645416,
      "memory(GiB)": 107.26,
      "step": 33495,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.66426578,
      "epoch": 0.8498224251648909,
      "grad_norm": 6.09375,
      "learning_rate": 6.624885804921425e-06,
      "loss": 1.49502258,
      "memory(GiB)": 107.26,
      "step": 33500,
      "train_speed(iter/s)": 1.636399
    },
    {
      "acc": 0.65908689,
      "epoch": 0.849949264332826,
      "grad_norm": 6.90625,
      "learning_rate": 6.623894062852673e-06,
      "loss": 1.55287924,
      "memory(GiB)": 107.26,
      "step": 33505,
      "train_speed(iter/s)": 1.636424
    },
    {
      "acc": 0.67805767,
      "epoch": 0.850076103500761,
      "grad_norm": 4.96875,
      "learning_rate": 6.62290224935781e-06,
      "loss": 1.53019915,
      "memory(GiB)": 107.26,
      "step": 33510,
      "train_speed(iter/s)": 1.636451
    },
    {
      "acc": 0.65787125,
      "epoch": 0.8502029426686961,
      "grad_norm": 4.625,
      "learning_rate": 6.621910364480461e-06,
      "loss": 1.57635317,
      "memory(GiB)": 107.26,
      "step": 33515,
      "train_speed(iter/s)": 1.636478
    },
    {
      "acc": 0.65534821,
      "epoch": 0.8503297818366311,
      "grad_norm": 5.8125,
      "learning_rate": 6.620918408264252e-06,
      "loss": 1.66582336,
      "memory(GiB)": 107.26,
      "step": 33520,
      "train_speed(iter/s)": 1.636505
    },
    {
      "acc": 0.64076109,
      "epoch": 0.8504566210045662,
      "grad_norm": 5.28125,
      "learning_rate": 6.6199263807528136e-06,
      "loss": 1.69195633,
      "memory(GiB)": 107.26,
      "step": 33525,
      "train_speed(iter/s)": 1.636531
    },
    {
      "acc": 0.65869346,
      "epoch": 0.8505834601725013,
      "grad_norm": 5.90625,
      "learning_rate": 6.618934281989783e-06,
      "loss": 1.64132214,
      "memory(GiB)": 107.26,
      "step": 33530,
      "train_speed(iter/s)": 1.63656
    },
    {
      "acc": 0.65224018,
      "epoch": 0.8507102993404363,
      "grad_norm": 5.6875,
      "learning_rate": 6.6179421120187915e-06,
      "loss": 1.56616573,
      "memory(GiB)": 107.26,
      "step": 33535,
      "train_speed(iter/s)": 1.636586
    },
    {
      "acc": 0.67532496,
      "epoch": 0.8508371385083714,
      "grad_norm": 5.90625,
      "learning_rate": 6.616949870883486e-06,
      "loss": 1.51041489,
      "memory(GiB)": 107.26,
      "step": 33540,
      "train_speed(iter/s)": 1.636614
    },
    {
      "acc": 0.64911766,
      "epoch": 0.8509639776763065,
      "grad_norm": 6.21875,
      "learning_rate": 6.615957558627503e-06,
      "loss": 1.66333809,
      "memory(GiB)": 107.26,
      "step": 33545,
      "train_speed(iter/s)": 1.63664
    },
    {
      "acc": 0.65912933,
      "epoch": 0.8510908168442415,
      "grad_norm": 5.8125,
      "learning_rate": 6.6149651752944945e-06,
      "loss": 1.57878094,
      "memory(GiB)": 107.26,
      "step": 33550,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.64721432,
      "epoch": 0.8512176560121766,
      "grad_norm": 6.0,
      "learning_rate": 6.613972720928105e-06,
      "loss": 1.61661415,
      "memory(GiB)": 107.26,
      "step": 33555,
      "train_speed(iter/s)": 1.636696
    },
    {
      "acc": 0.65262809,
      "epoch": 0.8513444951801116,
      "grad_norm": 6.4375,
      "learning_rate": 6.61298019557199e-06,
      "loss": 1.57162457,
      "memory(GiB)": 107.26,
      "step": 33560,
      "train_speed(iter/s)": 1.636723
    },
    {
      "acc": 0.67586737,
      "epoch": 0.8514713343480467,
      "grad_norm": 4.90625,
      "learning_rate": 6.6119875992698045e-06,
      "loss": 1.52930984,
      "memory(GiB)": 107.26,
      "step": 33565,
      "train_speed(iter/s)": 1.636749
    },
    {
      "acc": 0.66140804,
      "epoch": 0.8515981735159818,
      "grad_norm": 5.8125,
      "learning_rate": 6.610994932065207e-06,
      "loss": 1.5717617,
      "memory(GiB)": 107.26,
      "step": 33570,
      "train_speed(iter/s)": 1.636776
    },
    {
      "acc": 0.6553977,
      "epoch": 0.8517250126839168,
      "grad_norm": 5.65625,
      "learning_rate": 6.610002194001861e-06,
      "loss": 1.62787571,
      "memory(GiB)": 107.26,
      "step": 33575,
      "train_speed(iter/s)": 1.636802
    },
    {
      "acc": 0.66403141,
      "epoch": 0.8518518518518519,
      "grad_norm": 6.6875,
      "learning_rate": 6.609009385123429e-06,
      "loss": 1.52281885,
      "memory(GiB)": 107.26,
      "step": 33580,
      "train_speed(iter/s)": 1.636829
    },
    {
      "acc": 0.66344795,
      "epoch": 0.851978691019787,
      "grad_norm": 6.0625,
      "learning_rate": 6.608016505473582e-06,
      "loss": 1.57456589,
      "memory(GiB)": 107.26,
      "step": 33585,
      "train_speed(iter/s)": 1.636856
    },
    {
      "acc": 0.65338573,
      "epoch": 0.852105530187722,
      "grad_norm": 6.21875,
      "learning_rate": 6.60702355509599e-06,
      "loss": 1.58700981,
      "memory(GiB)": 107.26,
      "step": 33590,
      "train_speed(iter/s)": 1.636882
    },
    {
      "acc": 0.65575476,
      "epoch": 0.852232369355657,
      "grad_norm": 5.28125,
      "learning_rate": 6.606030534034326e-06,
      "loss": 1.61918526,
      "memory(GiB)": 107.26,
      "step": 33595,
      "train_speed(iter/s)": 1.636907
    },
    {
      "acc": 0.64912591,
      "epoch": 0.852359208523592,
      "grad_norm": 5.8125,
      "learning_rate": 6.6050374423322685e-06,
      "loss": 1.67333641,
      "memory(GiB)": 107.26,
      "step": 33600,
      "train_speed(iter/s)": 1.636934
    },
    {
      "acc": 0.63515944,
      "epoch": 0.8524860476915271,
      "grad_norm": 6.15625,
      "learning_rate": 6.604044280033498e-06,
      "loss": 1.67019806,
      "memory(GiB)": 107.26,
      "step": 33605,
      "train_speed(iter/s)": 1.636962
    },
    {
      "acc": 0.6477654,
      "epoch": 0.8526128868594622,
      "grad_norm": 6.8125,
      "learning_rate": 6.6030510471817e-06,
      "loss": 1.66130009,
      "memory(GiB)": 107.26,
      "step": 33610,
      "train_speed(iter/s)": 1.636991
    },
    {
      "acc": 0.6302866,
      "epoch": 0.8527397260273972,
      "grad_norm": 5.09375,
      "learning_rate": 6.602057743820558e-06,
      "loss": 1.69997482,
      "memory(GiB)": 107.26,
      "step": 33615,
      "train_speed(iter/s)": 1.637018
    },
    {
      "acc": 0.66915846,
      "epoch": 0.8528665651953323,
      "grad_norm": 8.5625,
      "learning_rate": 6.601064369993766e-06,
      "loss": 1.52528439,
      "memory(GiB)": 107.26,
      "step": 33620,
      "train_speed(iter/s)": 1.637045
    },
    {
      "acc": 0.64886618,
      "epoch": 0.8529934043632674,
      "grad_norm": 5.4375,
      "learning_rate": 6.600070925745012e-06,
      "loss": 1.59449921,
      "memory(GiB)": 107.26,
      "step": 33625,
      "train_speed(iter/s)": 1.63707
    },
    {
      "acc": 0.65051441,
      "epoch": 0.8531202435312024,
      "grad_norm": 5.65625,
      "learning_rate": 6.599077411117998e-06,
      "loss": 1.63744965,
      "memory(GiB)": 107.26,
      "step": 33630,
      "train_speed(iter/s)": 1.637098
    },
    {
      "acc": 0.66494141,
      "epoch": 0.8532470826991375,
      "grad_norm": 6.1875,
      "learning_rate": 6.598083826156418e-06,
      "loss": 1.53433905,
      "memory(GiB)": 107.26,
      "step": 33635,
      "train_speed(iter/s)": 1.637126
    },
    {
      "acc": 0.66096401,
      "epoch": 0.8533739218670725,
      "grad_norm": 5.375,
      "learning_rate": 6.597090170903977e-06,
      "loss": 1.62079887,
      "memory(GiB)": 107.26,
      "step": 33640,
      "train_speed(iter/s)": 1.637155
    },
    {
      "acc": 0.65777836,
      "epoch": 0.8535007610350076,
      "grad_norm": 8.0,
      "learning_rate": 6.596096445404381e-06,
      "loss": 1.61259346,
      "memory(GiB)": 107.26,
      "step": 33645,
      "train_speed(iter/s)": 1.637182
    },
    {
      "acc": 0.65028338,
      "epoch": 0.8536276002029427,
      "grad_norm": 5.625,
      "learning_rate": 6.595102649701336e-06,
      "loss": 1.60424461,
      "memory(GiB)": 107.26,
      "step": 33650,
      "train_speed(iter/s)": 1.637209
    },
    {
      "acc": 0.65557952,
      "epoch": 0.8537544393708777,
      "grad_norm": 5.78125,
      "learning_rate": 6.5941087838385545e-06,
      "loss": 1.60531502,
      "memory(GiB)": 107.26,
      "step": 33655,
      "train_speed(iter/s)": 1.637235
    },
    {
      "acc": 0.66563034,
      "epoch": 0.8538812785388128,
      "grad_norm": 5.5,
      "learning_rate": 6.593114847859752e-06,
      "loss": 1.55379219,
      "memory(GiB)": 107.26,
      "step": 33660,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.65028367,
      "epoch": 0.8540081177067479,
      "grad_norm": 7.4375,
      "learning_rate": 6.592120841808646e-06,
      "loss": 1.65940437,
      "memory(GiB)": 107.26,
      "step": 33665,
      "train_speed(iter/s)": 1.637291
    },
    {
      "acc": 0.66156449,
      "epoch": 0.8541349568746829,
      "grad_norm": 5.15625,
      "learning_rate": 6.5911267657289564e-06,
      "loss": 1.57866669,
      "memory(GiB)": 107.26,
      "step": 33670,
      "train_speed(iter/s)": 1.637315
    },
    {
      "acc": 0.65063314,
      "epoch": 0.854261796042618,
      "grad_norm": 5.25,
      "learning_rate": 6.590132619664408e-06,
      "loss": 1.61251144,
      "memory(GiB)": 107.26,
      "step": 33675,
      "train_speed(iter/s)": 1.63734
    },
    {
      "acc": 0.66867943,
      "epoch": 0.854388635210553,
      "grad_norm": 5.03125,
      "learning_rate": 6.589138403658728e-06,
      "loss": 1.56566391,
      "memory(GiB)": 107.26,
      "step": 33680,
      "train_speed(iter/s)": 1.637367
    },
    {
      "acc": 0.64111919,
      "epoch": 0.8545154743784881,
      "grad_norm": 5.46875,
      "learning_rate": 6.588144117755645e-06,
      "loss": 1.66057682,
      "memory(GiB)": 107.26,
      "step": 33685,
      "train_speed(iter/s)": 1.637393
    },
    {
      "acc": 0.65913076,
      "epoch": 0.8546423135464232,
      "grad_norm": 9.625,
      "learning_rate": 6.5871497619988945e-06,
      "loss": 1.58504915,
      "memory(GiB)": 107.26,
      "step": 33690,
      "train_speed(iter/s)": 1.637418
    },
    {
      "acc": 0.64105988,
      "epoch": 0.8547691527143582,
      "grad_norm": 5.25,
      "learning_rate": 6.586155336432211e-06,
      "loss": 1.65872593,
      "memory(GiB)": 107.26,
      "step": 33695,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.66763258,
      "epoch": 0.8548959918822933,
      "grad_norm": 5.375,
      "learning_rate": 6.585160841099333e-06,
      "loss": 1.53311768,
      "memory(GiB)": 107.26,
      "step": 33700,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.6632225,
      "epoch": 0.8550228310502284,
      "grad_norm": 5.71875,
      "learning_rate": 6.584166276044005e-06,
      "loss": 1.58094616,
      "memory(GiB)": 107.26,
      "step": 33705,
      "train_speed(iter/s)": 1.637497
    },
    {
      "acc": 0.64657121,
      "epoch": 0.8551496702181633,
      "grad_norm": 4.53125,
      "learning_rate": 6.583171641309971e-06,
      "loss": 1.6748003,
      "memory(GiB)": 107.26,
      "step": 33710,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.65904293,
      "epoch": 0.8552765093860984,
      "grad_norm": 6.6875,
      "learning_rate": 6.58217693694098e-06,
      "loss": 1.60534363,
      "memory(GiB)": 107.26,
      "step": 33715,
      "train_speed(iter/s)": 1.637548
    },
    {
      "acc": 0.6524539,
      "epoch": 0.8554033485540334,
      "grad_norm": 7.15625,
      "learning_rate": 6.581182162980784e-06,
      "loss": 1.5884263,
      "memory(GiB)": 107.26,
      "step": 33720,
      "train_speed(iter/s)": 1.637573
    },
    {
      "acc": 0.64203291,
      "epoch": 0.8555301877219685,
      "grad_norm": 5.0,
      "learning_rate": 6.580187319473137e-06,
      "loss": 1.65365448,
      "memory(GiB)": 107.26,
      "step": 33725,
      "train_speed(iter/s)": 1.6376
    },
    {
      "acc": 0.63745112,
      "epoch": 0.8556570268899036,
      "grad_norm": 5.25,
      "learning_rate": 6.579192406461796e-06,
      "loss": 1.66923809,
      "memory(GiB)": 107.26,
      "step": 33730,
      "train_speed(iter/s)": 1.637626
    },
    {
      "acc": 0.6508481,
      "epoch": 0.8557838660578386,
      "grad_norm": 6.625,
      "learning_rate": 6.5781974239905225e-06,
      "loss": 1.66174297,
      "memory(GiB)": 107.26,
      "step": 33735,
      "train_speed(iter/s)": 1.637651
    },
    {
      "acc": 0.63802643,
      "epoch": 0.8559107052257737,
      "grad_norm": 7.75,
      "learning_rate": 6.57720237210308e-06,
      "loss": 1.65425167,
      "memory(GiB)": 107.26,
      "step": 33740,
      "train_speed(iter/s)": 1.637676
    },
    {
      "acc": 0.64974241,
      "epoch": 0.8560375443937088,
      "grad_norm": 6.03125,
      "learning_rate": 6.576207250843235e-06,
      "loss": 1.58731642,
      "memory(GiB)": 107.26,
      "step": 33745,
      "train_speed(iter/s)": 1.637703
    },
    {
      "acc": 0.67265229,
      "epoch": 0.8561643835616438,
      "grad_norm": 5.1875,
      "learning_rate": 6.575212060254759e-06,
      "loss": 1.56792126,
      "memory(GiB)": 107.26,
      "step": 33750,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.66145802,
      "epoch": 0.8562912227295789,
      "grad_norm": 5.1875,
      "learning_rate": 6.574216800381424e-06,
      "loss": 1.65183334,
      "memory(GiB)": 107.26,
      "step": 33755,
      "train_speed(iter/s)": 1.637754
    },
    {
      "acc": 0.66072688,
      "epoch": 0.8564180618975139,
      "grad_norm": 6.21875,
      "learning_rate": 6.573221471267005e-06,
      "loss": 1.56241426,
      "memory(GiB)": 107.26,
      "step": 33760,
      "train_speed(iter/s)": 1.637779
    },
    {
      "acc": 0.65756068,
      "epoch": 0.856544901065449,
      "grad_norm": 4.9375,
      "learning_rate": 6.572226072955281e-06,
      "loss": 1.59077196,
      "memory(GiB)": 107.26,
      "step": 33765,
      "train_speed(iter/s)": 1.637803
    },
    {
      "acc": 0.67466726,
      "epoch": 0.8566717402333841,
      "grad_norm": 6.3125,
      "learning_rate": 6.571230605490036e-06,
      "loss": 1.52986774,
      "memory(GiB)": 107.26,
      "step": 33770,
      "train_speed(iter/s)": 1.637831
    },
    {
      "acc": 0.67146111,
      "epoch": 0.8567985794013191,
      "grad_norm": 5.84375,
      "learning_rate": 6.570235068915053e-06,
      "loss": 1.5328475,
      "memory(GiB)": 107.26,
      "step": 33775,
      "train_speed(iter/s)": 1.637855
    },
    {
      "acc": 0.65428061,
      "epoch": 0.8569254185692542,
      "grad_norm": 5.46875,
      "learning_rate": 6.569239463274122e-06,
      "loss": 1.65695744,
      "memory(GiB)": 107.26,
      "step": 33780,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.64167366,
      "epoch": 0.8570522577371893,
      "grad_norm": 4.8125,
      "learning_rate": 6.568243788611033e-06,
      "loss": 1.58303642,
      "memory(GiB)": 107.26,
      "step": 33785,
      "train_speed(iter/s)": 1.637907
    },
    {
      "acc": 0.65935373,
      "epoch": 0.8571790969051243,
      "grad_norm": 5.125,
      "learning_rate": 6.56724804496958e-06,
      "loss": 1.59171257,
      "memory(GiB)": 107.26,
      "step": 33790,
      "train_speed(iter/s)": 1.637933
    },
    {
      "acc": 0.6521513,
      "epoch": 0.8573059360730594,
      "grad_norm": 4.8125,
      "learning_rate": 6.566252232393561e-06,
      "loss": 1.62133522,
      "memory(GiB)": 107.26,
      "step": 33795,
      "train_speed(iter/s)": 1.637961
    },
    {
      "acc": 0.66416121,
      "epoch": 0.8574327752409944,
      "grad_norm": 5.46875,
      "learning_rate": 6.565256350926777e-06,
      "loss": 1.59280167,
      "memory(GiB)": 107.26,
      "step": 33800,
      "train_speed(iter/s)": 1.637986
    },
    {
      "acc": 0.66706371,
      "epoch": 0.8575596144089295,
      "grad_norm": 4.6875,
      "learning_rate": 6.5642604006130286e-06,
      "loss": 1.53438721,
      "memory(GiB)": 107.26,
      "step": 33805,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.65187678,
      "epoch": 0.8576864535768646,
      "grad_norm": 4.875,
      "learning_rate": 6.563264381496124e-06,
      "loss": 1.69195251,
      "memory(GiB)": 107.26,
      "step": 33810,
      "train_speed(iter/s)": 1.638037
    },
    {
      "acc": 0.65640607,
      "epoch": 0.8578132927447996,
      "grad_norm": 11.0625,
      "learning_rate": 6.562268293619872e-06,
      "loss": 1.6875267,
      "memory(GiB)": 107.26,
      "step": 33815,
      "train_speed(iter/s)": 1.638064
    },
    {
      "acc": 0.65623112,
      "epoch": 0.8579401319127347,
      "grad_norm": 7.96875,
      "learning_rate": 6.561272137028089e-06,
      "loss": 1.61683464,
      "memory(GiB)": 107.26,
      "step": 33820,
      "train_speed(iter/s)": 1.63809
    },
    {
      "acc": 0.64545979,
      "epoch": 0.8580669710806698,
      "grad_norm": 5.46875,
      "learning_rate": 6.560275911764582e-06,
      "loss": 1.58695049,
      "memory(GiB)": 107.26,
      "step": 33825,
      "train_speed(iter/s)": 1.638116
    },
    {
      "acc": 0.66298537,
      "epoch": 0.8581938102486047,
      "grad_norm": 6.96875,
      "learning_rate": 6.5592796178731776e-06,
      "loss": 1.58266106,
      "memory(GiB)": 107.26,
      "step": 33830,
      "train_speed(iter/s)": 1.638142
    },
    {
      "acc": 0.63866143,
      "epoch": 0.8583206494165398,
      "grad_norm": 5.71875,
      "learning_rate": 6.5582832553976924e-06,
      "loss": 1.60564728,
      "memory(GiB)": 107.26,
      "step": 33835,
      "train_speed(iter/s)": 1.638167
    },
    {
      "acc": 0.6707767,
      "epoch": 0.8584474885844748,
      "grad_norm": 4.78125,
      "learning_rate": 6.557286824381955e-06,
      "loss": 1.55806322,
      "memory(GiB)": 107.26,
      "step": 33840,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.65880113,
      "epoch": 0.8585743277524099,
      "grad_norm": 7.21875,
      "learning_rate": 6.556290324869786e-06,
      "loss": 1.60772247,
      "memory(GiB)": 107.26,
      "step": 33845,
      "train_speed(iter/s)": 1.638219
    },
    {
      "acc": 0.66736221,
      "epoch": 0.858701166920345,
      "grad_norm": 5.78125,
      "learning_rate": 6.555293756905024e-06,
      "loss": 1.58186092,
      "memory(GiB)": 107.26,
      "step": 33850,
      "train_speed(iter/s)": 1.638246
    },
    {
      "acc": 0.66110554,
      "epoch": 0.85882800608828,
      "grad_norm": 5.78125,
      "learning_rate": 6.554297120531497e-06,
      "loss": 1.51177158,
      "memory(GiB)": 107.26,
      "step": 33855,
      "train_speed(iter/s)": 1.638272
    },
    {
      "acc": 0.64942493,
      "epoch": 0.8589548452562151,
      "grad_norm": 4.78125,
      "learning_rate": 6.553300415793042e-06,
      "loss": 1.63577919,
      "memory(GiB)": 107.26,
      "step": 33860,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.64790277,
      "epoch": 0.8590816844241502,
      "grad_norm": 5.40625,
      "learning_rate": 6.552303642733502e-06,
      "loss": 1.61803627,
      "memory(GiB)": 107.26,
      "step": 33865,
      "train_speed(iter/s)": 1.638325
    },
    {
      "acc": 0.6589128,
      "epoch": 0.8592085235920852,
      "grad_norm": 5.25,
      "learning_rate": 6.551306801396715e-06,
      "loss": 1.61670704,
      "memory(GiB)": 107.26,
      "step": 33870,
      "train_speed(iter/s)": 1.638351
    },
    {
      "acc": 0.65412321,
      "epoch": 0.8593353627600203,
      "grad_norm": 5.625,
      "learning_rate": 6.550309891826531e-06,
      "loss": 1.55272522,
      "memory(GiB)": 107.26,
      "step": 33875,
      "train_speed(iter/s)": 1.638378
    },
    {
      "acc": 0.6536561,
      "epoch": 0.8594622019279553,
      "grad_norm": 6.0,
      "learning_rate": 6.5493129140667955e-06,
      "loss": 1.57131338,
      "memory(GiB)": 107.26,
      "step": 33880,
      "train_speed(iter/s)": 1.638403
    },
    {
      "acc": 0.64003358,
      "epoch": 0.8595890410958904,
      "grad_norm": 6.5625,
      "learning_rate": 6.54831586816136e-06,
      "loss": 1.64107819,
      "memory(GiB)": 107.26,
      "step": 33885,
      "train_speed(iter/s)": 1.638432
    },
    {
      "acc": 0.66077843,
      "epoch": 0.8597158802638255,
      "grad_norm": 5.53125,
      "learning_rate": 6.54731875415408e-06,
      "loss": 1.59336863,
      "memory(GiB)": 107.26,
      "step": 33890,
      "train_speed(iter/s)": 1.638458
    },
    {
      "acc": 0.66765656,
      "epoch": 0.8598427194317605,
      "grad_norm": 4.75,
      "learning_rate": 6.546321572088814e-06,
      "loss": 1.55714464,
      "memory(GiB)": 107.26,
      "step": 33895,
      "train_speed(iter/s)": 1.638485
    },
    {
      "acc": 0.65034971,
      "epoch": 0.8599695585996956,
      "grad_norm": 5.5625,
      "learning_rate": 6.545324322009421e-06,
      "loss": 1.63762589,
      "memory(GiB)": 107.26,
      "step": 33900,
      "train_speed(iter/s)": 1.638512
    },
    {
      "acc": 0.6678153,
      "epoch": 0.8600963977676307,
      "grad_norm": 6.03125,
      "learning_rate": 6.544327003959765e-06,
      "loss": 1.56251316,
      "memory(GiB)": 107.26,
      "step": 33905,
      "train_speed(iter/s)": 1.638539
    },
    {
      "acc": 0.66459317,
      "epoch": 0.8602232369355657,
      "grad_norm": 5.5625,
      "learning_rate": 6.543329617983713e-06,
      "loss": 1.58747234,
      "memory(GiB)": 107.26,
      "step": 33910,
      "train_speed(iter/s)": 1.638565
    },
    {
      "acc": 0.66931686,
      "epoch": 0.8603500761035008,
      "grad_norm": 5.65625,
      "learning_rate": 6.5423321641251316e-06,
      "loss": 1.61592941,
      "memory(GiB)": 107.26,
      "step": 33915,
      "train_speed(iter/s)": 1.638593
    },
    {
      "acc": 0.64662681,
      "epoch": 0.8604769152714358,
      "grad_norm": 6.28125,
      "learning_rate": 6.541334642427898e-06,
      "loss": 1.63952312,
      "memory(GiB)": 107.26,
      "step": 33920,
      "train_speed(iter/s)": 1.638619
    },
    {
      "acc": 0.65676203,
      "epoch": 0.8606037544393709,
      "grad_norm": 4.8125,
      "learning_rate": 6.540337052935884e-06,
      "loss": 1.58259211,
      "memory(GiB)": 107.26,
      "step": 33925,
      "train_speed(iter/s)": 1.638645
    },
    {
      "acc": 0.65153465,
      "epoch": 0.860730593607306,
      "grad_norm": 5.75,
      "learning_rate": 6.53933939569297e-06,
      "loss": 1.62081699,
      "memory(GiB)": 107.26,
      "step": 33930,
      "train_speed(iter/s)": 1.638673
    },
    {
      "acc": 0.65377936,
      "epoch": 0.860857432775241,
      "grad_norm": 5.5625,
      "learning_rate": 6.538341670743037e-06,
      "loss": 1.61512146,
      "memory(GiB)": 107.26,
      "step": 33935,
      "train_speed(iter/s)": 1.638698
    },
    {
      "acc": 0.66160526,
      "epoch": 0.8609842719431761,
      "grad_norm": 6.0625,
      "learning_rate": 6.537343878129969e-06,
      "loss": 1.57845831,
      "memory(GiB)": 107.26,
      "step": 33940,
      "train_speed(iter/s)": 1.638725
    },
    {
      "acc": 0.64678173,
      "epoch": 0.8611111111111112,
      "grad_norm": 6.46875,
      "learning_rate": 6.5363460178976524e-06,
      "loss": 1.64689846,
      "memory(GiB)": 107.26,
      "step": 33945,
      "train_speed(iter/s)": 1.63875
    },
    {
      "acc": 0.64583454,
      "epoch": 0.8612379502790461,
      "grad_norm": 6.1875,
      "learning_rate": 6.53534809008998e-06,
      "loss": 1.63763504,
      "memory(GiB)": 107.26,
      "step": 33950,
      "train_speed(iter/s)": 1.638776
    },
    {
      "acc": 0.655334,
      "epoch": 0.8613647894469812,
      "grad_norm": 4.84375,
      "learning_rate": 6.534350094750843e-06,
      "loss": 1.62189484,
      "memory(GiB)": 107.26,
      "step": 33955,
      "train_speed(iter/s)": 1.6388
    },
    {
      "acc": 0.64833665,
      "epoch": 0.8614916286149162,
      "grad_norm": 5.71875,
      "learning_rate": 6.5333520319241385e-06,
      "loss": 1.62575493,
      "memory(GiB)": 107.26,
      "step": 33960,
      "train_speed(iter/s)": 1.638826
    },
    {
      "acc": 0.63960791,
      "epoch": 0.8616184677828513,
      "grad_norm": 6.25,
      "learning_rate": 6.532353901653765e-06,
      "loss": 1.66806908,
      "memory(GiB)": 107.26,
      "step": 33965,
      "train_speed(iter/s)": 1.638854
    },
    {
      "acc": 0.64926233,
      "epoch": 0.8617453069507864,
      "grad_norm": 5.9375,
      "learning_rate": 6.531355703983627e-06,
      "loss": 1.66905327,
      "memory(GiB)": 107.26,
      "step": 33970,
      "train_speed(iter/s)": 1.638879
    },
    {
      "acc": 0.65363922,
      "epoch": 0.8618721461187214,
      "grad_norm": 5.84375,
      "learning_rate": 6.530357438957626e-06,
      "loss": 1.65205574,
      "memory(GiB)": 107.26,
      "step": 33975,
      "train_speed(iter/s)": 1.638906
    },
    {
      "acc": 0.65834751,
      "epoch": 0.8619989852866565,
      "grad_norm": 6.125,
      "learning_rate": 6.529359106619675e-06,
      "loss": 1.5987874,
      "memory(GiB)": 107.26,
      "step": 33980,
      "train_speed(iter/s)": 1.638933
    },
    {
      "acc": 0.64411855,
      "epoch": 0.8621258244545916,
      "grad_norm": 5.0,
      "learning_rate": 6.528360707013681e-06,
      "loss": 1.62621117,
      "memory(GiB)": 107.26,
      "step": 33985,
      "train_speed(iter/s)": 1.638958
    },
    {
      "acc": 0.6748682,
      "epoch": 0.8622526636225266,
      "grad_norm": 7.09375,
      "learning_rate": 6.52736224018356e-06,
      "loss": 1.64302311,
      "memory(GiB)": 107.26,
      "step": 33990,
      "train_speed(iter/s)": 1.638985
    },
    {
      "acc": 0.65918121,
      "epoch": 0.8623795027904617,
      "grad_norm": 4.90625,
      "learning_rate": 6.526363706173227e-06,
      "loss": 1.6388485,
      "memory(GiB)": 107.26,
      "step": 33995,
      "train_speed(iter/s)": 1.639009
    },
    {
      "acc": 0.67118726,
      "epoch": 0.8625063419583967,
      "grad_norm": 5.71875,
      "learning_rate": 6.525365105026605e-06,
      "loss": 1.47325811,
      "memory(GiB)": 107.26,
      "step": 34000,
      "train_speed(iter/s)": 1.639035
    },
    {
      "epoch": 0.8625063419583967,
      "eval_acc": 0.6457969707031536,
      "eval_loss": 1.5752389430999756,
      "eval_runtime": 58.1348,
      "eval_samples_per_second": 109.573,
      "eval_steps_per_second": 27.402,
      "step": 34000
    },
    {
      "acc": 0.6792944,
      "epoch": 0.8626331811263318,
      "grad_norm": 6.03125,
      "learning_rate": 6.524366436787615e-06,
      "loss": 1.48697662,
      "memory(GiB)": 107.26,
      "step": 34005,
      "train_speed(iter/s)": 1.634143
    },
    {
      "acc": 0.65556364,
      "epoch": 0.8627600202942669,
      "grad_norm": 5.25,
      "learning_rate": 6.523367701500183e-06,
      "loss": 1.60516357,
      "memory(GiB)": 107.26,
      "step": 34010,
      "train_speed(iter/s)": 1.634165
    },
    {
      "acc": 0.65507517,
      "epoch": 0.8628868594622019,
      "grad_norm": 5.03125,
      "learning_rate": 6.5223688992082375e-06,
      "loss": 1.56654625,
      "memory(GiB)": 107.26,
      "step": 34015,
      "train_speed(iter/s)": 1.634189
    },
    {
      "acc": 0.64477711,
      "epoch": 0.863013698630137,
      "grad_norm": 6.46875,
      "learning_rate": 6.521370029955713e-06,
      "loss": 1.65110779,
      "memory(GiB)": 107.26,
      "step": 34020,
      "train_speed(iter/s)": 1.634213
    },
    {
      "acc": 0.64231291,
      "epoch": 0.8631405377980721,
      "grad_norm": 5.125,
      "learning_rate": 6.520371093786541e-06,
      "loss": 1.69971352,
      "memory(GiB)": 107.26,
      "step": 34025,
      "train_speed(iter/s)": 1.634237
    },
    {
      "acc": 0.67796497,
      "epoch": 0.8632673769660071,
      "grad_norm": 6.71875,
      "learning_rate": 6.51937209074466e-06,
      "loss": 1.56233158,
      "memory(GiB)": 107.26,
      "step": 34030,
      "train_speed(iter/s)": 1.63426
    },
    {
      "acc": 0.65515461,
      "epoch": 0.8633942161339422,
      "grad_norm": 5.59375,
      "learning_rate": 6.51837302087401e-06,
      "loss": 1.5657732,
      "memory(GiB)": 107.26,
      "step": 34035,
      "train_speed(iter/s)": 1.634287
    },
    {
      "acc": 0.64916468,
      "epoch": 0.8635210553018772,
      "grad_norm": 7.0625,
      "learning_rate": 6.517373884218539e-06,
      "loss": 1.67203236,
      "memory(GiB)": 107.26,
      "step": 34040,
      "train_speed(iter/s)": 1.634311
    },
    {
      "acc": 0.64695168,
      "epoch": 0.8636478944698123,
      "grad_norm": 6.1875,
      "learning_rate": 6.5163746808221865e-06,
      "loss": 1.65379772,
      "memory(GiB)": 107.26,
      "step": 34045,
      "train_speed(iter/s)": 1.634336
    },
    {
      "acc": 0.63686333,
      "epoch": 0.8637747336377474,
      "grad_norm": 4.6875,
      "learning_rate": 6.515375410728907e-06,
      "loss": 1.58830204,
      "memory(GiB)": 107.26,
      "step": 34050,
      "train_speed(iter/s)": 1.634363
    },
    {
      "acc": 0.65573053,
      "epoch": 0.8639015728056824,
      "grad_norm": 4.65625,
      "learning_rate": 6.51437607398265e-06,
      "loss": 1.62066689,
      "memory(GiB)": 107.26,
      "step": 34055,
      "train_speed(iter/s)": 1.634388
    },
    {
      "acc": 0.67124023,
      "epoch": 0.8640284119736175,
      "grad_norm": 6.875,
      "learning_rate": 6.513376670627374e-06,
      "loss": 1.54675541,
      "memory(GiB)": 107.26,
      "step": 34060,
      "train_speed(iter/s)": 1.634412
    },
    {
      "acc": 0.64677529,
      "epoch": 0.8641552511415526,
      "grad_norm": 8.125,
      "learning_rate": 6.512377200707033e-06,
      "loss": 1.640695,
      "memory(GiB)": 107.26,
      "step": 34065,
      "train_speed(iter/s)": 1.634439
    },
    {
      "acc": 0.6600338,
      "epoch": 0.8642820903094875,
      "grad_norm": 6.59375,
      "learning_rate": 6.511377664265591e-06,
      "loss": 1.56802282,
      "memory(GiB)": 107.26,
      "step": 34070,
      "train_speed(iter/s)": 1.634466
    },
    {
      "acc": 0.65844598,
      "epoch": 0.8644089294774226,
      "grad_norm": 5.4375,
      "learning_rate": 6.510378061347013e-06,
      "loss": 1.60723152,
      "memory(GiB)": 107.26,
      "step": 34075,
      "train_speed(iter/s)": 1.634491
    },
    {
      "acc": 0.65013862,
      "epoch": 0.8645357686453576,
      "grad_norm": 7.875,
      "learning_rate": 6.509378391995264e-06,
      "loss": 1.6730938,
      "memory(GiB)": 107.26,
      "step": 34080,
      "train_speed(iter/s)": 1.634516
    },
    {
      "acc": 0.65103164,
      "epoch": 0.8646626078132927,
      "grad_norm": 4.71875,
      "learning_rate": 6.508378656254314e-06,
      "loss": 1.69251461,
      "memory(GiB)": 107.26,
      "step": 34085,
      "train_speed(iter/s)": 1.634541
    },
    {
      "acc": 0.66417913,
      "epoch": 0.8647894469812278,
      "grad_norm": 5.4375,
      "learning_rate": 6.507378854168136e-06,
      "loss": 1.55832195,
      "memory(GiB)": 107.26,
      "step": 34090,
      "train_speed(iter/s)": 1.634566
    },
    {
      "acc": 0.64776597,
      "epoch": 0.8649162861491628,
      "grad_norm": 6.09375,
      "learning_rate": 6.506378985780707e-06,
      "loss": 1.60819626,
      "memory(GiB)": 107.26,
      "step": 34095,
      "train_speed(iter/s)": 1.63459
    },
    {
      "acc": 0.66627274,
      "epoch": 0.8650431253170979,
      "grad_norm": 5.46875,
      "learning_rate": 6.505379051136004e-06,
      "loss": 1.57564068,
      "memory(GiB)": 107.26,
      "step": 34100,
      "train_speed(iter/s)": 1.634617
    },
    {
      "acc": 0.66107988,
      "epoch": 0.865169964485033,
      "grad_norm": 5.5,
      "learning_rate": 6.504379050278009e-06,
      "loss": 1.58241034,
      "memory(GiB)": 107.26,
      "step": 34105,
      "train_speed(iter/s)": 1.634641
    },
    {
      "acc": 0.64033117,
      "epoch": 0.865296803652968,
      "grad_norm": 6.21875,
      "learning_rate": 6.503378983250707e-06,
      "loss": 1.72999649,
      "memory(GiB)": 107.26,
      "step": 34110,
      "train_speed(iter/s)": 1.634668
    },
    {
      "acc": 0.65916958,
      "epoch": 0.8654236428209031,
      "grad_norm": 9.0,
      "learning_rate": 6.5023788500980855e-06,
      "loss": 1.59703093,
      "memory(GiB)": 107.26,
      "step": 34115,
      "train_speed(iter/s)": 1.634694
    },
    {
      "acc": 0.64947586,
      "epoch": 0.8655504819888381,
      "grad_norm": 5.3125,
      "learning_rate": 6.501378650864135e-06,
      "loss": 1.64080048,
      "memory(GiB)": 107.26,
      "step": 34120,
      "train_speed(iter/s)": 1.634719
    },
    {
      "acc": 0.65534706,
      "epoch": 0.8656773211567732,
      "grad_norm": 5.46875,
      "learning_rate": 6.500378385592847e-06,
      "loss": 1.60876083,
      "memory(GiB)": 107.26,
      "step": 34125,
      "train_speed(iter/s)": 1.634746
    },
    {
      "acc": 0.66416488,
      "epoch": 0.8658041603247083,
      "grad_norm": 7.0,
      "learning_rate": 6.49937805432822e-06,
      "loss": 1.58953152,
      "memory(GiB)": 107.26,
      "step": 34130,
      "train_speed(iter/s)": 1.634772
    },
    {
      "acc": 0.66077566,
      "epoch": 0.8659309994926433,
      "grad_norm": 6.03125,
      "learning_rate": 6.498377657114251e-06,
      "loss": 1.55426388,
      "memory(GiB)": 107.26,
      "step": 34135,
      "train_speed(iter/s)": 1.634798
    },
    {
      "acc": 0.66569037,
      "epoch": 0.8660578386605784,
      "grad_norm": 5.59375,
      "learning_rate": 6.497377193994944e-06,
      "loss": 1.5657814,
      "memory(GiB)": 107.26,
      "step": 34140,
      "train_speed(iter/s)": 1.634823
    },
    {
      "acc": 0.66869922,
      "epoch": 0.8661846778285135,
      "grad_norm": 5.625,
      "learning_rate": 6.496376665014301e-06,
      "loss": 1.5821269,
      "memory(GiB)": 107.26,
      "step": 34145,
      "train_speed(iter/s)": 1.634849
    },
    {
      "acc": 0.64816437,
      "epoch": 0.8663115169964485,
      "grad_norm": 4.71875,
      "learning_rate": 6.4953760702163325e-06,
      "loss": 1.65185356,
      "memory(GiB)": 107.26,
      "step": 34150,
      "train_speed(iter/s)": 1.634874
    },
    {
      "acc": 0.65632424,
      "epoch": 0.8664383561643836,
      "grad_norm": 4.625,
      "learning_rate": 6.494375409645049e-06,
      "loss": 1.57895212,
      "memory(GiB)": 107.26,
      "step": 34155,
      "train_speed(iter/s)": 1.634899
    },
    {
      "acc": 0.65427155,
      "epoch": 0.8665651953323186,
      "grad_norm": 5.71875,
      "learning_rate": 6.493374683344462e-06,
      "loss": 1.59411678,
      "memory(GiB)": 107.26,
      "step": 34160,
      "train_speed(iter/s)": 1.634921
    },
    {
      "acc": 0.66368279,
      "epoch": 0.8666920345002537,
      "grad_norm": 7.53125,
      "learning_rate": 6.492373891358589e-06,
      "loss": 1.54134436,
      "memory(GiB)": 107.26,
      "step": 34165,
      "train_speed(iter/s)": 1.634948
    },
    {
      "acc": 0.65509548,
      "epoch": 0.8668188736681888,
      "grad_norm": 7.0625,
      "learning_rate": 6.4913730337314495e-06,
      "loss": 1.60821648,
      "memory(GiB)": 107.26,
      "step": 34170,
      "train_speed(iter/s)": 1.634973
    },
    {
      "acc": 0.6473949,
      "epoch": 0.8669457128361238,
      "grad_norm": 4.6875,
      "learning_rate": 6.490372110507066e-06,
      "loss": 1.65505047,
      "memory(GiB)": 107.26,
      "step": 34175,
      "train_speed(iter/s)": 1.634998
    },
    {
      "acc": 0.66014204,
      "epoch": 0.8670725520040589,
      "grad_norm": 5.46875,
      "learning_rate": 6.489371121729462e-06,
      "loss": 1.57256222,
      "memory(GiB)": 107.26,
      "step": 34180,
      "train_speed(iter/s)": 1.635025
    },
    {
      "acc": 0.65014925,
      "epoch": 0.867199391171994,
      "grad_norm": 5.96875,
      "learning_rate": 6.4883700674426666e-06,
      "loss": 1.57869263,
      "memory(GiB)": 107.26,
      "step": 34185,
      "train_speed(iter/s)": 1.635051
    },
    {
      "acc": 0.65479174,
      "epoch": 0.867326230339929,
      "grad_norm": 5.3125,
      "learning_rate": 6.4873689476907105e-06,
      "loss": 1.59951153,
      "memory(GiB)": 107.26,
      "step": 34190,
      "train_speed(iter/s)": 1.635075
    },
    {
      "acc": 0.65862598,
      "epoch": 0.867453069507864,
      "grad_norm": 7.09375,
      "learning_rate": 6.486367762517628e-06,
      "loss": 1.62387295,
      "memory(GiB)": 107.26,
      "step": 34195,
      "train_speed(iter/s)": 1.635094
    },
    {
      "acc": 0.64710178,
      "epoch": 0.867579908675799,
      "grad_norm": 5.1875,
      "learning_rate": 6.4853665119674556e-06,
      "loss": 1.6497982,
      "memory(GiB)": 107.26,
      "step": 34200,
      "train_speed(iter/s)": 1.63512
    },
    {
      "acc": 0.64454632,
      "epoch": 0.8677067478437341,
      "grad_norm": 4.65625,
      "learning_rate": 6.484365196084231e-06,
      "loss": 1.58651199,
      "memory(GiB)": 107.26,
      "step": 34205,
      "train_speed(iter/s)": 1.635146
    },
    {
      "acc": 0.65895886,
      "epoch": 0.8678335870116692,
      "grad_norm": 4.90625,
      "learning_rate": 6.4833638149119985e-06,
      "loss": 1.62382412,
      "memory(GiB)": 107.26,
      "step": 34210,
      "train_speed(iter/s)": 1.63517
    },
    {
      "acc": 0.65482121,
      "epoch": 0.8679604261796042,
      "grad_norm": 6.34375,
      "learning_rate": 6.4823623684948034e-06,
      "loss": 1.62744713,
      "memory(GiB)": 107.26,
      "step": 34215,
      "train_speed(iter/s)": 1.635196
    },
    {
      "acc": 0.66440773,
      "epoch": 0.8680872653475393,
      "grad_norm": 5.59375,
      "learning_rate": 6.4813608568766924e-06,
      "loss": 1.64382381,
      "memory(GiB)": 107.26,
      "step": 34220,
      "train_speed(iter/s)": 1.635221
    },
    {
      "acc": 0.65206661,
      "epoch": 0.8682141045154744,
      "grad_norm": 5.84375,
      "learning_rate": 6.480359280101717e-06,
      "loss": 1.59996347,
      "memory(GiB)": 107.26,
      "step": 34225,
      "train_speed(iter/s)": 1.635247
    },
    {
      "acc": 0.65976686,
      "epoch": 0.8683409436834094,
      "grad_norm": 5.71875,
      "learning_rate": 6.479357638213931e-06,
      "loss": 1.54963942,
      "memory(GiB)": 107.26,
      "step": 34230,
      "train_speed(iter/s)": 1.635271
    },
    {
      "acc": 0.65868626,
      "epoch": 0.8684677828513445,
      "grad_norm": 6.0625,
      "learning_rate": 6.478355931257392e-06,
      "loss": 1.59762096,
      "memory(GiB)": 107.26,
      "step": 34235,
      "train_speed(iter/s)": 1.635296
    },
    {
      "acc": 0.66919794,
      "epoch": 0.8685946220192795,
      "grad_norm": 5.75,
      "learning_rate": 6.477354159276158e-06,
      "loss": 1.55229073,
      "memory(GiB)": 107.26,
      "step": 34240,
      "train_speed(iter/s)": 1.635323
    },
    {
      "acc": 0.64872255,
      "epoch": 0.8687214611872146,
      "grad_norm": 5.46875,
      "learning_rate": 6.476352322314292e-06,
      "loss": 1.61540604,
      "memory(GiB)": 107.26,
      "step": 34245,
      "train_speed(iter/s)": 1.63535
    },
    {
      "acc": 0.64695129,
      "epoch": 0.8688483003551497,
      "grad_norm": 6.0,
      "learning_rate": 6.47535042041586e-06,
      "loss": 1.66697693,
      "memory(GiB)": 107.26,
      "step": 34250,
      "train_speed(iter/s)": 1.635376
    },
    {
      "acc": 0.63026171,
      "epoch": 0.8689751395230847,
      "grad_norm": 5.21875,
      "learning_rate": 6.474348453624929e-06,
      "loss": 1.70388355,
      "memory(GiB)": 107.26,
      "step": 34255,
      "train_speed(iter/s)": 1.635403
    },
    {
      "acc": 0.6526309,
      "epoch": 0.8691019786910198,
      "grad_norm": 7.0,
      "learning_rate": 6.473346421985571e-06,
      "loss": 1.63642807,
      "memory(GiB)": 107.26,
      "step": 34260,
      "train_speed(iter/s)": 1.635427
    },
    {
      "acc": 0.6500998,
      "epoch": 0.8692288178589549,
      "grad_norm": 6.21875,
      "learning_rate": 6.472344325541859e-06,
      "loss": 1.65020237,
      "memory(GiB)": 107.26,
      "step": 34265,
      "train_speed(iter/s)": 1.635453
    },
    {
      "acc": 0.64811931,
      "epoch": 0.8693556570268899,
      "grad_norm": 5.1875,
      "learning_rate": 6.4713421643378715e-06,
      "loss": 1.65410061,
      "memory(GiB)": 107.26,
      "step": 34270,
      "train_speed(iter/s)": 1.635479
    },
    {
      "acc": 0.64889007,
      "epoch": 0.869482496194825,
      "grad_norm": 5.90625,
      "learning_rate": 6.470339938417685e-06,
      "loss": 1.57758446,
      "memory(GiB)": 107.26,
      "step": 34275,
      "train_speed(iter/s)": 1.635505
    },
    {
      "acc": 0.67047734,
      "epoch": 0.86960933536276,
      "grad_norm": 5.875,
      "learning_rate": 6.469337647825384e-06,
      "loss": 1.56917114,
      "memory(GiB)": 107.26,
      "step": 34280,
      "train_speed(iter/s)": 1.63553
    },
    {
      "acc": 0.64510756,
      "epoch": 0.8697361745306951,
      "grad_norm": 9.625,
      "learning_rate": 6.468335292605053e-06,
      "loss": 1.61944275,
      "memory(GiB)": 107.26,
      "step": 34285,
      "train_speed(iter/s)": 1.635556
    },
    {
      "acc": 0.67003632,
      "epoch": 0.8698630136986302,
      "grad_norm": 5.78125,
      "learning_rate": 6.467332872800779e-06,
      "loss": 1.51040487,
      "memory(GiB)": 107.26,
      "step": 34290,
      "train_speed(iter/s)": 1.635582
    },
    {
      "acc": 0.65759735,
      "epoch": 0.8699898528665652,
      "grad_norm": 5.375,
      "learning_rate": 6.466330388456655e-06,
      "loss": 1.61880112,
      "memory(GiB)": 107.26,
      "step": 34295,
      "train_speed(iter/s)": 1.635608
    },
    {
      "acc": 0.64413924,
      "epoch": 0.8701166920345003,
      "grad_norm": 7.28125,
      "learning_rate": 6.465327839616774e-06,
      "loss": 1.654953,
      "memory(GiB)": 107.26,
      "step": 34300,
      "train_speed(iter/s)": 1.635632
    },
    {
      "acc": 0.66525888,
      "epoch": 0.8702435312024354,
      "grad_norm": 6.09375,
      "learning_rate": 6.464325226325232e-06,
      "loss": 1.59572163,
      "memory(GiB)": 107.26,
      "step": 34305,
      "train_speed(iter/s)": 1.635656
    },
    {
      "acc": 0.64500017,
      "epoch": 0.8703703703703703,
      "grad_norm": 7.15625,
      "learning_rate": 6.46332254862613e-06,
      "loss": 1.67218018,
      "memory(GiB)": 107.26,
      "step": 34310,
      "train_speed(iter/s)": 1.635682
    },
    {
      "acc": 0.65909228,
      "epoch": 0.8704972095383054,
      "grad_norm": 4.65625,
      "learning_rate": 6.462319806563568e-06,
      "loss": 1.60085297,
      "memory(GiB)": 107.26,
      "step": 34315,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.64792891,
      "epoch": 0.8706240487062404,
      "grad_norm": 5.75,
      "learning_rate": 6.461317000181653e-06,
      "loss": 1.5514185,
      "memory(GiB)": 107.26,
      "step": 34320,
      "train_speed(iter/s)": 1.635731
    },
    {
      "acc": 0.65251832,
      "epoch": 0.8707508878741755,
      "grad_norm": 6.4375,
      "learning_rate": 6.460314129524491e-06,
      "loss": 1.65665169,
      "memory(GiB)": 107.26,
      "step": 34325,
      "train_speed(iter/s)": 1.635755
    },
    {
      "acc": 0.65071449,
      "epoch": 0.8708777270421106,
      "grad_norm": 5.0625,
      "learning_rate": 6.4593111946361945e-06,
      "loss": 1.60932198,
      "memory(GiB)": 107.26,
      "step": 34330,
      "train_speed(iter/s)": 1.635781
    },
    {
      "acc": 0.64631996,
      "epoch": 0.8710045662100456,
      "grad_norm": 5.28125,
      "learning_rate": 6.458308195560874e-06,
      "loss": 1.6017498,
      "memory(GiB)": 107.26,
      "step": 34335,
      "train_speed(iter/s)": 1.635808
    },
    {
      "acc": 0.64003296,
      "epoch": 0.8711314053779807,
      "grad_norm": 6.875,
      "learning_rate": 6.4573051323426515e-06,
      "loss": 1.66052971,
      "memory(GiB)": 107.26,
      "step": 34340,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.65305729,
      "epoch": 0.8712582445459158,
      "grad_norm": 6.875,
      "learning_rate": 6.456302005025641e-06,
      "loss": 1.66724777,
      "memory(GiB)": 107.26,
      "step": 34345,
      "train_speed(iter/s)": 1.635858
    },
    {
      "acc": 0.6336185,
      "epoch": 0.8713850837138508,
      "grad_norm": 6.65625,
      "learning_rate": 6.4552988136539675e-06,
      "loss": 1.67614632,
      "memory(GiB)": 107.26,
      "step": 34350,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.66357269,
      "epoch": 0.8715119228817859,
      "grad_norm": 6.28125,
      "learning_rate": 6.454295558271752e-06,
      "loss": 1.56703577,
      "memory(GiB)": 107.26,
      "step": 34355,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.65190077,
      "epoch": 0.8716387620497209,
      "grad_norm": 5.375,
      "learning_rate": 6.4532922389231275e-06,
      "loss": 1.57342186,
      "memory(GiB)": 107.26,
      "step": 34360,
      "train_speed(iter/s)": 1.635933
    },
    {
      "acc": 0.66487818,
      "epoch": 0.871765601217656,
      "grad_norm": 5.40625,
      "learning_rate": 6.452288855652222e-06,
      "loss": 1.56768627,
      "memory(GiB)": 107.26,
      "step": 34365,
      "train_speed(iter/s)": 1.63596
    },
    {
      "acc": 0.65949831,
      "epoch": 0.8718924403855911,
      "grad_norm": 4.90625,
      "learning_rate": 6.451285408503167e-06,
      "loss": 1.56575298,
      "memory(GiB)": 107.26,
      "step": 34370,
      "train_speed(iter/s)": 1.635984
    },
    {
      "acc": 0.64705176,
      "epoch": 0.8720192795535261,
      "grad_norm": 6.0,
      "learning_rate": 6.450281897520102e-06,
      "loss": 1.64136944,
      "memory(GiB)": 107.26,
      "step": 34375,
      "train_speed(iter/s)": 1.636008
    },
    {
      "acc": 0.66802015,
      "epoch": 0.8721461187214612,
      "grad_norm": 5.5625,
      "learning_rate": 6.449278322747164e-06,
      "loss": 1.59115458,
      "memory(GiB)": 107.26,
      "step": 34380,
      "train_speed(iter/s)": 1.636032
    },
    {
      "acc": 0.6672204,
      "epoch": 0.8722729578893963,
      "grad_norm": 5.78125,
      "learning_rate": 6.448274684228494e-06,
      "loss": 1.6125803,
      "memory(GiB)": 107.26,
      "step": 34385,
      "train_speed(iter/s)": 1.636057
    },
    {
      "acc": 0.67955103,
      "epoch": 0.8723997970573313,
      "grad_norm": 6.5,
      "learning_rate": 6.447270982008237e-06,
      "loss": 1.52753544,
      "memory(GiB)": 107.26,
      "step": 34390,
      "train_speed(iter/s)": 1.636081
    },
    {
      "acc": 0.65122099,
      "epoch": 0.8725266362252664,
      "grad_norm": 6.0,
      "learning_rate": 6.446267216130541e-06,
      "loss": 1.63484802,
      "memory(GiB)": 107.26,
      "step": 34395,
      "train_speed(iter/s)": 1.636106
    },
    {
      "acc": 0.65745192,
      "epoch": 0.8726534753932014,
      "grad_norm": 5.125,
      "learning_rate": 6.4452633866395555e-06,
      "loss": 1.59543476,
      "memory(GiB)": 107.26,
      "step": 34400,
      "train_speed(iter/s)": 1.636133
    },
    {
      "acc": 0.66621742,
      "epoch": 0.8727803145611365,
      "grad_norm": 4.875,
      "learning_rate": 6.444259493579433e-06,
      "loss": 1.58463697,
      "memory(GiB)": 107.26,
      "step": 34405,
      "train_speed(iter/s)": 1.63616
    },
    {
      "acc": 0.66239519,
      "epoch": 0.8729071537290716,
      "grad_norm": 5.875,
      "learning_rate": 6.443255536994331e-06,
      "loss": 1.58281994,
      "memory(GiB)": 107.26,
      "step": 34410,
      "train_speed(iter/s)": 1.636186
    },
    {
      "acc": 0.66718121,
      "epoch": 0.8730339928970066,
      "grad_norm": 5.75,
      "learning_rate": 6.442251516928406e-06,
      "loss": 1.61974564,
      "memory(GiB)": 107.26,
      "step": 34415,
      "train_speed(iter/s)": 1.636213
    },
    {
      "acc": 0.67059326,
      "epoch": 0.8731608320649417,
      "grad_norm": 4.5625,
      "learning_rate": 6.441247433425821e-06,
      "loss": 1.55733433,
      "memory(GiB)": 107.26,
      "step": 34420,
      "train_speed(iter/s)": 1.636239
    },
    {
      "acc": 0.64994831,
      "epoch": 0.8732876712328768,
      "grad_norm": 5.5625,
      "learning_rate": 6.4402432865307384e-06,
      "loss": 1.64555511,
      "memory(GiB)": 107.26,
      "step": 34425,
      "train_speed(iter/s)": 1.636265
    },
    {
      "acc": 0.65168667,
      "epoch": 0.8734145104008117,
      "grad_norm": 5.3125,
      "learning_rate": 6.439239076287327e-06,
      "loss": 1.6625782,
      "memory(GiB)": 107.26,
      "step": 34430,
      "train_speed(iter/s)": 1.636292
    },
    {
      "acc": 0.65405111,
      "epoch": 0.8735413495687468,
      "grad_norm": 6.40625,
      "learning_rate": 6.438234802739753e-06,
      "loss": 1.60159836,
      "memory(GiB)": 107.26,
      "step": 34435,
      "train_speed(iter/s)": 1.636317
    },
    {
      "acc": 0.66004095,
      "epoch": 0.8736681887366818,
      "grad_norm": 6.53125,
      "learning_rate": 6.4372304659321935e-06,
      "loss": 1.6271347,
      "memory(GiB)": 107.26,
      "step": 34440,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65722857,
      "epoch": 0.8737950279046169,
      "grad_norm": 5.71875,
      "learning_rate": 6.43622606590882e-06,
      "loss": 1.6335743,
      "memory(GiB)": 107.26,
      "step": 34445,
      "train_speed(iter/s)": 1.636369
    },
    {
      "acc": 0.64141674,
      "epoch": 0.873921867072552,
      "grad_norm": 5.59375,
      "learning_rate": 6.4352216027138125e-06,
      "loss": 1.58018131,
      "memory(GiB)": 107.26,
      "step": 34450,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.64842167,
      "epoch": 0.874048706240487,
      "grad_norm": 6.375,
      "learning_rate": 6.434217076391351e-06,
      "loss": 1.61761398,
      "memory(GiB)": 107.26,
      "step": 34455,
      "train_speed(iter/s)": 1.636421
    },
    {
      "acc": 0.6538404,
      "epoch": 0.8741755454084221,
      "grad_norm": 5.40625,
      "learning_rate": 6.433212486985618e-06,
      "loss": 1.60644569,
      "memory(GiB)": 107.26,
      "step": 34460,
      "train_speed(iter/s)": 1.636447
    },
    {
      "acc": 0.65105915,
      "epoch": 0.8743023845763572,
      "grad_norm": 4.90625,
      "learning_rate": 6.432207834540802e-06,
      "loss": 1.61576042,
      "memory(GiB)": 107.26,
      "step": 34465,
      "train_speed(iter/s)": 1.636473
    },
    {
      "acc": 0.66844931,
      "epoch": 0.8744292237442922,
      "grad_norm": 4.875,
      "learning_rate": 6.431203119101093e-06,
      "loss": 1.58085232,
      "memory(GiB)": 107.26,
      "step": 34470,
      "train_speed(iter/s)": 1.636497
    },
    {
      "acc": 0.64565792,
      "epoch": 0.8745560629122273,
      "grad_norm": 5.78125,
      "learning_rate": 6.430198340710677e-06,
      "loss": 1.59852791,
      "memory(GiB)": 107.26,
      "step": 34475,
      "train_speed(iter/s)": 1.636523
    },
    {
      "acc": 0.66981173,
      "epoch": 0.8746829020801623,
      "grad_norm": 5.90625,
      "learning_rate": 6.4291934994137566e-06,
      "loss": 1.56081276,
      "memory(GiB)": 107.26,
      "step": 34480,
      "train_speed(iter/s)": 1.636549
    },
    {
      "acc": 0.66475592,
      "epoch": 0.8748097412480974,
      "grad_norm": 6.65625,
      "learning_rate": 6.428188595254521e-06,
      "loss": 1.58975334,
      "memory(GiB)": 107.26,
      "step": 34485,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.67063007,
      "epoch": 0.8749365804160325,
      "grad_norm": 7.03125,
      "learning_rate": 6.427183628277178e-06,
      "loss": 1.61651955,
      "memory(GiB)": 107.26,
      "step": 34490,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.6530767,
      "epoch": 0.8750634195839675,
      "grad_norm": 5.21875,
      "learning_rate": 6.426178598525925e-06,
      "loss": 1.65785427,
      "memory(GiB)": 107.26,
      "step": 34495,
      "train_speed(iter/s)": 1.636628
    },
    {
      "acc": 0.65333533,
      "epoch": 0.8751902587519026,
      "grad_norm": 4.8125,
      "learning_rate": 6.4251735060449725e-06,
      "loss": 1.66282806,
      "memory(GiB)": 107.26,
      "step": 34500,
      "train_speed(iter/s)": 1.636656
    },
    {
      "acc": 0.64274106,
      "epoch": 0.8753170979198377,
      "grad_norm": 5.34375,
      "learning_rate": 6.424168350878524e-06,
      "loss": 1.62472115,
      "memory(GiB)": 107.26,
      "step": 34505,
      "train_speed(iter/s)": 1.636681
    },
    {
      "acc": 0.66486292,
      "epoch": 0.8754439370877727,
      "grad_norm": 6.875,
      "learning_rate": 6.423163133070792e-06,
      "loss": 1.53483925,
      "memory(GiB)": 107.26,
      "step": 34510,
      "train_speed(iter/s)": 1.636708
    },
    {
      "acc": 0.66005802,
      "epoch": 0.8755707762557078,
      "grad_norm": 5.75,
      "learning_rate": 6.422157852665993e-06,
      "loss": 1.66071949,
      "memory(GiB)": 107.26,
      "step": 34515,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.65810995,
      "epoch": 0.8756976154236428,
      "grad_norm": 5.59375,
      "learning_rate": 6.421152509708342e-06,
      "loss": 1.53953047,
      "memory(GiB)": 107.26,
      "step": 34520,
      "train_speed(iter/s)": 1.636759
    },
    {
      "acc": 0.6590312,
      "epoch": 0.8758244545915779,
      "grad_norm": 6.625,
      "learning_rate": 6.4201471042420595e-06,
      "loss": 1.578829,
      "memory(GiB)": 107.26,
      "step": 34525,
      "train_speed(iter/s)": 1.636786
    },
    {
      "acc": 0.65072832,
      "epoch": 0.875951293759513,
      "grad_norm": 5.125,
      "learning_rate": 6.419141636311366e-06,
      "loss": 1.56273546,
      "memory(GiB)": 107.26,
      "step": 34530,
      "train_speed(iter/s)": 1.636812
    },
    {
      "acc": 0.65237608,
      "epoch": 0.876078132927448,
      "grad_norm": 5.90625,
      "learning_rate": 6.4181361059604875e-06,
      "loss": 1.5652914,
      "memory(GiB)": 107.26,
      "step": 34535,
      "train_speed(iter/s)": 1.636838
    },
    {
      "acc": 0.64299164,
      "epoch": 0.8762049720953831,
      "grad_norm": 5.59375,
      "learning_rate": 6.4171305132336515e-06,
      "loss": 1.64306889,
      "memory(GiB)": 107.26,
      "step": 34540,
      "train_speed(iter/s)": 1.636864
    },
    {
      "acc": 0.65462313,
      "epoch": 0.8763318112633182,
      "grad_norm": 5.0,
      "learning_rate": 6.416124858175088e-06,
      "loss": 1.57467413,
      "memory(GiB)": 107.26,
      "step": 34545,
      "train_speed(iter/s)": 1.636891
    },
    {
      "acc": 0.6615304,
      "epoch": 0.8764586504312532,
      "grad_norm": 5.59375,
      "learning_rate": 6.415119140829031e-06,
      "loss": 1.6883213,
      "memory(GiB)": 107.26,
      "step": 34550,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.6573936,
      "epoch": 0.8765854895991883,
      "grad_norm": 5.59375,
      "learning_rate": 6.414113361239715e-06,
      "loss": 1.5804142,
      "memory(GiB)": 107.26,
      "step": 34555,
      "train_speed(iter/s)": 1.636944
    },
    {
      "acc": 0.66936827,
      "epoch": 0.8767123287671232,
      "grad_norm": 5.6875,
      "learning_rate": 6.4131075194513825e-06,
      "loss": 1.5039279,
      "memory(GiB)": 107.26,
      "step": 34560,
      "train_speed(iter/s)": 1.636969
    },
    {
      "acc": 0.67082405,
      "epoch": 0.8768391679350583,
      "grad_norm": 4.90625,
      "learning_rate": 6.41210161550827e-06,
      "loss": 1.40923691,
      "memory(GiB)": 107.26,
      "step": 34565,
      "train_speed(iter/s)": 1.636995
    },
    {
      "acc": 0.65074606,
      "epoch": 0.8769660071029934,
      "grad_norm": 5.03125,
      "learning_rate": 6.411095649454626e-06,
      "loss": 1.58685188,
      "memory(GiB)": 107.26,
      "step": 34570,
      "train_speed(iter/s)": 1.637021
    },
    {
      "acc": 0.65319657,
      "epoch": 0.8770928462709284,
      "grad_norm": 6.15625,
      "learning_rate": 6.410089621334693e-06,
      "loss": 1.63718605,
      "memory(GiB)": 107.26,
      "step": 34575,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.65627975,
      "epoch": 0.8772196854388635,
      "grad_norm": 5.15625,
      "learning_rate": 6.4090835311927236e-06,
      "loss": 1.55061054,
      "memory(GiB)": 107.26,
      "step": 34580,
      "train_speed(iter/s)": 1.637071
    },
    {
      "acc": 0.67677021,
      "epoch": 0.8773465246067986,
      "grad_norm": 5.28125,
      "learning_rate": 6.40807737907297e-06,
      "loss": 1.58040161,
      "memory(GiB)": 107.26,
      "step": 34585,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.65182686,
      "epoch": 0.8774733637747336,
      "grad_norm": 7.15625,
      "learning_rate": 6.407071165019686e-06,
      "loss": 1.64803467,
      "memory(GiB)": 107.26,
      "step": 34590,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.66373453,
      "epoch": 0.8776002029426687,
      "grad_norm": 5.125,
      "learning_rate": 6.40606488907713e-06,
      "loss": 1.58467455,
      "memory(GiB)": 107.26,
      "step": 34595,
      "train_speed(iter/s)": 1.637148
    },
    {
      "acc": 0.64267673,
      "epoch": 0.8777270421106037,
      "grad_norm": 5.03125,
      "learning_rate": 6.4050585512895624e-06,
      "loss": 1.63575554,
      "memory(GiB)": 107.26,
      "step": 34600,
      "train_speed(iter/s)": 1.637174
    },
    {
      "acc": 0.66226158,
      "epoch": 0.8778538812785388,
      "grad_norm": 5.4375,
      "learning_rate": 6.4040521517012475e-06,
      "loss": 1.55161991,
      "memory(GiB)": 107.26,
      "step": 34605,
      "train_speed(iter/s)": 1.6372
    },
    {
      "acc": 0.64370594,
      "epoch": 0.8779807204464739,
      "grad_norm": 6.25,
      "learning_rate": 6.40304569035645e-06,
      "loss": 1.62557526,
      "memory(GiB)": 107.26,
      "step": 34610,
      "train_speed(iter/s)": 1.637226
    },
    {
      "acc": 0.65586767,
      "epoch": 0.8781075596144089,
      "grad_norm": 5.6875,
      "learning_rate": 6.402039167299439e-06,
      "loss": 1.61429729,
      "memory(GiB)": 107.26,
      "step": 34615,
      "train_speed(iter/s)": 1.637254
    },
    {
      "acc": 0.64672518,
      "epoch": 0.878234398782344,
      "grad_norm": 5.53125,
      "learning_rate": 6.401032582574485e-06,
      "loss": 1.62737198,
      "memory(GiB)": 107.26,
      "step": 34620,
      "train_speed(iter/s)": 1.63728
    },
    {
      "acc": 0.65626516,
      "epoch": 0.8783612379502791,
      "grad_norm": 6.96875,
      "learning_rate": 6.400025936225862e-06,
      "loss": 1.58581991,
      "memory(GiB)": 107.26,
      "step": 34625,
      "train_speed(iter/s)": 1.637306
    },
    {
      "acc": 0.64161911,
      "epoch": 0.8784880771182141,
      "grad_norm": 5.21875,
      "learning_rate": 6.399019228297851e-06,
      "loss": 1.65909843,
      "memory(GiB)": 107.26,
      "step": 34630,
      "train_speed(iter/s)": 1.637332
    },
    {
      "acc": 0.66053019,
      "epoch": 0.8786149162861492,
      "grad_norm": 8.0,
      "learning_rate": 6.398012458834724e-06,
      "loss": 1.57179422,
      "memory(GiB)": 107.26,
      "step": 34635,
      "train_speed(iter/s)": 1.637357
    },
    {
      "acc": 0.64921684,
      "epoch": 0.8787417554540842,
      "grad_norm": 4.96875,
      "learning_rate": 6.397005627880771e-06,
      "loss": 1.5992384,
      "memory(GiB)": 107.26,
      "step": 34640,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.65615377,
      "epoch": 0.8788685946220193,
      "grad_norm": 6.5,
      "learning_rate": 6.395998735480271e-06,
      "loss": 1.64360905,
      "memory(GiB)": 107.26,
      "step": 34645,
      "train_speed(iter/s)": 1.637408
    },
    {
      "acc": 0.65786705,
      "epoch": 0.8789954337899544,
      "grad_norm": 5.0625,
      "learning_rate": 6.394991781677516e-06,
      "loss": 1.53924789,
      "memory(GiB)": 107.26,
      "step": 34650,
      "train_speed(iter/s)": 1.637431
    },
    {
      "acc": 0.64682426,
      "epoch": 0.8791222729578894,
      "grad_norm": 5.6875,
      "learning_rate": 6.393984766516792e-06,
      "loss": 1.64797173,
      "memory(GiB)": 107.26,
      "step": 34655,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.64263763,
      "epoch": 0.8792491121258245,
      "grad_norm": 5.8125,
      "learning_rate": 6.392977690042395e-06,
      "loss": 1.73442726,
      "memory(GiB)": 107.26,
      "step": 34660,
      "train_speed(iter/s)": 1.637484
    },
    {
      "acc": 0.65896811,
      "epoch": 0.8793759512937596,
      "grad_norm": 6.84375,
      "learning_rate": 6.3919705522986205e-06,
      "loss": 1.59436302,
      "memory(GiB)": 107.26,
      "step": 34665,
      "train_speed(iter/s)": 1.637508
    },
    {
      "acc": 0.6781909,
      "epoch": 0.8795027904616946,
      "grad_norm": 5.40625,
      "learning_rate": 6.390963353329767e-06,
      "loss": 1.49957829,
      "memory(GiB)": 107.26,
      "step": 34670,
      "train_speed(iter/s)": 1.637534
    },
    {
      "acc": 0.63663464,
      "epoch": 0.8796296296296297,
      "grad_norm": 5.125,
      "learning_rate": 6.389956093180134e-06,
      "loss": 1.66632843,
      "memory(GiB)": 107.26,
      "step": 34675,
      "train_speed(iter/s)": 1.637559
    },
    {
      "acc": 0.66338162,
      "epoch": 0.8797564687975646,
      "grad_norm": 5.34375,
      "learning_rate": 6.388948771894025e-06,
      "loss": 1.58040829,
      "memory(GiB)": 107.26,
      "step": 34680,
      "train_speed(iter/s)": 1.637585
    },
    {
      "acc": 0.66679783,
      "epoch": 0.8798833079654997,
      "grad_norm": 5.15625,
      "learning_rate": 6.38794138951575e-06,
      "loss": 1.58340797,
      "memory(GiB)": 107.26,
      "step": 34685,
      "train_speed(iter/s)": 1.637611
    },
    {
      "acc": 0.66828232,
      "epoch": 0.8800101471334348,
      "grad_norm": 6.21875,
      "learning_rate": 6.386933946089615e-06,
      "loss": 1.58165216,
      "memory(GiB)": 107.26,
      "step": 34690,
      "train_speed(iter/s)": 1.637637
    },
    {
      "acc": 0.66285686,
      "epoch": 0.8801369863013698,
      "grad_norm": 6.46875,
      "learning_rate": 6.385926441659933e-06,
      "loss": 1.61361046,
      "memory(GiB)": 107.26,
      "step": 34695,
      "train_speed(iter/s)": 1.637663
    },
    {
      "acc": 0.65112848,
      "epoch": 0.8802638254693049,
      "grad_norm": 6.59375,
      "learning_rate": 6.38491887627102e-06,
      "loss": 1.65015106,
      "memory(GiB)": 107.26,
      "step": 34700,
      "train_speed(iter/s)": 1.637687
    },
    {
      "acc": 0.65056801,
      "epoch": 0.88039066463724,
      "grad_norm": 6.65625,
      "learning_rate": 6.383911249967188e-06,
      "loss": 1.61523705,
      "memory(GiB)": 107.26,
      "step": 34705,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.65150981,
      "epoch": 0.880517503805175,
      "grad_norm": 5.34375,
      "learning_rate": 6.382903562792764e-06,
      "loss": 1.6159996,
      "memory(GiB)": 107.26,
      "step": 34710,
      "train_speed(iter/s)": 1.637736
    },
    {
      "acc": 0.63804379,
      "epoch": 0.8806443429731101,
      "grad_norm": 6.5625,
      "learning_rate": 6.381895814792065e-06,
      "loss": 1.66233826,
      "memory(GiB)": 107.26,
      "step": 34715,
      "train_speed(iter/s)": 1.637762
    },
    {
      "acc": 0.65599394,
      "epoch": 0.8807711821410451,
      "grad_norm": 5.5,
      "learning_rate": 6.38088800600942e-06,
      "loss": 1.60345058,
      "memory(GiB)": 107.26,
      "step": 34720,
      "train_speed(iter/s)": 1.637789
    },
    {
      "acc": 0.65587921,
      "epoch": 0.8808980213089802,
      "grad_norm": 5.46875,
      "learning_rate": 6.3798801364891535e-06,
      "loss": 1.63268394,
      "memory(GiB)": 107.26,
      "step": 34725,
      "train_speed(iter/s)": 1.637815
    },
    {
      "acc": 0.64889307,
      "epoch": 0.8810248604769153,
      "grad_norm": 6.09375,
      "learning_rate": 6.378872206275599e-06,
      "loss": 1.61623306,
      "memory(GiB)": 107.26,
      "step": 34730,
      "train_speed(iter/s)": 1.637841
    },
    {
      "acc": 0.6586792,
      "epoch": 0.8811516996448503,
      "grad_norm": 5.6875,
      "learning_rate": 6.377864215413088e-06,
      "loss": 1.55771284,
      "memory(GiB)": 107.26,
      "step": 34735,
      "train_speed(iter/s)": 1.637865
    },
    {
      "acc": 0.66447821,
      "epoch": 0.8812785388127854,
      "grad_norm": 6.84375,
      "learning_rate": 6.376856163945957e-06,
      "loss": 1.63500576,
      "memory(GiB)": 107.26,
      "step": 34740,
      "train_speed(iter/s)": 1.63789
    },
    {
      "acc": 0.65884371,
      "epoch": 0.8814053779807205,
      "grad_norm": 6.53125,
      "learning_rate": 6.375848051918546e-06,
      "loss": 1.59303045,
      "memory(GiB)": 107.26,
      "step": 34745,
      "train_speed(iter/s)": 1.637915
    },
    {
      "acc": 0.65938687,
      "epoch": 0.8815322171486555,
      "grad_norm": 5.21875,
      "learning_rate": 6.374839879375194e-06,
      "loss": 1.56345606,
      "memory(GiB)": 107.26,
      "step": 34750,
      "train_speed(iter/s)": 1.637938
    },
    {
      "acc": 0.64880505,
      "epoch": 0.8816590563165906,
      "grad_norm": 5.0625,
      "learning_rate": 6.373831646360245e-06,
      "loss": 1.5778017,
      "memory(GiB)": 107.26,
      "step": 34755,
      "train_speed(iter/s)": 1.637965
    },
    {
      "acc": 0.65841331,
      "epoch": 0.8817858954845256,
      "grad_norm": 5.9375,
      "learning_rate": 6.372823352918048e-06,
      "loss": 1.59769955,
      "memory(GiB)": 107.26,
      "step": 34760,
      "train_speed(iter/s)": 1.637992
    },
    {
      "acc": 0.65031161,
      "epoch": 0.8819127346524607,
      "grad_norm": 5.59375,
      "learning_rate": 6.371814999092951e-06,
      "loss": 1.60031586,
      "memory(GiB)": 107.26,
      "step": 34765,
      "train_speed(iter/s)": 1.638018
    },
    {
      "acc": 0.65365033,
      "epoch": 0.8820395738203958,
      "grad_norm": 6.53125,
      "learning_rate": 6.370806584929305e-06,
      "loss": 1.64977074,
      "memory(GiB)": 107.26,
      "step": 34770,
      "train_speed(iter/s)": 1.638045
    },
    {
      "acc": 0.65816007,
      "epoch": 0.8821664129883308,
      "grad_norm": 4.5625,
      "learning_rate": 6.369798110471463e-06,
      "loss": 1.59827347,
      "memory(GiB)": 107.26,
      "step": 34775,
      "train_speed(iter/s)": 1.638069
    },
    {
      "acc": 0.65530052,
      "epoch": 0.8822932521562659,
      "grad_norm": 6.90625,
      "learning_rate": 6.368789575763787e-06,
      "loss": 1.56892929,
      "memory(GiB)": 107.26,
      "step": 34780,
      "train_speed(iter/s)": 1.638094
    },
    {
      "acc": 0.65165534,
      "epoch": 0.882420091324201,
      "grad_norm": 6.3125,
      "learning_rate": 6.367780980850633e-06,
      "loss": 1.57746811,
      "memory(GiB)": 107.26,
      "step": 34785,
      "train_speed(iter/s)": 1.638119
    },
    {
      "acc": 0.65830736,
      "epoch": 0.882546930492136,
      "grad_norm": 6.59375,
      "learning_rate": 6.366772325776367e-06,
      "loss": 1.65471096,
      "memory(GiB)": 107.26,
      "step": 34790,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.65235801,
      "epoch": 0.882673769660071,
      "grad_norm": 5.9375,
      "learning_rate": 6.365763610585349e-06,
      "loss": 1.5687458,
      "memory(GiB)": 107.26,
      "step": 34795,
      "train_speed(iter/s)": 1.638169
    },
    {
      "acc": 0.67540703,
      "epoch": 0.882800608828006,
      "grad_norm": 6.25,
      "learning_rate": 6.3647548353219515e-06,
      "loss": 1.57441368,
      "memory(GiB)": 107.26,
      "step": 34800,
      "train_speed(iter/s)": 1.638196
    },
    {
      "acc": 0.66891437,
      "epoch": 0.8829274479959411,
      "grad_norm": 6.625,
      "learning_rate": 6.363746000030543e-06,
      "loss": 1.63154259,
      "memory(GiB)": 107.26,
      "step": 34805,
      "train_speed(iter/s)": 1.638221
    },
    {
      "acc": 0.65212326,
      "epoch": 0.8830542871638762,
      "grad_norm": 6.53125,
      "learning_rate": 6.362737104755497e-06,
      "loss": 1.62655983,
      "memory(GiB)": 107.26,
      "step": 34810,
      "train_speed(iter/s)": 1.638248
    },
    {
      "acc": 0.66751871,
      "epoch": 0.8831811263318112,
      "grad_norm": 6.28125,
      "learning_rate": 6.361728149541188e-06,
      "loss": 1.59664602,
      "memory(GiB)": 107.26,
      "step": 34815,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.6539156,
      "epoch": 0.8833079654997463,
      "grad_norm": 5.59375,
      "learning_rate": 6.360719134431995e-06,
      "loss": 1.6347332,
      "memory(GiB)": 107.26,
      "step": 34820,
      "train_speed(iter/s)": 1.638298
    },
    {
      "acc": 0.67178602,
      "epoch": 0.8834348046676814,
      "grad_norm": 7.25,
      "learning_rate": 6.359710059472299e-06,
      "loss": 1.59146099,
      "memory(GiB)": 107.26,
      "step": 34825,
      "train_speed(iter/s)": 1.638324
    },
    {
      "acc": 0.65588737,
      "epoch": 0.8835616438356164,
      "grad_norm": 5.90625,
      "learning_rate": 6.358700924706486e-06,
      "loss": 1.61237297,
      "memory(GiB)": 107.26,
      "step": 34830,
      "train_speed(iter/s)": 1.638349
    },
    {
      "acc": 0.65609603,
      "epoch": 0.8836884830035515,
      "grad_norm": 5.9375,
      "learning_rate": 6.357691730178939e-06,
      "loss": 1.61122551,
      "memory(GiB)": 107.26,
      "step": 34835,
      "train_speed(iter/s)": 1.638374
    },
    {
      "acc": 0.64841795,
      "epoch": 0.8838153221714865,
      "grad_norm": 5.1875,
      "learning_rate": 6.356682475934048e-06,
      "loss": 1.67296028,
      "memory(GiB)": 107.26,
      "step": 34840,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.65464005,
      "epoch": 0.8839421613394216,
      "grad_norm": 6.375,
      "learning_rate": 6.3556731620162036e-06,
      "loss": 1.58997316,
      "memory(GiB)": 107.26,
      "step": 34845,
      "train_speed(iter/s)": 1.638424
    },
    {
      "acc": 0.66694059,
      "epoch": 0.8840690005073567,
      "grad_norm": 7.03125,
      "learning_rate": 6.354663788469803e-06,
      "loss": 1.59241447,
      "memory(GiB)": 107.26,
      "step": 34850,
      "train_speed(iter/s)": 1.638448
    },
    {
      "acc": 0.65776868,
      "epoch": 0.8841958396752917,
      "grad_norm": 6.25,
      "learning_rate": 6.353654355339238e-06,
      "loss": 1.63055115,
      "memory(GiB)": 107.26,
      "step": 34855,
      "train_speed(iter/s)": 1.638473
    },
    {
      "acc": 0.65780544,
      "epoch": 0.8843226788432268,
      "grad_norm": 6.09375,
      "learning_rate": 6.352644862668914e-06,
      "loss": 1.58074188,
      "memory(GiB)": 107.26,
      "step": 34860,
      "train_speed(iter/s)": 1.638496
    },
    {
      "acc": 0.6692008,
      "epoch": 0.8844495180111619,
      "grad_norm": 5.5625,
      "learning_rate": 6.351635310503228e-06,
      "loss": 1.58006744,
      "memory(GiB)": 107.26,
      "step": 34865,
      "train_speed(iter/s)": 1.63852
    },
    {
      "acc": 0.66118078,
      "epoch": 0.8845763571790969,
      "grad_norm": 6.875,
      "learning_rate": 6.3506256988865865e-06,
      "loss": 1.5850481,
      "memory(GiB)": 107.26,
      "step": 34870,
      "train_speed(iter/s)": 1.638548
    },
    {
      "acc": 0.67588987,
      "epoch": 0.884703196347032,
      "grad_norm": 5.15625,
      "learning_rate": 6.349616027863397e-06,
      "loss": 1.51588545,
      "memory(GiB)": 107.26,
      "step": 34875,
      "train_speed(iter/s)": 1.638571
    },
    {
      "acc": 0.65689621,
      "epoch": 0.884830035514967,
      "grad_norm": 5.15625,
      "learning_rate": 6.34860629747807e-06,
      "loss": 1.60547981,
      "memory(GiB)": 107.26,
      "step": 34880,
      "train_speed(iter/s)": 1.638597
    },
    {
      "acc": 0.64912934,
      "epoch": 0.8849568746829021,
      "grad_norm": 7.53125,
      "learning_rate": 6.347596507775016e-06,
      "loss": 1.60005016,
      "memory(GiB)": 107.26,
      "step": 34885,
      "train_speed(iter/s)": 1.638622
    },
    {
      "acc": 0.65871048,
      "epoch": 0.8850837138508372,
      "grad_norm": 5.09375,
      "learning_rate": 6.3465866587986505e-06,
      "loss": 1.51602583,
      "memory(GiB)": 107.26,
      "step": 34890,
      "train_speed(iter/s)": 1.638645
    },
    {
      "acc": 0.66990614,
      "epoch": 0.8852105530187722,
      "grad_norm": 5.0625,
      "learning_rate": 6.345576750593392e-06,
      "loss": 1.55591345,
      "memory(GiB)": 107.26,
      "step": 34895,
      "train_speed(iter/s)": 1.638669
    },
    {
      "acc": 0.64289408,
      "epoch": 0.8853373921867073,
      "grad_norm": 7.125,
      "learning_rate": 6.34456678320366e-06,
      "loss": 1.67922363,
      "memory(GiB)": 107.26,
      "step": 34900,
      "train_speed(iter/s)": 1.638695
    },
    {
      "acc": 0.64900303,
      "epoch": 0.8854642313546424,
      "grad_norm": 4.875,
      "learning_rate": 6.343556756673879e-06,
      "loss": 1.64148579,
      "memory(GiB)": 107.26,
      "step": 34905,
      "train_speed(iter/s)": 1.63872
    },
    {
      "acc": 0.65437651,
      "epoch": 0.8855910705225774,
      "grad_norm": 4.75,
      "learning_rate": 6.3425466710484726e-06,
      "loss": 1.59104872,
      "memory(GiB)": 107.26,
      "step": 34910,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.64971294,
      "epoch": 0.8857179096905125,
      "grad_norm": 5.8125,
      "learning_rate": 6.3415365263718686e-06,
      "loss": 1.6510519,
      "memory(GiB)": 107.26,
      "step": 34915,
      "train_speed(iter/s)": 1.638768
    },
    {
      "acc": 0.65020671,
      "epoch": 0.8858447488584474,
      "grad_norm": 5.3125,
      "learning_rate": 6.340526322688501e-06,
      "loss": 1.58353453,
      "memory(GiB)": 107.26,
      "step": 34920,
      "train_speed(iter/s)": 1.638793
    },
    {
      "acc": 0.66052647,
      "epoch": 0.8859715880263825,
      "grad_norm": 6.6875,
      "learning_rate": 6.339516060042798e-06,
      "loss": 1.62916718,
      "memory(GiB)": 107.26,
      "step": 34925,
      "train_speed(iter/s)": 1.638817
    },
    {
      "acc": 0.64058785,
      "epoch": 0.8860984271943176,
      "grad_norm": 5.625,
      "learning_rate": 6.3385057384792e-06,
      "loss": 1.64945335,
      "memory(GiB)": 107.26,
      "step": 34930,
      "train_speed(iter/s)": 1.638842
    },
    {
      "acc": 0.64525433,
      "epoch": 0.8862252663622526,
      "grad_norm": 6.6875,
      "learning_rate": 6.337495358042143e-06,
      "loss": 1.6395565,
      "memory(GiB)": 107.26,
      "step": 34935,
      "train_speed(iter/s)": 1.638868
    },
    {
      "acc": 0.65324221,
      "epoch": 0.8863521055301877,
      "grad_norm": 5.4375,
      "learning_rate": 6.336484918776069e-06,
      "loss": 1.63604259,
      "memory(GiB)": 107.26,
      "step": 34940,
      "train_speed(iter/s)": 1.638894
    },
    {
      "acc": 0.66482482,
      "epoch": 0.8864789446981228,
      "grad_norm": 5.0625,
      "learning_rate": 6.335474420725421e-06,
      "loss": 1.59167957,
      "memory(GiB)": 107.26,
      "step": 34945,
      "train_speed(iter/s)": 1.63892
    },
    {
      "acc": 0.64228411,
      "epoch": 0.8866057838660578,
      "grad_norm": 5.40625,
      "learning_rate": 6.334463863934646e-06,
      "loss": 1.64272976,
      "memory(GiB)": 107.26,
      "step": 34950,
      "train_speed(iter/s)": 1.638946
    },
    {
      "acc": 0.65288992,
      "epoch": 0.8867326230339929,
      "grad_norm": 5.09375,
      "learning_rate": 6.333453248448192e-06,
      "loss": 1.5728159,
      "memory(GiB)": 107.26,
      "step": 34955,
      "train_speed(iter/s)": 1.63897
    },
    {
      "acc": 0.63415837,
      "epoch": 0.8868594622019279,
      "grad_norm": 5.59375,
      "learning_rate": 6.33244257431051e-06,
      "loss": 1.76514473,
      "memory(GiB)": 107.26,
      "step": 34960,
      "train_speed(iter/s)": 1.638997
    },
    {
      "acc": 0.65475917,
      "epoch": 0.886986301369863,
      "grad_norm": 5.21875,
      "learning_rate": 6.331431841566056e-06,
      "loss": 1.63356133,
      "memory(GiB)": 107.26,
      "step": 34965,
      "train_speed(iter/s)": 1.639023
    },
    {
      "acc": 0.65832453,
      "epoch": 0.8871131405377981,
      "grad_norm": 6.09375,
      "learning_rate": 6.330421050259283e-06,
      "loss": 1.61892357,
      "memory(GiB)": 107.26,
      "step": 34970,
      "train_speed(iter/s)": 1.63905
    },
    {
      "acc": 0.65398064,
      "epoch": 0.8872399797057331,
      "grad_norm": 6.53125,
      "learning_rate": 6.329410200434655e-06,
      "loss": 1.54905472,
      "memory(GiB)": 107.26,
      "step": 34975,
      "train_speed(iter/s)": 1.639074
    },
    {
      "acc": 0.65093765,
      "epoch": 0.8873668188736682,
      "grad_norm": 5.65625,
      "learning_rate": 6.328399292136629e-06,
      "loss": 1.66408138,
      "memory(GiB)": 107.26,
      "step": 34980,
      "train_speed(iter/s)": 1.639098
    },
    {
      "acc": 0.65008602,
      "epoch": 0.8874936580416033,
      "grad_norm": 6.15625,
      "learning_rate": 6.327388325409672e-06,
      "loss": 1.64595585,
      "memory(GiB)": 107.26,
      "step": 34985,
      "train_speed(iter/s)": 1.639124
    },
    {
      "acc": 0.63527822,
      "epoch": 0.8876204972095383,
      "grad_norm": 5.09375,
      "learning_rate": 6.326377300298251e-06,
      "loss": 1.63354034,
      "memory(GiB)": 107.26,
      "step": 34990,
      "train_speed(iter/s)": 1.639149
    },
    {
      "acc": 0.66591129,
      "epoch": 0.8877473363774734,
      "grad_norm": 5.5625,
      "learning_rate": 6.325366216846832e-06,
      "loss": 1.58777866,
      "memory(GiB)": 107.26,
      "step": 34995,
      "train_speed(iter/s)": 1.639175
    },
    {
      "acc": 0.66433449,
      "epoch": 0.8878741755454084,
      "grad_norm": 6.59375,
      "learning_rate": 6.324355075099893e-06,
      "loss": 1.65761032,
      "memory(GiB)": 107.26,
      "step": 35000,
      "train_speed(iter/s)": 1.6392
    },
    {
      "epoch": 0.8878741755454084,
      "eval_acc": 0.6458274599723925,
      "eval_loss": 1.5753097534179688,
      "eval_runtime": 57.8911,
      "eval_samples_per_second": 110.034,
      "eval_steps_per_second": 27.517,
      "step": 35000
    },
    {
      "acc": 0.65280848,
      "epoch": 0.8880010147133435,
      "grad_norm": 5.28125,
      "learning_rate": 6.3233438751019016e-06,
      "loss": 1.63273716,
      "memory(GiB)": 107.26,
      "step": 35005,
      "train_speed(iter/s)": 1.634472
    },
    {
      "acc": 0.66387553,
      "epoch": 0.8881278538812786,
      "grad_norm": 7.71875,
      "learning_rate": 6.322332616897341e-06,
      "loss": 1.57154799,
      "memory(GiB)": 107.26,
      "step": 35010,
      "train_speed(iter/s)": 1.634498
    },
    {
      "acc": 0.66874261,
      "epoch": 0.8882546930492136,
      "grad_norm": 7.3125,
      "learning_rate": 6.321321300530685e-06,
      "loss": 1.46993999,
      "memory(GiB)": 107.26,
      "step": 35015,
      "train_speed(iter/s)": 1.634523
    },
    {
      "acc": 0.66598988,
      "epoch": 0.8883815322171487,
      "grad_norm": 6.5625,
      "learning_rate": 6.320309926046421e-06,
      "loss": 1.58416843,
      "memory(GiB)": 107.26,
      "step": 35020,
      "train_speed(iter/s)": 1.63455
    },
    {
      "acc": 0.63891273,
      "epoch": 0.8885083713850838,
      "grad_norm": 6.71875,
      "learning_rate": 6.319298493489032e-06,
      "loss": 1.6747057,
      "memory(GiB)": 107.26,
      "step": 35025,
      "train_speed(iter/s)": 1.634576
    },
    {
      "acc": 0.65055408,
      "epoch": 0.8886352105530188,
      "grad_norm": 6.5625,
      "learning_rate": 6.318287002903004e-06,
      "loss": 1.58125954,
      "memory(GiB)": 107.26,
      "step": 35030,
      "train_speed(iter/s)": 1.634601
    },
    {
      "acc": 0.63083124,
      "epoch": 0.8887620497209539,
      "grad_norm": 6.0,
      "learning_rate": 6.317275454332829e-06,
      "loss": 1.64551086,
      "memory(GiB)": 107.26,
      "step": 35035,
      "train_speed(iter/s)": 1.634627
    },
    {
      "acc": 0.64888906,
      "epoch": 0.8888888888888888,
      "grad_norm": 9.6875,
      "learning_rate": 6.3162638478229965e-06,
      "loss": 1.64853363,
      "memory(GiB)": 107.26,
      "step": 35040,
      "train_speed(iter/s)": 1.634653
    },
    {
      "acc": 0.65330877,
      "epoch": 0.8890157280568239,
      "grad_norm": 6.125,
      "learning_rate": 6.315252183418005e-06,
      "loss": 1.60603943,
      "memory(GiB)": 107.26,
      "step": 35045,
      "train_speed(iter/s)": 1.634678
    },
    {
      "acc": 0.64801445,
      "epoch": 0.889142567224759,
      "grad_norm": 5.125,
      "learning_rate": 6.31424046116235e-06,
      "loss": 1.71050892,
      "memory(GiB)": 107.26,
      "step": 35050,
      "train_speed(iter/s)": 1.634707
    },
    {
      "acc": 0.65057969,
      "epoch": 0.889269406392694,
      "grad_norm": 5.21875,
      "learning_rate": 6.313228681100532e-06,
      "loss": 1.62970982,
      "memory(GiB)": 107.26,
      "step": 35055,
      "train_speed(iter/s)": 1.634732
    },
    {
      "acc": 0.6642458,
      "epoch": 0.8893962455606291,
      "grad_norm": 5.625,
      "learning_rate": 6.312216843277052e-06,
      "loss": 1.56750851,
      "memory(GiB)": 107.26,
      "step": 35060,
      "train_speed(iter/s)": 1.634758
    },
    {
      "acc": 0.620889,
      "epoch": 0.8895230847285642,
      "grad_norm": 5.78125,
      "learning_rate": 6.3112049477364165e-06,
      "loss": 1.71920853,
      "memory(GiB)": 107.26,
      "step": 35065,
      "train_speed(iter/s)": 1.634784
    },
    {
      "acc": 0.65542688,
      "epoch": 0.8896499238964992,
      "grad_norm": 6.625,
      "learning_rate": 6.310192994523137e-06,
      "loss": 1.64014091,
      "memory(GiB)": 107.26,
      "step": 35070,
      "train_speed(iter/s)": 1.634809
    },
    {
      "acc": 0.64823637,
      "epoch": 0.8897767630644343,
      "grad_norm": 5.5625,
      "learning_rate": 6.309180983681716e-06,
      "loss": 1.61887169,
      "memory(GiB)": 107.26,
      "step": 35075,
      "train_speed(iter/s)": 1.634835
    },
    {
      "acc": 0.64675283,
      "epoch": 0.8899036022323693,
      "grad_norm": 5.375,
      "learning_rate": 6.308168915256671e-06,
      "loss": 1.67565269,
      "memory(GiB)": 107.26,
      "step": 35080,
      "train_speed(iter/s)": 1.634858
    },
    {
      "acc": 0.6456954,
      "epoch": 0.8900304414003044,
      "grad_norm": 6.28125,
      "learning_rate": 6.307156789292518e-06,
      "loss": 1.65382309,
      "memory(GiB)": 107.26,
      "step": 35085,
      "train_speed(iter/s)": 1.634886
    },
    {
      "acc": 0.66452851,
      "epoch": 0.8901572805682395,
      "grad_norm": 5.625,
      "learning_rate": 6.306144605833773e-06,
      "loss": 1.57362566,
      "memory(GiB)": 107.26,
      "step": 35090,
      "train_speed(iter/s)": 1.634915
    },
    {
      "acc": 0.63521528,
      "epoch": 0.8902841197361745,
      "grad_norm": 6.03125,
      "learning_rate": 6.305132364924955e-06,
      "loss": 1.62159691,
      "memory(GiB)": 107.26,
      "step": 35095,
      "train_speed(iter/s)": 1.634941
    },
    {
      "acc": 0.64541359,
      "epoch": 0.8904109589041096,
      "grad_norm": 6.53125,
      "learning_rate": 6.3041200666105905e-06,
      "loss": 1.59821301,
      "memory(GiB)": 107.26,
      "step": 35100,
      "train_speed(iter/s)": 1.634968
    },
    {
      "acc": 0.65029306,
      "epoch": 0.8905377980720447,
      "grad_norm": 5.21875,
      "learning_rate": 6.303107710935202e-06,
      "loss": 1.62801361,
      "memory(GiB)": 107.26,
      "step": 35105,
      "train_speed(iter/s)": 1.634994
    },
    {
      "acc": 0.63913221,
      "epoch": 0.8906646372399797,
      "grad_norm": 6.28125,
      "learning_rate": 6.302095297943319e-06,
      "loss": 1.58933678,
      "memory(GiB)": 107.26,
      "step": 35110,
      "train_speed(iter/s)": 1.63502
    },
    {
      "acc": 0.6435895,
      "epoch": 0.8907914764079148,
      "grad_norm": 5.0,
      "learning_rate": 6.301082827679472e-06,
      "loss": 1.62738838,
      "memory(GiB)": 107.26,
      "step": 35115,
      "train_speed(iter/s)": 1.635047
    },
    {
      "acc": 0.6804801,
      "epoch": 0.8909183155758498,
      "grad_norm": 7.75,
      "learning_rate": 6.300070300188192e-06,
      "loss": 1.55457878,
      "memory(GiB)": 107.26,
      "step": 35120,
      "train_speed(iter/s)": 1.635074
    },
    {
      "acc": 0.65929065,
      "epoch": 0.8910451547437849,
      "grad_norm": 7.40625,
      "learning_rate": 6.2990577155140164e-06,
      "loss": 1.62332287,
      "memory(GiB)": 107.26,
      "step": 35125,
      "train_speed(iter/s)": 1.635102
    },
    {
      "acc": 0.65902109,
      "epoch": 0.89117199391172,
      "grad_norm": 6.03125,
      "learning_rate": 6.298045073701483e-06,
      "loss": 1.57639027,
      "memory(GiB)": 107.26,
      "step": 35130,
      "train_speed(iter/s)": 1.635128
    },
    {
      "acc": 0.65303311,
      "epoch": 0.891298833079655,
      "grad_norm": 5.25,
      "learning_rate": 6.29703237479513e-06,
      "loss": 1.59200039,
      "memory(GiB)": 107.26,
      "step": 35135,
      "train_speed(iter/s)": 1.635152
    },
    {
      "acc": 0.64996223,
      "epoch": 0.8914256722475901,
      "grad_norm": 5.875,
      "learning_rate": 6.296019618839505e-06,
      "loss": 1.64976425,
      "memory(GiB)": 107.26,
      "step": 35140,
      "train_speed(iter/s)": 1.635177
    },
    {
      "acc": 0.66934071,
      "epoch": 0.8915525114155252,
      "grad_norm": 5.03125,
      "learning_rate": 6.295006805879149e-06,
      "loss": 1.54726419,
      "memory(GiB)": 107.26,
      "step": 35145,
      "train_speed(iter/s)": 1.635203
    },
    {
      "acc": 0.65224061,
      "epoch": 0.8916793505834602,
      "grad_norm": 5.5,
      "learning_rate": 6.293993935958613e-06,
      "loss": 1.56719608,
      "memory(GiB)": 107.26,
      "step": 35150,
      "train_speed(iter/s)": 1.635228
    },
    {
      "acc": 0.65799284,
      "epoch": 0.8918061897513953,
      "grad_norm": 5.09375,
      "learning_rate": 6.292981009122445e-06,
      "loss": 1.5790575,
      "memory(GiB)": 107.26,
      "step": 35155,
      "train_speed(iter/s)": 1.635254
    },
    {
      "acc": 0.6496316,
      "epoch": 0.8919330289193302,
      "grad_norm": 6.4375,
      "learning_rate": 6.291968025415202e-06,
      "loss": 1.64907455,
      "memory(GiB)": 107.26,
      "step": 35160,
      "train_speed(iter/s)": 1.635278
    },
    {
      "acc": 0.65701466,
      "epoch": 0.8920598680872653,
      "grad_norm": 5.96875,
      "learning_rate": 6.290954984881434e-06,
      "loss": 1.59845505,
      "memory(GiB)": 107.26,
      "step": 35165,
      "train_speed(iter/s)": 1.635302
    },
    {
      "acc": 0.65795808,
      "epoch": 0.8921867072552004,
      "grad_norm": 6.21875,
      "learning_rate": 6.289941887565703e-06,
      "loss": 1.63652306,
      "memory(GiB)": 107.26,
      "step": 35170,
      "train_speed(iter/s)": 1.635329
    },
    {
      "acc": 0.67381649,
      "epoch": 0.8923135464231354,
      "grad_norm": 5.90625,
      "learning_rate": 6.288928733512569e-06,
      "loss": 1.5697506,
      "memory(GiB)": 107.26,
      "step": 35175,
      "train_speed(iter/s)": 1.635356
    },
    {
      "acc": 0.66548557,
      "epoch": 0.8924403855910705,
      "grad_norm": 6.15625,
      "learning_rate": 6.287915522766596e-06,
      "loss": 1.52901688,
      "memory(GiB)": 107.26,
      "step": 35180,
      "train_speed(iter/s)": 1.635383
    },
    {
      "acc": 0.65016232,
      "epoch": 0.8925672247590056,
      "grad_norm": 6.5,
      "learning_rate": 6.2869022553723465e-06,
      "loss": 1.59212151,
      "memory(GiB)": 107.26,
      "step": 35185,
      "train_speed(iter/s)": 1.63541
    },
    {
      "acc": 0.66490364,
      "epoch": 0.8926940639269406,
      "grad_norm": 5.1875,
      "learning_rate": 6.285888931374391e-06,
      "loss": 1.57499943,
      "memory(GiB)": 107.26,
      "step": 35190,
      "train_speed(iter/s)": 1.635439
    },
    {
      "acc": 0.6520072,
      "epoch": 0.8928209030948757,
      "grad_norm": 6.125,
      "learning_rate": 6.284875550817299e-06,
      "loss": 1.56602497,
      "memory(GiB)": 107.26,
      "step": 35195,
      "train_speed(iter/s)": 1.635466
    },
    {
      "acc": 0.6499917,
      "epoch": 0.8929477422628107,
      "grad_norm": 5.59375,
      "learning_rate": 6.2838621137456425e-06,
      "loss": 1.60754051,
      "memory(GiB)": 107.26,
      "step": 35200,
      "train_speed(iter/s)": 1.635492
    },
    {
      "acc": 0.65284986,
      "epoch": 0.8930745814307458,
      "grad_norm": 4.4375,
      "learning_rate": 6.282848620203999e-06,
      "loss": 1.58412743,
      "memory(GiB)": 107.26,
      "step": 35205,
      "train_speed(iter/s)": 1.635519
    },
    {
      "acc": 0.65429516,
      "epoch": 0.8932014205986809,
      "grad_norm": 5.59375,
      "learning_rate": 6.2818350702369466e-06,
      "loss": 1.60597458,
      "memory(GiB)": 107.26,
      "step": 35210,
      "train_speed(iter/s)": 1.635546
    },
    {
      "acc": 0.66010389,
      "epoch": 0.8933282597666159,
      "grad_norm": 7.8125,
      "learning_rate": 6.280821463889063e-06,
      "loss": 1.61679363,
      "memory(GiB)": 107.26,
      "step": 35215,
      "train_speed(iter/s)": 1.635572
    },
    {
      "acc": 0.66756291,
      "epoch": 0.893455098934551,
      "grad_norm": 5.53125,
      "learning_rate": 6.279807801204936e-06,
      "loss": 1.58928738,
      "memory(GiB)": 107.26,
      "step": 35220,
      "train_speed(iter/s)": 1.635597
    },
    {
      "acc": 0.67244115,
      "epoch": 0.8935819381024861,
      "grad_norm": 5.46875,
      "learning_rate": 6.278794082229145e-06,
      "loss": 1.58225079,
      "memory(GiB)": 107.26,
      "step": 35225,
      "train_speed(iter/s)": 1.635622
    },
    {
      "acc": 0.66417904,
      "epoch": 0.8937087772704211,
      "grad_norm": 5.9375,
      "learning_rate": 6.2777803070062825e-06,
      "loss": 1.60984669,
      "memory(GiB)": 107.26,
      "step": 35230,
      "train_speed(iter/s)": 1.635647
    },
    {
      "acc": 0.65566659,
      "epoch": 0.8938356164383562,
      "grad_norm": 5.65625,
      "learning_rate": 6.276766475580935e-06,
      "loss": 1.57599201,
      "memory(GiB)": 107.26,
      "step": 35235,
      "train_speed(iter/s)": 1.635673
    },
    {
      "acc": 0.63814325,
      "epoch": 0.8939624556062912,
      "grad_norm": 5.90625,
      "learning_rate": 6.2757525879977e-06,
      "loss": 1.70150547,
      "memory(GiB)": 107.26,
      "step": 35240,
      "train_speed(iter/s)": 1.6357
    },
    {
      "acc": 0.65425034,
      "epoch": 0.8940892947742263,
      "grad_norm": 6.25,
      "learning_rate": 6.27473864430117e-06,
      "loss": 1.6459486,
      "memory(GiB)": 107.26,
      "step": 35245,
      "train_speed(iter/s)": 1.635726
    },
    {
      "acc": 0.63694177,
      "epoch": 0.8942161339421614,
      "grad_norm": 5.71875,
      "learning_rate": 6.273724644535942e-06,
      "loss": 1.65102577,
      "memory(GiB)": 107.26,
      "step": 35250,
      "train_speed(iter/s)": 1.635754
    },
    {
      "acc": 0.65684929,
      "epoch": 0.8943429731100964,
      "grad_norm": 5.125,
      "learning_rate": 6.272710588746619e-06,
      "loss": 1.56203547,
      "memory(GiB)": 107.26,
      "step": 35255,
      "train_speed(iter/s)": 1.635779
    },
    {
      "acc": 0.64609299,
      "epoch": 0.8944698122780315,
      "grad_norm": 5.15625,
      "learning_rate": 6.271696476977801e-06,
      "loss": 1.65582352,
      "memory(GiB)": 107.26,
      "step": 35260,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.64698038,
      "epoch": 0.8945966514459666,
      "grad_norm": 5.65625,
      "learning_rate": 6.270682309274094e-06,
      "loss": 1.56543884,
      "memory(GiB)": 107.26,
      "step": 35265,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.65742726,
      "epoch": 0.8947234906139016,
      "grad_norm": 5.6875,
      "learning_rate": 6.269668085680106e-06,
      "loss": 1.58538094,
      "memory(GiB)": 107.26,
      "step": 35270,
      "train_speed(iter/s)": 1.63586
    },
    {
      "acc": 0.64529839,
      "epoch": 0.8948503297818367,
      "grad_norm": 6.8125,
      "learning_rate": 6.268653806240448e-06,
      "loss": 1.66950684,
      "memory(GiB)": 107.26,
      "step": 35275,
      "train_speed(iter/s)": 1.635888
    },
    {
      "acc": 0.65966301,
      "epoch": 0.8949771689497716,
      "grad_norm": 5.40625,
      "learning_rate": 6.26763947099973e-06,
      "loss": 1.58397522,
      "memory(GiB)": 107.26,
      "step": 35280,
      "train_speed(iter/s)": 1.635914
    },
    {
      "acc": 0.64616308,
      "epoch": 0.8951040081177067,
      "grad_norm": 6.46875,
      "learning_rate": 6.266625080002569e-06,
      "loss": 1.70310516,
      "memory(GiB)": 107.26,
      "step": 35285,
      "train_speed(iter/s)": 1.635941
    },
    {
      "acc": 0.64480777,
      "epoch": 0.8952308472856418,
      "grad_norm": 5.25,
      "learning_rate": 6.265610633293582e-06,
      "loss": 1.62725906,
      "memory(GiB)": 107.26,
      "step": 35290,
      "train_speed(iter/s)": 1.635968
    },
    {
      "acc": 0.64881401,
      "epoch": 0.8953576864535768,
      "grad_norm": 6.375,
      "learning_rate": 6.264596130917389e-06,
      "loss": 1.57150612,
      "memory(GiB)": 107.26,
      "step": 35295,
      "train_speed(iter/s)": 1.635994
    },
    {
      "acc": 0.67461615,
      "epoch": 0.8954845256215119,
      "grad_norm": 6.375,
      "learning_rate": 6.2635815729186124e-06,
      "loss": 1.5082366,
      "memory(GiB)": 107.26,
      "step": 35300,
      "train_speed(iter/s)": 1.63602
    },
    {
      "acc": 0.65389862,
      "epoch": 0.895611364789447,
      "grad_norm": 6.375,
      "learning_rate": 6.2625669593418744e-06,
      "loss": 1.60202541,
      "memory(GiB)": 107.26,
      "step": 35305,
      "train_speed(iter/s)": 1.636048
    },
    {
      "acc": 0.6394526,
      "epoch": 0.895738203957382,
      "grad_norm": 6.8125,
      "learning_rate": 6.261552290231807e-06,
      "loss": 1.61742516,
      "memory(GiB)": 107.26,
      "step": 35310,
      "train_speed(iter/s)": 1.636075
    },
    {
      "acc": 0.66084189,
      "epoch": 0.8958650431253171,
      "grad_norm": 5.0625,
      "learning_rate": 6.260537565633037e-06,
      "loss": 1.54237661,
      "memory(GiB)": 107.26,
      "step": 35315,
      "train_speed(iter/s)": 1.636101
    },
    {
      "acc": 0.65795479,
      "epoch": 0.8959918822932521,
      "grad_norm": 5.6875,
      "learning_rate": 6.259522785590197e-06,
      "loss": 1.61592808,
      "memory(GiB)": 107.26,
      "step": 35320,
      "train_speed(iter/s)": 1.636128
    },
    {
      "acc": 0.63821888,
      "epoch": 0.8961187214611872,
      "grad_norm": 7.5,
      "learning_rate": 6.2585079501479205e-06,
      "loss": 1.67298851,
      "memory(GiB)": 107.26,
      "step": 35325,
      "train_speed(iter/s)": 1.636156
    },
    {
      "acc": 0.65477619,
      "epoch": 0.8962455606291223,
      "grad_norm": 5.8125,
      "learning_rate": 6.257493059350848e-06,
      "loss": 1.57751865,
      "memory(GiB)": 107.26,
      "step": 35330,
      "train_speed(iter/s)": 1.636184
    },
    {
      "acc": 0.65645504,
      "epoch": 0.8963723997970573,
      "grad_norm": 4.6875,
      "learning_rate": 6.256478113243613e-06,
      "loss": 1.62918262,
      "memory(GiB)": 107.26,
      "step": 35335,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.65731502,
      "epoch": 0.8964992389649924,
      "grad_norm": 6.21875,
      "learning_rate": 6.255463111870864e-06,
      "loss": 1.5873189,
      "memory(GiB)": 107.26,
      "step": 35340,
      "train_speed(iter/s)": 1.636237
    },
    {
      "acc": 0.65349741,
      "epoch": 0.8966260781329275,
      "grad_norm": 6.3125,
      "learning_rate": 6.25444805527724e-06,
      "loss": 1.61542244,
      "memory(GiB)": 107.26,
      "step": 35345,
      "train_speed(iter/s)": 1.636264
    },
    {
      "acc": 0.64976134,
      "epoch": 0.8967529173008625,
      "grad_norm": 6.0625,
      "learning_rate": 6.253432943507391e-06,
      "loss": 1.68300591,
      "memory(GiB)": 107.26,
      "step": 35350,
      "train_speed(iter/s)": 1.636292
    },
    {
      "acc": 0.65394454,
      "epoch": 0.8968797564687976,
      "grad_norm": 6.125,
      "learning_rate": 6.252417776605964e-06,
      "loss": 1.56053839,
      "memory(GiB)": 107.26,
      "step": 35355,
      "train_speed(iter/s)": 1.636318
    },
    {
      "acc": 0.66816502,
      "epoch": 0.8970065956367326,
      "grad_norm": 5.625,
      "learning_rate": 6.251402554617613e-06,
      "loss": 1.53104286,
      "memory(GiB)": 107.26,
      "step": 35360,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.63301992,
      "epoch": 0.8971334348046677,
      "grad_norm": 7.71875,
      "learning_rate": 6.2503872775869886e-06,
      "loss": 1.72977295,
      "memory(GiB)": 107.26,
      "step": 35365,
      "train_speed(iter/s)": 1.636371
    },
    {
      "acc": 0.67164698,
      "epoch": 0.8972602739726028,
      "grad_norm": 5.40625,
      "learning_rate": 6.249371945558751e-06,
      "loss": 1.55892744,
      "memory(GiB)": 107.26,
      "step": 35370,
      "train_speed(iter/s)": 1.636397
    },
    {
      "acc": 0.66241236,
      "epoch": 0.8973871131405378,
      "grad_norm": 5.5,
      "learning_rate": 6.248356558577555e-06,
      "loss": 1.57639856,
      "memory(GiB)": 107.26,
      "step": 35375,
      "train_speed(iter/s)": 1.636423
    },
    {
      "acc": 0.66527195,
      "epoch": 0.8975139523084729,
      "grad_norm": 7.5,
      "learning_rate": 6.247341116688067e-06,
      "loss": 1.61928024,
      "memory(GiB)": 107.26,
      "step": 35380,
      "train_speed(iter/s)": 1.63645
    },
    {
      "acc": 0.65958743,
      "epoch": 0.897640791476408,
      "grad_norm": 5.46875,
      "learning_rate": 6.246325619934945e-06,
      "loss": 1.56521568,
      "memory(GiB)": 107.26,
      "step": 35385,
      "train_speed(iter/s)": 1.636476
    },
    {
      "acc": 0.65656252,
      "epoch": 0.897767630644343,
      "grad_norm": 6.71875,
      "learning_rate": 6.245310068362859e-06,
      "loss": 1.57629642,
      "memory(GiB)": 107.26,
      "step": 35390,
      "train_speed(iter/s)": 1.636505
    },
    {
      "acc": 0.65723815,
      "epoch": 0.897894469812278,
      "grad_norm": 5.46875,
      "learning_rate": 6.244294462016476e-06,
      "loss": 1.62229996,
      "memory(GiB)": 107.26,
      "step": 35395,
      "train_speed(iter/s)": 1.636533
    },
    {
      "acc": 0.67403202,
      "epoch": 0.898021308980213,
      "grad_norm": 6.8125,
      "learning_rate": 6.243278800940468e-06,
      "loss": 1.52510862,
      "memory(GiB)": 107.26,
      "step": 35400,
      "train_speed(iter/s)": 1.636561
    },
    {
      "acc": 0.66067467,
      "epoch": 0.8981481481481481,
      "grad_norm": 6.5,
      "learning_rate": 6.242263085179506e-06,
      "loss": 1.60486584,
      "memory(GiB)": 107.26,
      "step": 35405,
      "train_speed(iter/s)": 1.636589
    },
    {
      "acc": 0.6519989,
      "epoch": 0.8982749873160832,
      "grad_norm": 5.0,
      "learning_rate": 6.241247314778269e-06,
      "loss": 1.60743599,
      "memory(GiB)": 107.26,
      "step": 35410,
      "train_speed(iter/s)": 1.636616
    },
    {
      "acc": 0.64827452,
      "epoch": 0.8984018264840182,
      "grad_norm": 4.84375,
      "learning_rate": 6.240231489781432e-06,
      "loss": 1.67846203,
      "memory(GiB)": 107.26,
      "step": 35415,
      "train_speed(iter/s)": 1.636644
    },
    {
      "acc": 0.64432755,
      "epoch": 0.8985286656519533,
      "grad_norm": 5.96875,
      "learning_rate": 6.239215610233678e-06,
      "loss": 1.60612183,
      "memory(GiB)": 107.26,
      "step": 35420,
      "train_speed(iter/s)": 1.636673
    },
    {
      "acc": 0.66481633,
      "epoch": 0.8986555048198884,
      "grad_norm": 6.71875,
      "learning_rate": 6.238199676179688e-06,
      "loss": 1.5848711,
      "memory(GiB)": 107.26,
      "step": 35425,
      "train_speed(iter/s)": 1.6367
    },
    {
      "acc": 0.66628904,
      "epoch": 0.8987823439878234,
      "grad_norm": 8.0625,
      "learning_rate": 6.2371836876641475e-06,
      "loss": 1.56425457,
      "memory(GiB)": 107.26,
      "step": 35430,
      "train_speed(iter/s)": 1.636726
    },
    {
      "acc": 0.65049715,
      "epoch": 0.8989091831557585,
      "grad_norm": 6.46875,
      "learning_rate": 6.236167644731745e-06,
      "loss": 1.62554779,
      "memory(GiB)": 107.26,
      "step": 35435,
      "train_speed(iter/s)": 1.636753
    },
    {
      "acc": 0.66369848,
      "epoch": 0.8990360223236935,
      "grad_norm": 5.6875,
      "learning_rate": 6.235151547427172e-06,
      "loss": 1.58174276,
      "memory(GiB)": 107.26,
      "step": 35440,
      "train_speed(iter/s)": 1.636781
    },
    {
      "acc": 0.65990949,
      "epoch": 0.8991628614916286,
      "grad_norm": 5.40625,
      "learning_rate": 6.2341353957951165e-06,
      "loss": 1.62984047,
      "memory(GiB)": 107.26,
      "step": 35445,
      "train_speed(iter/s)": 1.636808
    },
    {
      "acc": 0.63817658,
      "epoch": 0.8992897006595637,
      "grad_norm": 7.40625,
      "learning_rate": 6.233119189880279e-06,
      "loss": 1.64208374,
      "memory(GiB)": 107.26,
      "step": 35450,
      "train_speed(iter/s)": 1.636837
    },
    {
      "acc": 0.64803524,
      "epoch": 0.8994165398274987,
      "grad_norm": 5.40625,
      "learning_rate": 6.232102929727353e-06,
      "loss": 1.63871555,
      "memory(GiB)": 107.26,
      "step": 35455,
      "train_speed(iter/s)": 1.636864
    },
    {
      "acc": 0.64808092,
      "epoch": 0.8995433789954338,
      "grad_norm": 4.90625,
      "learning_rate": 6.231086615381039e-06,
      "loss": 1.67996273,
      "memory(GiB)": 107.26,
      "step": 35460,
      "train_speed(iter/s)": 1.636892
    },
    {
      "acc": 0.66335268,
      "epoch": 0.8996702181633689,
      "grad_norm": 6.40625,
      "learning_rate": 6.2300702468860385e-06,
      "loss": 1.60317669,
      "memory(GiB)": 107.26,
      "step": 35465,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.6618619,
      "epoch": 0.8997970573313039,
      "grad_norm": 6.0625,
      "learning_rate": 6.229053824287058e-06,
      "loss": 1.56038418,
      "memory(GiB)": 107.26,
      "step": 35470,
      "train_speed(iter/s)": 1.636947
    },
    {
      "acc": 0.64557924,
      "epoch": 0.899923896499239,
      "grad_norm": 6.125,
      "learning_rate": 6.228037347628803e-06,
      "loss": 1.62468853,
      "memory(GiB)": 107.26,
      "step": 35475,
      "train_speed(iter/s)": 1.636975
    },
    {
      "acc": 0.66901484,
      "epoch": 0.900050735667174,
      "grad_norm": 4.8125,
      "learning_rate": 6.227020816955982e-06,
      "loss": 1.51534071,
      "memory(GiB)": 107.26,
      "step": 35480,
      "train_speed(iter/s)": 1.637002
    },
    {
      "acc": 0.6630837,
      "epoch": 0.9001775748351091,
      "grad_norm": 5.1875,
      "learning_rate": 6.226004232313308e-06,
      "loss": 1.58505068,
      "memory(GiB)": 107.26,
      "step": 35485,
      "train_speed(iter/s)": 1.63703
    },
    {
      "acc": 0.64404302,
      "epoch": 0.9003044140030442,
      "grad_norm": 5.5625,
      "learning_rate": 6.224987593745493e-06,
      "loss": 1.62261753,
      "memory(GiB)": 107.26,
      "step": 35490,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.64504671,
      "epoch": 0.9004312531709792,
      "grad_norm": 6.53125,
      "learning_rate": 6.223970901297255e-06,
      "loss": 1.58707047,
      "memory(GiB)": 107.26,
      "step": 35495,
      "train_speed(iter/s)": 1.637086
    },
    {
      "acc": 0.65822105,
      "epoch": 0.9005580923389143,
      "grad_norm": 6.28125,
      "learning_rate": 6.222954155013312e-06,
      "loss": 1.60274563,
      "memory(GiB)": 107.26,
      "step": 35500,
      "train_speed(iter/s)": 1.637114
    },
    {
      "acc": 0.66814632,
      "epoch": 0.9006849315068494,
      "grad_norm": 5.625,
      "learning_rate": 6.221937354938386e-06,
      "loss": 1.56985779,
      "memory(GiB)": 107.26,
      "step": 35505,
      "train_speed(iter/s)": 1.637142
    },
    {
      "acc": 0.65876322,
      "epoch": 0.9008117706747844,
      "grad_norm": 5.4375,
      "learning_rate": 6.2209205011171995e-06,
      "loss": 1.61329308,
      "memory(GiB)": 107.26,
      "step": 35510,
      "train_speed(iter/s)": 1.637168
    },
    {
      "acc": 0.63096466,
      "epoch": 0.9009386098427195,
      "grad_norm": 5.59375,
      "learning_rate": 6.219903593594476e-06,
      "loss": 1.6204073,
      "memory(GiB)": 107.26,
      "step": 35515,
      "train_speed(iter/s)": 1.637196
    },
    {
      "acc": 0.66494236,
      "epoch": 0.9010654490106544,
      "grad_norm": 6.625,
      "learning_rate": 6.218886632414949e-06,
      "loss": 1.55846291,
      "memory(GiB)": 107.26,
      "step": 35520,
      "train_speed(iter/s)": 1.637223
    },
    {
      "acc": 0.65294962,
      "epoch": 0.9011922881785895,
      "grad_norm": 5.40625,
      "learning_rate": 6.217869617623343e-06,
      "loss": 1.60607605,
      "memory(GiB)": 107.26,
      "step": 35525,
      "train_speed(iter/s)": 1.63725
    },
    {
      "acc": 0.67328677,
      "epoch": 0.9013191273465246,
      "grad_norm": 5.53125,
      "learning_rate": 6.216852549264396e-06,
      "loss": 1.58206215,
      "memory(GiB)": 107.26,
      "step": 35530,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.66554885,
      "epoch": 0.9014459665144596,
      "grad_norm": 5.9375,
      "learning_rate": 6.215835427382842e-06,
      "loss": 1.56776257,
      "memory(GiB)": 107.26,
      "step": 35535,
      "train_speed(iter/s)": 1.637305
    },
    {
      "acc": 0.65261765,
      "epoch": 0.9015728056823947,
      "grad_norm": 5.40625,
      "learning_rate": 6.214818252023415e-06,
      "loss": 1.60929527,
      "memory(GiB)": 107.26,
      "step": 35540,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.66164408,
      "epoch": 0.9016996448503298,
      "grad_norm": 5.03125,
      "learning_rate": 6.2138010232308585e-06,
      "loss": 1.53942204,
      "memory(GiB)": 107.26,
      "step": 35545,
      "train_speed(iter/s)": 1.63736
    },
    {
      "acc": 0.67017527,
      "epoch": 0.9018264840182648,
      "grad_norm": 6.28125,
      "learning_rate": 6.212783741049915e-06,
      "loss": 1.55215244,
      "memory(GiB)": 107.26,
      "step": 35550,
      "train_speed(iter/s)": 1.637387
    },
    {
      "acc": 0.6511302,
      "epoch": 0.9019533231861999,
      "grad_norm": 4.6875,
      "learning_rate": 6.211766405525326e-06,
      "loss": 1.57251072,
      "memory(GiB)": 107.26,
      "step": 35555,
      "train_speed(iter/s)": 1.637413
    },
    {
      "acc": 0.64634433,
      "epoch": 0.9020801623541349,
      "grad_norm": 6.25,
      "learning_rate": 6.210749016701842e-06,
      "loss": 1.64773903,
      "memory(GiB)": 107.26,
      "step": 35560,
      "train_speed(iter/s)": 1.637441
    },
    {
      "acc": 0.64436336,
      "epoch": 0.90220700152207,
      "grad_norm": 6.3125,
      "learning_rate": 6.2097315746242095e-06,
      "loss": 1.65151157,
      "memory(GiB)": 107.26,
      "step": 35565,
      "train_speed(iter/s)": 1.637466
    },
    {
      "acc": 0.63314352,
      "epoch": 0.9023338406900051,
      "grad_norm": 5.4375,
      "learning_rate": 6.208714079337181e-06,
      "loss": 1.69086132,
      "memory(GiB)": 107.26,
      "step": 35570,
      "train_speed(iter/s)": 1.637491
    },
    {
      "acc": 0.64559937,
      "epoch": 0.9024606798579401,
      "grad_norm": 5.03125,
      "learning_rate": 6.207696530885511e-06,
      "loss": 1.61990757,
      "memory(GiB)": 107.26,
      "step": 35575,
      "train_speed(iter/s)": 1.63752
    },
    {
      "acc": 0.65296516,
      "epoch": 0.9025875190258752,
      "grad_norm": 6.28125,
      "learning_rate": 6.2066789293139565e-06,
      "loss": 1.6086504,
      "memory(GiB)": 107.26,
      "step": 35580,
      "train_speed(iter/s)": 1.637546
    },
    {
      "acc": 0.63693833,
      "epoch": 0.9027143581938103,
      "grad_norm": 5.5,
      "learning_rate": 6.2056612746672736e-06,
      "loss": 1.67460136,
      "memory(GiB)": 107.26,
      "step": 35585,
      "train_speed(iter/s)": 1.637572
    },
    {
      "acc": 0.65801229,
      "epoch": 0.9028411973617453,
      "grad_norm": 5.46875,
      "learning_rate": 6.204643566990227e-06,
      "loss": 1.62313728,
      "memory(GiB)": 107.26,
      "step": 35590,
      "train_speed(iter/s)": 1.6376
    },
    {
      "acc": 0.65059948,
      "epoch": 0.9029680365296804,
      "grad_norm": 5.84375,
      "learning_rate": 6.2036258063275764e-06,
      "loss": 1.57591248,
      "memory(GiB)": 107.26,
      "step": 35595,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.65701418,
      "epoch": 0.9030948756976154,
      "grad_norm": 6.84375,
      "learning_rate": 6.20260799272409e-06,
      "loss": 1.61003704,
      "memory(GiB)": 107.26,
      "step": 35600,
      "train_speed(iter/s)": 1.637654
    },
    {
      "acc": 0.66342473,
      "epoch": 0.9032217148655505,
      "grad_norm": 5.0625,
      "learning_rate": 6.201590126224534e-06,
      "loss": 1.60203552,
      "memory(GiB)": 107.26,
      "step": 35605,
      "train_speed(iter/s)": 1.63768
    },
    {
      "acc": 0.66524825,
      "epoch": 0.9033485540334856,
      "grad_norm": 4.875,
      "learning_rate": 6.20057220687368e-06,
      "loss": 1.51242161,
      "memory(GiB)": 107.26,
      "step": 35610,
      "train_speed(iter/s)": 1.637707
    },
    {
      "acc": 0.65484581,
      "epoch": 0.9034753932014206,
      "grad_norm": 5.1875,
      "learning_rate": 6.199554234716301e-06,
      "loss": 1.58962402,
      "memory(GiB)": 107.26,
      "step": 35615,
      "train_speed(iter/s)": 1.637734
    },
    {
      "acc": 0.65649681,
      "epoch": 0.9036022323693557,
      "grad_norm": 6.0,
      "learning_rate": 6.19853620979717e-06,
      "loss": 1.56722755,
      "memory(GiB)": 107.26,
      "step": 35620,
      "train_speed(iter/s)": 1.637762
    },
    {
      "acc": 0.65528669,
      "epoch": 0.9037290715372908,
      "grad_norm": 5.40625,
      "learning_rate": 6.1975181321610655e-06,
      "loss": 1.58421974,
      "memory(GiB)": 107.26,
      "step": 35625,
      "train_speed(iter/s)": 1.637788
    },
    {
      "acc": 0.64572115,
      "epoch": 0.9038559107052258,
      "grad_norm": 6.40625,
      "learning_rate": 6.1965000018527676e-06,
      "loss": 1.64173851,
      "memory(GiB)": 107.26,
      "step": 35630,
      "train_speed(iter/s)": 1.637816
    },
    {
      "acc": 0.64437065,
      "epoch": 0.9039827498731609,
      "grad_norm": 5.25,
      "learning_rate": 6.195481818917057e-06,
      "loss": 1.62821884,
      "memory(GiB)": 107.26,
      "step": 35635,
      "train_speed(iter/s)": 1.637843
    },
    {
      "acc": 0.65158839,
      "epoch": 0.9041095890410958,
      "grad_norm": 5.59375,
      "learning_rate": 6.194463583398719e-06,
      "loss": 1.61533279,
      "memory(GiB)": 107.26,
      "step": 35640,
      "train_speed(iter/s)": 1.63787
    },
    {
      "acc": 0.65070362,
      "epoch": 0.9042364282090309,
      "grad_norm": 5.875,
      "learning_rate": 6.193445295342538e-06,
      "loss": 1.59141836,
      "memory(GiB)": 107.26,
      "step": 35645,
      "train_speed(iter/s)": 1.637897
    },
    {
      "acc": 0.63709927,
      "epoch": 0.904363267376966,
      "grad_norm": 4.90625,
      "learning_rate": 6.192426954793308e-06,
      "loss": 1.63582268,
      "memory(GiB)": 107.26,
      "step": 35650,
      "train_speed(iter/s)": 1.637925
    },
    {
      "acc": 0.65651522,
      "epoch": 0.904490106544901,
      "grad_norm": 5.75,
      "learning_rate": 6.1914085617958135e-06,
      "loss": 1.61126938,
      "memory(GiB)": 107.26,
      "step": 35655,
      "train_speed(iter/s)": 1.637952
    },
    {
      "acc": 0.66297903,
      "epoch": 0.9046169457128361,
      "grad_norm": 5.25,
      "learning_rate": 6.190390116394853e-06,
      "loss": 1.59218988,
      "memory(GiB)": 107.26,
      "step": 35660,
      "train_speed(iter/s)": 1.637976
    },
    {
      "acc": 0.63934641,
      "epoch": 0.9047437848807712,
      "grad_norm": 9.25,
      "learning_rate": 6.189371618635219e-06,
      "loss": 1.64230671,
      "memory(GiB)": 107.26,
      "step": 35665,
      "train_speed(iter/s)": 1.638004
    },
    {
      "acc": 0.64188948,
      "epoch": 0.9048706240487062,
      "grad_norm": 5.03125,
      "learning_rate": 6.188353068561714e-06,
      "loss": 1.60472813,
      "memory(GiB)": 107.26,
      "step": 35670,
      "train_speed(iter/s)": 1.63803
    },
    {
      "acc": 0.65712948,
      "epoch": 0.9049974632166413,
      "grad_norm": 6.0,
      "learning_rate": 6.187334466219133e-06,
      "loss": 1.55190039,
      "memory(GiB)": 107.26,
      "step": 35675,
      "train_speed(iter/s)": 1.638059
    },
    {
      "acc": 0.65382495,
      "epoch": 0.9051243023845763,
      "grad_norm": 6.46875,
      "learning_rate": 6.18631581165228e-06,
      "loss": 1.55043144,
      "memory(GiB)": 107.26,
      "step": 35680,
      "train_speed(iter/s)": 1.638086
    },
    {
      "acc": 0.64673357,
      "epoch": 0.9052511415525114,
      "grad_norm": 6.0625,
      "learning_rate": 6.185297104905963e-06,
      "loss": 1.62627296,
      "memory(GiB)": 107.26,
      "step": 35685,
      "train_speed(iter/s)": 1.638111
    },
    {
      "acc": 0.64564052,
      "epoch": 0.9053779807204465,
      "grad_norm": 6.1875,
      "learning_rate": 6.184278346024988e-06,
      "loss": 1.64137344,
      "memory(GiB)": 107.26,
      "step": 35690,
      "train_speed(iter/s)": 1.638141
    },
    {
      "acc": 0.6500391,
      "epoch": 0.9055048198883815,
      "grad_norm": 4.90625,
      "learning_rate": 6.183259535054163e-06,
      "loss": 1.57553253,
      "memory(GiB)": 107.26,
      "step": 35695,
      "train_speed(iter/s)": 1.638168
    },
    {
      "acc": 0.63960915,
      "epoch": 0.9056316590563166,
      "grad_norm": 5.4375,
      "learning_rate": 6.1822406720383e-06,
      "loss": 1.60085945,
      "memory(GiB)": 107.26,
      "step": 35700,
      "train_speed(iter/s)": 1.638196
    },
    {
      "acc": 0.65586462,
      "epoch": 0.9057584982242517,
      "grad_norm": 4.96875,
      "learning_rate": 6.181221757022215e-06,
      "loss": 1.60654297,
      "memory(GiB)": 107.26,
      "step": 35705,
      "train_speed(iter/s)": 1.638221
    },
    {
      "acc": 0.65102749,
      "epoch": 0.9058853373921867,
      "grad_norm": 6.3125,
      "learning_rate": 6.180202790050724e-06,
      "loss": 1.59761124,
      "memory(GiB)": 107.26,
      "step": 35710,
      "train_speed(iter/s)": 1.638249
    },
    {
      "acc": 0.65822906,
      "epoch": 0.9060121765601218,
      "grad_norm": 5.375,
      "learning_rate": 6.179183771168643e-06,
      "loss": 1.6333868,
      "memory(GiB)": 107.26,
      "step": 35715,
      "train_speed(iter/s)": 1.638275
    },
    {
      "acc": 0.65344114,
      "epoch": 0.9061390157280568,
      "grad_norm": 5.5625,
      "learning_rate": 6.1781647004207965e-06,
      "loss": 1.64196205,
      "memory(GiB)": 107.26,
      "step": 35720,
      "train_speed(iter/s)": 1.638302
    },
    {
      "acc": 0.64389219,
      "epoch": 0.9062658548959919,
      "grad_norm": 6.6875,
      "learning_rate": 6.177145577852005e-06,
      "loss": 1.684412,
      "memory(GiB)": 107.26,
      "step": 35725,
      "train_speed(iter/s)": 1.638328
    },
    {
      "acc": 0.64120736,
      "epoch": 0.906392694063927,
      "grad_norm": 6.0625,
      "learning_rate": 6.176126403507097e-06,
      "loss": 1.64522171,
      "memory(GiB)": 107.26,
      "step": 35730,
      "train_speed(iter/s)": 1.638357
    },
    {
      "acc": 0.66335068,
      "epoch": 0.906519533231862,
      "grad_norm": 5.46875,
      "learning_rate": 6.175107177430897e-06,
      "loss": 1.563762,
      "memory(GiB)": 107.26,
      "step": 35735,
      "train_speed(iter/s)": 1.638386
    },
    {
      "acc": 0.64939804,
      "epoch": 0.9066463723997971,
      "grad_norm": 5.625,
      "learning_rate": 6.17408789966824e-06,
      "loss": 1.57620907,
      "memory(GiB)": 107.26,
      "step": 35740,
      "train_speed(iter/s)": 1.638412
    },
    {
      "acc": 0.64524412,
      "epoch": 0.9067732115677322,
      "grad_norm": 5.25,
      "learning_rate": 6.173068570263951e-06,
      "loss": 1.61273613,
      "memory(GiB)": 107.26,
      "step": 35745,
      "train_speed(iter/s)": 1.638438
    },
    {
      "acc": 0.65329552,
      "epoch": 0.9069000507356672,
      "grad_norm": 5.9375,
      "learning_rate": 6.172049189262872e-06,
      "loss": 1.62465801,
      "memory(GiB)": 107.26,
      "step": 35750,
      "train_speed(iter/s)": 1.638466
    },
    {
      "acc": 0.66614685,
      "epoch": 0.9070268899036023,
      "grad_norm": 5.125,
      "learning_rate": 6.1710297567098354e-06,
      "loss": 1.59923954,
      "memory(GiB)": 107.26,
      "step": 35755,
      "train_speed(iter/s)": 1.638493
    },
    {
      "acc": 0.65149946,
      "epoch": 0.9071537290715372,
      "grad_norm": 7.21875,
      "learning_rate": 6.170010272649682e-06,
      "loss": 1.63270988,
      "memory(GiB)": 107.26,
      "step": 35760,
      "train_speed(iter/s)": 1.63852
    },
    {
      "acc": 0.64007444,
      "epoch": 0.9072805682394723,
      "grad_norm": 11.9375,
      "learning_rate": 6.168990737127254e-06,
      "loss": 1.65108566,
      "memory(GiB)": 107.26,
      "step": 35765,
      "train_speed(iter/s)": 1.638545
    },
    {
      "acc": 0.65903435,
      "epoch": 0.9074074074074074,
      "grad_norm": 5.5625,
      "learning_rate": 6.167971150187394e-06,
      "loss": 1.587397,
      "memory(GiB)": 107.26,
      "step": 35770,
      "train_speed(iter/s)": 1.638572
    },
    {
      "acc": 0.63616629,
      "epoch": 0.9075342465753424,
      "grad_norm": 6.0625,
      "learning_rate": 6.166951511874948e-06,
      "loss": 1.66819973,
      "memory(GiB)": 107.26,
      "step": 35775,
      "train_speed(iter/s)": 1.638599
    },
    {
      "acc": 0.6569768,
      "epoch": 0.9076610857432775,
      "grad_norm": 6.6875,
      "learning_rate": 6.165931822234764e-06,
      "loss": 1.58296509,
      "memory(GiB)": 107.26,
      "step": 35780,
      "train_speed(iter/s)": 1.638627
    },
    {
      "acc": 0.65801291,
      "epoch": 0.9077879249112126,
      "grad_norm": 7.15625,
      "learning_rate": 6.164912081311694e-06,
      "loss": 1.65813446,
      "memory(GiB)": 107.26,
      "step": 35785,
      "train_speed(iter/s)": 1.638654
    },
    {
      "acc": 0.65638576,
      "epoch": 0.9079147640791476,
      "grad_norm": 5.84375,
      "learning_rate": 6.163892289150588e-06,
      "loss": 1.56630287,
      "memory(GiB)": 107.26,
      "step": 35790,
      "train_speed(iter/s)": 1.638681
    },
    {
      "acc": 0.66190662,
      "epoch": 0.9080416032470827,
      "grad_norm": 6.25,
      "learning_rate": 6.162872445796303e-06,
      "loss": 1.57662067,
      "memory(GiB)": 107.26,
      "step": 35795,
      "train_speed(iter/s)": 1.638708
    },
    {
      "acc": 0.64458818,
      "epoch": 0.9081684424150177,
      "grad_norm": 5.03125,
      "learning_rate": 6.161852551293697e-06,
      "loss": 1.6509428,
      "memory(GiB)": 107.26,
      "step": 35800,
      "train_speed(iter/s)": 1.638735
    },
    {
      "acc": 0.65495453,
      "epoch": 0.9082952815829528,
      "grad_norm": 6.90625,
      "learning_rate": 6.160832605687628e-06,
      "loss": 1.63957863,
      "memory(GiB)": 107.26,
      "step": 35805,
      "train_speed(iter/s)": 1.638761
    },
    {
      "acc": 0.65895343,
      "epoch": 0.9084221207508879,
      "grad_norm": 6.0625,
      "learning_rate": 6.159812609022961e-06,
      "loss": 1.59441814,
      "memory(GiB)": 107.26,
      "step": 35810,
      "train_speed(iter/s)": 1.638788
    },
    {
      "acc": 0.64820228,
      "epoch": 0.9085489599188229,
      "grad_norm": 4.46875,
      "learning_rate": 6.158792561344553e-06,
      "loss": 1.64654045,
      "memory(GiB)": 107.26,
      "step": 35815,
      "train_speed(iter/s)": 1.638814
    },
    {
      "acc": 0.65361223,
      "epoch": 0.908675799086758,
      "grad_norm": 6.90625,
      "learning_rate": 6.157772462697277e-06,
      "loss": 1.63154316,
      "memory(GiB)": 107.26,
      "step": 35820,
      "train_speed(iter/s)": 1.638841
    },
    {
      "acc": 0.65724802,
      "epoch": 0.9088026382546931,
      "grad_norm": 6.375,
      "learning_rate": 6.156752313125998e-06,
      "loss": 1.6546814,
      "memory(GiB)": 107.26,
      "step": 35825,
      "train_speed(iter/s)": 1.638868
    },
    {
      "acc": 0.6573822,
      "epoch": 0.9089294774226281,
      "grad_norm": 6.40625,
      "learning_rate": 6.155732112675587e-06,
      "loss": 1.57991428,
      "memory(GiB)": 107.26,
      "step": 35830,
      "train_speed(iter/s)": 1.638894
    },
    {
      "acc": 0.66967187,
      "epoch": 0.9090563165905632,
      "grad_norm": 6.1875,
      "learning_rate": 6.154711861390919e-06,
      "loss": 1.52042522,
      "memory(GiB)": 107.26,
      "step": 35835,
      "train_speed(iter/s)": 1.63892
    },
    {
      "acc": 0.66354117,
      "epoch": 0.9091831557584982,
      "grad_norm": 4.84375,
      "learning_rate": 6.153691559316868e-06,
      "loss": 1.60629234,
      "memory(GiB)": 107.26,
      "step": 35840,
      "train_speed(iter/s)": 1.638946
    },
    {
      "acc": 0.6384511,
      "epoch": 0.9093099949264333,
      "grad_norm": 6.25,
      "learning_rate": 6.152671206498311e-06,
      "loss": 1.70595627,
      "memory(GiB)": 107.26,
      "step": 35845,
      "train_speed(iter/s)": 1.638972
    },
    {
      "acc": 0.66828499,
      "epoch": 0.9094368340943684,
      "grad_norm": 5.5,
      "learning_rate": 6.151650802980128e-06,
      "loss": 1.55587339,
      "memory(GiB)": 107.26,
      "step": 35850,
      "train_speed(iter/s)": 1.638996
    },
    {
      "acc": 0.64764853,
      "epoch": 0.9095636732623034,
      "grad_norm": 5.84375,
      "learning_rate": 6.150630348807201e-06,
      "loss": 1.66778984,
      "memory(GiB)": 107.26,
      "step": 35855,
      "train_speed(iter/s)": 1.639023
    },
    {
      "acc": 0.63880033,
      "epoch": 0.9096905124302385,
      "grad_norm": 7.78125,
      "learning_rate": 6.149609844024413e-06,
      "loss": 1.6870472,
      "memory(GiB)": 107.26,
      "step": 35860,
      "train_speed(iter/s)": 1.639052
    },
    {
      "acc": 0.6494606,
      "epoch": 0.9098173515981736,
      "grad_norm": 5.03125,
      "learning_rate": 6.148589288676652e-06,
      "loss": 1.67009392,
      "memory(GiB)": 107.26,
      "step": 35865,
      "train_speed(iter/s)": 1.639079
    },
    {
      "acc": 0.65264816,
      "epoch": 0.9099441907661086,
      "grad_norm": 7.09375,
      "learning_rate": 6.147568682808808e-06,
      "loss": 1.60761185,
      "memory(GiB)": 107.26,
      "step": 35870,
      "train_speed(iter/s)": 1.639107
    },
    {
      "acc": 0.66067986,
      "epoch": 0.9100710299340437,
      "grad_norm": 6.09375,
      "learning_rate": 6.146548026465766e-06,
      "loss": 1.63269272,
      "memory(GiB)": 107.26,
      "step": 35875,
      "train_speed(iter/s)": 1.639135
    },
    {
      "acc": 0.66022673,
      "epoch": 0.9101978691019786,
      "grad_norm": 6.53125,
      "learning_rate": 6.145527319692427e-06,
      "loss": 1.54779644,
      "memory(GiB)": 107.26,
      "step": 35880,
      "train_speed(iter/s)": 1.639162
    },
    {
      "acc": 0.65475273,
      "epoch": 0.9103247082699137,
      "grad_norm": 5.84375,
      "learning_rate": 6.144506562533678e-06,
      "loss": 1.5719347,
      "memory(GiB)": 107.26,
      "step": 35885,
      "train_speed(iter/s)": 1.63919
    },
    {
      "acc": 0.65353661,
      "epoch": 0.9104515474378488,
      "grad_norm": 6.625,
      "learning_rate": 6.143485755034425e-06,
      "loss": 1.58170137,
      "memory(GiB)": 107.26,
      "step": 35890,
      "train_speed(iter/s)": 1.639217
    },
    {
      "acc": 0.65148792,
      "epoch": 0.9105783866057838,
      "grad_norm": 6.15625,
      "learning_rate": 6.14246489723956e-06,
      "loss": 1.58532286,
      "memory(GiB)": 107.26,
      "step": 35895,
      "train_speed(iter/s)": 1.639244
    },
    {
      "acc": 0.65632925,
      "epoch": 0.9107052257737189,
      "grad_norm": 7.25,
      "learning_rate": 6.141443989193988e-06,
      "loss": 1.61256351,
      "memory(GiB)": 107.26,
      "step": 35900,
      "train_speed(iter/s)": 1.639274
    },
    {
      "acc": 0.63929205,
      "epoch": 0.910832064941654,
      "grad_norm": 6.25,
      "learning_rate": 6.140423030942615e-06,
      "loss": 1.72920818,
      "memory(GiB)": 107.26,
      "step": 35905,
      "train_speed(iter/s)": 1.6393
    },
    {
      "acc": 0.65873332,
      "epoch": 0.910958904109589,
      "grad_norm": 5.25,
      "learning_rate": 6.139402022530344e-06,
      "loss": 1.62159214,
      "memory(GiB)": 107.26,
      "step": 35910,
      "train_speed(iter/s)": 1.639329
    },
    {
      "acc": 0.64691191,
      "epoch": 0.9110857432775241,
      "grad_norm": 5.84375,
      "learning_rate": 6.138380964002087e-06,
      "loss": 1.62927647,
      "memory(GiB)": 107.26,
      "step": 35915,
      "train_speed(iter/s)": 1.639356
    },
    {
      "acc": 0.67530079,
      "epoch": 0.9112125824454591,
      "grad_norm": 9.375,
      "learning_rate": 6.13735985540275e-06,
      "loss": 1.54399252,
      "memory(GiB)": 107.26,
      "step": 35920,
      "train_speed(iter/s)": 1.639384
    },
    {
      "acc": 0.65794725,
      "epoch": 0.9113394216133942,
      "grad_norm": 7.09375,
      "learning_rate": 6.13633869677725e-06,
      "loss": 1.55421505,
      "memory(GiB)": 117.38,
      "step": 35925,
      "train_speed(iter/s)": 1.639409
    },
    {
      "acc": 0.64901609,
      "epoch": 0.9114662607813293,
      "grad_norm": 5.3125,
      "learning_rate": 6.1353174881705e-06,
      "loss": 1.63148918,
      "memory(GiB)": 117.38,
      "step": 35930,
      "train_speed(iter/s)": 1.639434
    },
    {
      "acc": 0.64009352,
      "epoch": 0.9115930999492643,
      "grad_norm": 5.78125,
      "learning_rate": 6.134296229627419e-06,
      "loss": 1.69694366,
      "memory(GiB)": 117.38,
      "step": 35935,
      "train_speed(iter/s)": 1.639461
    },
    {
      "acc": 0.65991778,
      "epoch": 0.9117199391171994,
      "grad_norm": 5.5,
      "learning_rate": 6.1332749211929255e-06,
      "loss": 1.59715118,
      "memory(GiB)": 117.38,
      "step": 35940,
      "train_speed(iter/s)": 1.639488
    },
    {
      "acc": 0.67639484,
      "epoch": 0.9118467782851345,
      "grad_norm": 6.09375,
      "learning_rate": 6.132253562911941e-06,
      "loss": 1.56201153,
      "memory(GiB)": 117.38,
      "step": 35945,
      "train_speed(iter/s)": 1.639514
    },
    {
      "acc": 0.66358671,
      "epoch": 0.9119736174530695,
      "grad_norm": 5.6875,
      "learning_rate": 6.1312321548293895e-06,
      "loss": 1.55208702,
      "memory(GiB)": 117.38,
      "step": 35950,
      "train_speed(iter/s)": 1.639542
    },
    {
      "acc": 0.65372281,
      "epoch": 0.9121004566210046,
      "grad_norm": 5.375,
      "learning_rate": 6.130210696990197e-06,
      "loss": 1.68709469,
      "memory(GiB)": 117.38,
      "step": 35955,
      "train_speed(iter/s)": 1.639569
    },
    {
      "acc": 0.65379219,
      "epoch": 0.9122272957889396,
      "grad_norm": 6.15625,
      "learning_rate": 6.129189189439293e-06,
      "loss": 1.64993248,
      "memory(GiB)": 117.38,
      "step": 35960,
      "train_speed(iter/s)": 1.639598
    },
    {
      "acc": 0.65376654,
      "epoch": 0.9123541349568747,
      "grad_norm": 7.96875,
      "learning_rate": 6.128167632221605e-06,
      "loss": 1.6426651,
      "memory(GiB)": 117.38,
      "step": 35965,
      "train_speed(iter/s)": 1.639626
    },
    {
      "acc": 0.65782461,
      "epoch": 0.9124809741248098,
      "grad_norm": 5.375,
      "learning_rate": 6.127146025382069e-06,
      "loss": 1.55912247,
      "memory(GiB)": 117.38,
      "step": 35970,
      "train_speed(iter/s)": 1.639654
    },
    {
      "acc": 0.64602838,
      "epoch": 0.9126078132927448,
      "grad_norm": 5.8125,
      "learning_rate": 6.126124368965619e-06,
      "loss": 1.65744972,
      "memory(GiB)": 117.38,
      "step": 35975,
      "train_speed(iter/s)": 1.639683
    },
    {
      "acc": 0.63015456,
      "epoch": 0.9127346524606799,
      "grad_norm": 5.46875,
      "learning_rate": 6.125102663017191e-06,
      "loss": 1.72497082,
      "memory(GiB)": 117.38,
      "step": 35980,
      "train_speed(iter/s)": 1.63971
    },
    {
      "acc": 0.66336951,
      "epoch": 0.912861491628615,
      "grad_norm": 5.03125,
      "learning_rate": 6.124080907581724e-06,
      "loss": 1.57774143,
      "memory(GiB)": 117.38,
      "step": 35985,
      "train_speed(iter/s)": 1.639736
    },
    {
      "acc": 0.66320405,
      "epoch": 0.91298833079655,
      "grad_norm": 5.15625,
      "learning_rate": 6.1230591027041605e-06,
      "loss": 1.58957319,
      "memory(GiB)": 117.38,
      "step": 35990,
      "train_speed(iter/s)": 1.639762
    },
    {
      "acc": 0.6647799,
      "epoch": 0.913115169964485,
      "grad_norm": 5.1875,
      "learning_rate": 6.1220372484294444e-06,
      "loss": 1.56535702,
      "memory(GiB)": 117.38,
      "step": 35995,
      "train_speed(iter/s)": 1.63979
    },
    {
      "acc": 0.65342131,
      "epoch": 0.91324200913242,
      "grad_norm": 5.90625,
      "learning_rate": 6.12101534480252e-06,
      "loss": 1.62986031,
      "memory(GiB)": 117.38,
      "step": 36000,
      "train_speed(iter/s)": 1.639816
    },
    {
      "epoch": 0.91324200913242,
      "eval_acc": 0.6460513263876272,
      "eval_loss": 1.5751882791519165,
      "eval_runtime": 58.2682,
      "eval_samples_per_second": 109.322,
      "eval_steps_per_second": 27.339,
      "step": 36000
    },
    {
      "acc": 0.65351839,
      "epoch": 0.9133688483003551,
      "grad_norm": 6.34375,
      "learning_rate": 6.119993391868335e-06,
      "loss": 1.60084076,
      "memory(GiB)": 117.38,
      "step": 36005,
      "train_speed(iter/s)": 1.635184
    },
    {
      "acc": 0.67257266,
      "epoch": 0.9134956874682902,
      "grad_norm": 5.125,
      "learning_rate": 6.118971389671842e-06,
      "loss": 1.58251028,
      "memory(GiB)": 117.38,
      "step": 36010,
      "train_speed(iter/s)": 1.635209
    },
    {
      "acc": 0.64846458,
      "epoch": 0.9136225266362252,
      "grad_norm": 4.78125,
      "learning_rate": 6.117949338257989e-06,
      "loss": 1.63050919,
      "memory(GiB)": 117.38,
      "step": 36015,
      "train_speed(iter/s)": 1.635233
    },
    {
      "acc": 0.66973658,
      "epoch": 0.9137493658041603,
      "grad_norm": 5.96875,
      "learning_rate": 6.116927237671735e-06,
      "loss": 1.56487503,
      "memory(GiB)": 117.38,
      "step": 36020,
      "train_speed(iter/s)": 1.635256
    },
    {
      "acc": 0.65058022,
      "epoch": 0.9138762049720954,
      "grad_norm": 6.625,
      "learning_rate": 6.115905087958032e-06,
      "loss": 1.59680977,
      "memory(GiB)": 117.38,
      "step": 36025,
      "train_speed(iter/s)": 1.63528
    },
    {
      "acc": 0.65396318,
      "epoch": 0.9140030441400304,
      "grad_norm": 6.4375,
      "learning_rate": 6.114882889161844e-06,
      "loss": 1.64201965,
      "memory(GiB)": 117.38,
      "step": 36030,
      "train_speed(iter/s)": 1.635304
    },
    {
      "acc": 0.66737761,
      "epoch": 0.9141298833079655,
      "grad_norm": 6.15625,
      "learning_rate": 6.113860641328127e-06,
      "loss": 1.60068932,
      "memory(GiB)": 117.38,
      "step": 36035,
      "train_speed(iter/s)": 1.635328
    },
    {
      "acc": 0.65820255,
      "epoch": 0.9142567224759005,
      "grad_norm": 5.84375,
      "learning_rate": 6.112838344501846e-06,
      "loss": 1.55943222,
      "memory(GiB)": 117.38,
      "step": 36040,
      "train_speed(iter/s)": 1.635352
    },
    {
      "acc": 0.64755988,
      "epoch": 0.9143835616438356,
      "grad_norm": 5.5625,
      "learning_rate": 6.111815998727966e-06,
      "loss": 1.65948486,
      "memory(GiB)": 117.38,
      "step": 36045,
      "train_speed(iter/s)": 1.635377
    },
    {
      "acc": 0.6622467,
      "epoch": 0.9145104008117707,
      "grad_norm": 6.875,
      "learning_rate": 6.110793604051455e-06,
      "loss": 1.56997089,
      "memory(GiB)": 117.38,
      "step": 36050,
      "train_speed(iter/s)": 1.635403
    },
    {
      "acc": 0.66383572,
      "epoch": 0.9146372399797057,
      "grad_norm": 6.28125,
      "learning_rate": 6.109771160517283e-06,
      "loss": 1.60202141,
      "memory(GiB)": 117.38,
      "step": 36055,
      "train_speed(iter/s)": 1.635426
    },
    {
      "acc": 0.65538373,
      "epoch": 0.9147640791476408,
      "grad_norm": 6.4375,
      "learning_rate": 6.108748668170419e-06,
      "loss": 1.62757683,
      "memory(GiB)": 117.38,
      "step": 36060,
      "train_speed(iter/s)": 1.63545
    },
    {
      "acc": 0.6619308,
      "epoch": 0.9148909183155759,
      "grad_norm": 5.03125,
      "learning_rate": 6.1077261270558385e-06,
      "loss": 1.59994726,
      "memory(GiB)": 117.38,
      "step": 36065,
      "train_speed(iter/s)": 1.635475
    },
    {
      "acc": 0.65563016,
      "epoch": 0.9150177574835109,
      "grad_norm": 4.8125,
      "learning_rate": 6.106703537218518e-06,
      "loss": 1.63084145,
      "memory(GiB)": 117.38,
      "step": 36070,
      "train_speed(iter/s)": 1.6355
    },
    {
      "acc": 0.64853458,
      "epoch": 0.915144596651446,
      "grad_norm": 5.21875,
      "learning_rate": 6.105680898703434e-06,
      "loss": 1.62785873,
      "memory(GiB)": 117.38,
      "step": 36075,
      "train_speed(iter/s)": 1.635525
    },
    {
      "acc": 0.64009867,
      "epoch": 0.915271435819381,
      "grad_norm": 4.8125,
      "learning_rate": 6.104658211555568e-06,
      "loss": 1.68101349,
      "memory(GiB)": 117.38,
      "step": 36080,
      "train_speed(iter/s)": 1.635549
    },
    {
      "acc": 0.65485687,
      "epoch": 0.9153982749873161,
      "grad_norm": 6.6875,
      "learning_rate": 6.103635475819902e-06,
      "loss": 1.63143883,
      "memory(GiB)": 117.38,
      "step": 36085,
      "train_speed(iter/s)": 1.635575
    },
    {
      "acc": 0.65299535,
      "epoch": 0.9155251141552512,
      "grad_norm": 5.03125,
      "learning_rate": 6.102612691541422e-06,
      "loss": 1.61650009,
      "memory(GiB)": 117.38,
      "step": 36090,
      "train_speed(iter/s)": 1.635603
    },
    {
      "acc": 0.63208127,
      "epoch": 0.9156519533231862,
      "grad_norm": 5.4375,
      "learning_rate": 6.10158985876511e-06,
      "loss": 1.58753653,
      "memory(GiB)": 117.38,
      "step": 36095,
      "train_speed(iter/s)": 1.63563
    },
    {
      "acc": 0.65565453,
      "epoch": 0.9157787924911213,
      "grad_norm": 5.875,
      "learning_rate": 6.10056697753596e-06,
      "loss": 1.58026237,
      "memory(GiB)": 117.38,
      "step": 36100,
      "train_speed(iter/s)": 1.635655
    },
    {
      "acc": 0.65064707,
      "epoch": 0.9159056316590564,
      "grad_norm": 4.96875,
      "learning_rate": 6.0995440478989595e-06,
      "loss": 1.60773468,
      "memory(GiB)": 117.38,
      "step": 36105,
      "train_speed(iter/s)": 1.63568
    },
    {
      "acc": 0.67643147,
      "epoch": 0.9160324708269914,
      "grad_norm": 7.59375,
      "learning_rate": 6.098521069899104e-06,
      "loss": 1.49202375,
      "memory(GiB)": 117.38,
      "step": 36110,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.65322771,
      "epoch": 0.9161593099949265,
      "grad_norm": 7.0625,
      "learning_rate": 6.097498043581385e-06,
      "loss": 1.65457554,
      "memory(GiB)": 117.38,
      "step": 36115,
      "train_speed(iter/s)": 1.635732
    },
    {
      "acc": 0.66424999,
      "epoch": 0.9162861491628614,
      "grad_norm": 6.40625,
      "learning_rate": 6.096474968990804e-06,
      "loss": 1.5746685,
      "memory(GiB)": 117.38,
      "step": 36120,
      "train_speed(iter/s)": 1.635757
    },
    {
      "acc": 0.66097641,
      "epoch": 0.9164129883307965,
      "grad_norm": 4.90625,
      "learning_rate": 6.095451846172358e-06,
      "loss": 1.59456024,
      "memory(GiB)": 117.38,
      "step": 36125,
      "train_speed(iter/s)": 1.635781
    },
    {
      "acc": 0.64653678,
      "epoch": 0.9165398274987316,
      "grad_norm": 6.4375,
      "learning_rate": 6.094428675171049e-06,
      "loss": 1.64743042,
      "memory(GiB)": 117.38,
      "step": 36130,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.65295224,
      "epoch": 0.9166666666666666,
      "grad_norm": 6.34375,
      "learning_rate": 6.09340545603188e-06,
      "loss": 1.63160591,
      "memory(GiB)": 117.38,
      "step": 36135,
      "train_speed(iter/s)": 1.635832
    },
    {
      "acc": 0.67040863,
      "epoch": 0.9167935058346017,
      "grad_norm": 5.8125,
      "learning_rate": 6.092382188799858e-06,
      "loss": 1.58281822,
      "memory(GiB)": 117.38,
      "step": 36140,
      "train_speed(iter/s)": 1.635856
    },
    {
      "acc": 0.66300859,
      "epoch": 0.9169203450025368,
      "grad_norm": 5.625,
      "learning_rate": 6.09135887351999e-06,
      "loss": 1.59765091,
      "memory(GiB)": 117.38,
      "step": 36145,
      "train_speed(iter/s)": 1.635881
    },
    {
      "acc": 0.65275917,
      "epoch": 0.9170471841704718,
      "grad_norm": 8.4375,
      "learning_rate": 6.090335510237286e-06,
      "loss": 1.61115723,
      "memory(GiB)": 117.38,
      "step": 36150,
      "train_speed(iter/s)": 1.635906
    },
    {
      "acc": 0.66676817,
      "epoch": 0.9171740233384069,
      "grad_norm": 5.9375,
      "learning_rate": 6.089312098996758e-06,
      "loss": 1.56021061,
      "memory(GiB)": 117.38,
      "step": 36155,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.65913329,
      "epoch": 0.9173008625063419,
      "grad_norm": 7.5,
      "learning_rate": 6.088288639843422e-06,
      "loss": 1.62115898,
      "memory(GiB)": 117.38,
      "step": 36160,
      "train_speed(iter/s)": 1.635956
    },
    {
      "acc": 0.6623313,
      "epoch": 0.917427701674277,
      "grad_norm": 6.3125,
      "learning_rate": 6.08726513282229e-06,
      "loss": 1.58611641,
      "memory(GiB)": 117.38,
      "step": 36165,
      "train_speed(iter/s)": 1.635979
    },
    {
      "acc": 0.65777364,
      "epoch": 0.9175545408422121,
      "grad_norm": 5.46875,
      "learning_rate": 6.0862415779783855e-06,
      "loss": 1.61070251,
      "memory(GiB)": 117.38,
      "step": 36170,
      "train_speed(iter/s)": 1.636002
    },
    {
      "acc": 0.64643722,
      "epoch": 0.9176813800101471,
      "grad_norm": 7.0,
      "learning_rate": 6.085217975356726e-06,
      "loss": 1.64098854,
      "memory(GiB)": 117.38,
      "step": 36175,
      "train_speed(iter/s)": 1.636025
    },
    {
      "acc": 0.64263363,
      "epoch": 0.9178082191780822,
      "grad_norm": 4.71875,
      "learning_rate": 6.084194325002335e-06,
      "loss": 1.68528385,
      "memory(GiB)": 117.38,
      "step": 36180,
      "train_speed(iter/s)": 1.636048
    },
    {
      "acc": 0.66855478,
      "epoch": 0.9179350583460173,
      "grad_norm": 5.875,
      "learning_rate": 6.083170626960237e-06,
      "loss": 1.56983757,
      "memory(GiB)": 117.38,
      "step": 36185,
      "train_speed(iter/s)": 1.63607
    },
    {
      "acc": 0.66414213,
      "epoch": 0.9180618975139523,
      "grad_norm": 4.65625,
      "learning_rate": 6.082146881275458e-06,
      "loss": 1.59255886,
      "memory(GiB)": 117.38,
      "step": 36190,
      "train_speed(iter/s)": 1.636096
    },
    {
      "acc": 0.65383101,
      "epoch": 0.9181887366818874,
      "grad_norm": 5.46875,
      "learning_rate": 6.081123087993028e-06,
      "loss": 1.59816837,
      "memory(GiB)": 117.38,
      "step": 36195,
      "train_speed(iter/s)": 1.636122
    },
    {
      "acc": 0.638269,
      "epoch": 0.9183155758498224,
      "grad_norm": 6.28125,
      "learning_rate": 6.0800992471579775e-06,
      "loss": 1.63397884,
      "memory(GiB)": 117.38,
      "step": 36200,
      "train_speed(iter/s)": 1.636147
    },
    {
      "acc": 0.66062183,
      "epoch": 0.9184424150177575,
      "grad_norm": 5.71875,
      "learning_rate": 6.079075358815341e-06,
      "loss": 1.56029272,
      "memory(GiB)": 117.38,
      "step": 36205,
      "train_speed(iter/s)": 1.636173
    },
    {
      "acc": 0.64855075,
      "epoch": 0.9185692541856926,
      "grad_norm": 5.6875,
      "learning_rate": 6.078051423010152e-06,
      "loss": 1.63212471,
      "memory(GiB)": 117.38,
      "step": 36210,
      "train_speed(iter/s)": 1.636198
    },
    {
      "acc": 0.64342275,
      "epoch": 0.9186960933536276,
      "grad_norm": 6.09375,
      "learning_rate": 6.077027439787448e-06,
      "loss": 1.68921242,
      "memory(GiB)": 117.38,
      "step": 36215,
      "train_speed(iter/s)": 1.636222
    },
    {
      "acc": 0.65709906,
      "epoch": 0.9188229325215627,
      "grad_norm": 6.15625,
      "learning_rate": 6.076003409192268e-06,
      "loss": 1.62773457,
      "memory(GiB)": 117.38,
      "step": 36220,
      "train_speed(iter/s)": 1.636242
    },
    {
      "acc": 0.65981035,
      "epoch": 0.9189497716894978,
      "grad_norm": 5.625,
      "learning_rate": 6.074979331269656e-06,
      "loss": 1.59904537,
      "memory(GiB)": 117.38,
      "step": 36225,
      "train_speed(iter/s)": 1.636269
    },
    {
      "acc": 0.65500669,
      "epoch": 0.9190766108574328,
      "grad_norm": 5.28125,
      "learning_rate": 6.0739552060646525e-06,
      "loss": 1.67204399,
      "memory(GiB)": 117.38,
      "step": 36230,
      "train_speed(iter/s)": 1.636296
    },
    {
      "acc": 0.66228189,
      "epoch": 0.9192034500253679,
      "grad_norm": 6.46875,
      "learning_rate": 6.0729310336223025e-06,
      "loss": 1.58430262,
      "memory(GiB)": 117.38,
      "step": 36235,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.66498661,
      "epoch": 0.9193302891933028,
      "grad_norm": 5.3125,
      "learning_rate": 6.071906813987658e-06,
      "loss": 1.58328104,
      "memory(GiB)": 117.38,
      "step": 36240,
      "train_speed(iter/s)": 1.636347
    },
    {
      "acc": 0.66475973,
      "epoch": 0.9194571283612379,
      "grad_norm": 6.375,
      "learning_rate": 6.070882547205764e-06,
      "loss": 1.61242313,
      "memory(GiB)": 117.38,
      "step": 36245,
      "train_speed(iter/s)": 1.636371
    },
    {
      "acc": 0.67056246,
      "epoch": 0.919583967529173,
      "grad_norm": 5.875,
      "learning_rate": 6.069858233321677e-06,
      "loss": 1.52447662,
      "memory(GiB)": 117.38,
      "step": 36250,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.67167501,
      "epoch": 0.919710806697108,
      "grad_norm": 5.90625,
      "learning_rate": 6.068833872380445e-06,
      "loss": 1.54489784,
      "memory(GiB)": 117.38,
      "step": 36255,
      "train_speed(iter/s)": 1.636419
    },
    {
      "acc": 0.63946757,
      "epoch": 0.9198376458650431,
      "grad_norm": 6.21875,
      "learning_rate": 6.067809464427129e-06,
      "loss": 1.66090279,
      "memory(GiB)": 117.38,
      "step": 36260,
      "train_speed(iter/s)": 1.636444
    },
    {
      "acc": 0.66337266,
      "epoch": 0.9199644850329782,
      "grad_norm": 5.75,
      "learning_rate": 6.066785009506786e-06,
      "loss": 1.578654,
      "memory(GiB)": 117.38,
      "step": 36265,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.65782366,
      "epoch": 0.9200913242009132,
      "grad_norm": 5.75,
      "learning_rate": 6.065760507664474e-06,
      "loss": 1.60326195,
      "memory(GiB)": 117.38,
      "step": 36270,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.6633647,
      "epoch": 0.9202181633688483,
      "grad_norm": 7.0625,
      "learning_rate": 6.064735958945258e-06,
      "loss": 1.60193977,
      "memory(GiB)": 117.38,
      "step": 36275,
      "train_speed(iter/s)": 1.636522
    },
    {
      "acc": 0.66284847,
      "epoch": 0.9203450025367833,
      "grad_norm": 8.0,
      "learning_rate": 6.0637113633942006e-06,
      "loss": 1.65656433,
      "memory(GiB)": 117.38,
      "step": 36280,
      "train_speed(iter/s)": 1.636546
    },
    {
      "acc": 0.64009738,
      "epoch": 0.9204718417047184,
      "grad_norm": 5.3125,
      "learning_rate": 6.0626867210563675e-06,
      "loss": 1.71321259,
      "memory(GiB)": 117.38,
      "step": 36285,
      "train_speed(iter/s)": 1.63657
    },
    {
      "acc": 0.67031155,
      "epoch": 0.9205986808726535,
      "grad_norm": 4.9375,
      "learning_rate": 6.061662031976828e-06,
      "loss": 1.55949631,
      "memory(GiB)": 117.38,
      "step": 36290,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.65030584,
      "epoch": 0.9207255200405885,
      "grad_norm": 5.5,
      "learning_rate": 6.0606372962006534e-06,
      "loss": 1.56111631,
      "memory(GiB)": 117.38,
      "step": 36295,
      "train_speed(iter/s)": 1.636619
    },
    {
      "acc": 0.64955354,
      "epoch": 0.9208523592085236,
      "grad_norm": 5.1875,
      "learning_rate": 6.0596125137729145e-06,
      "loss": 1.61581497,
      "memory(GiB)": 117.38,
      "step": 36300,
      "train_speed(iter/s)": 1.636642
    },
    {
      "acc": 0.64847479,
      "epoch": 0.9209791983764587,
      "grad_norm": 6.0,
      "learning_rate": 6.058587684738685e-06,
      "loss": 1.66633797,
      "memory(GiB)": 117.38,
      "step": 36305,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.66139259,
      "epoch": 0.9211060375443937,
      "grad_norm": 6.625,
      "learning_rate": 6.057562809143045e-06,
      "loss": 1.56020889,
      "memory(GiB)": 117.38,
      "step": 36310,
      "train_speed(iter/s)": 1.636691
    },
    {
      "acc": 0.65074482,
      "epoch": 0.9212328767123288,
      "grad_norm": 5.53125,
      "learning_rate": 6.056537887031069e-06,
      "loss": 1.61470642,
      "memory(GiB)": 117.38,
      "step": 36315,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.66339464,
      "epoch": 0.9213597158802638,
      "grad_norm": 5.03125,
      "learning_rate": 6.055512918447841e-06,
      "loss": 1.60316505,
      "memory(GiB)": 117.38,
      "step": 36320,
      "train_speed(iter/s)": 1.636367
    },
    {
      "acc": 0.64431586,
      "epoch": 0.9214865550481989,
      "grad_norm": 5.03125,
      "learning_rate": 6.054487903438442e-06,
      "loss": 1.6006237,
      "memory(GiB)": 117.38,
      "step": 36325,
      "train_speed(iter/s)": 1.636392
    },
    {
      "acc": 0.65243015,
      "epoch": 0.921613394216134,
      "grad_norm": 5.875,
      "learning_rate": 6.0534628420479576e-06,
      "loss": 1.63444786,
      "memory(GiB)": 117.38,
      "step": 36330,
      "train_speed(iter/s)": 1.636418
    },
    {
      "acc": 0.6644886,
      "epoch": 0.921740233384069,
      "grad_norm": 5.4375,
      "learning_rate": 6.0524377343214724e-06,
      "loss": 1.62916622,
      "memory(GiB)": 117.38,
      "step": 36335,
      "train_speed(iter/s)": 1.636443
    },
    {
      "acc": 0.64734612,
      "epoch": 0.9218670725520041,
      "grad_norm": 5.875,
      "learning_rate": 6.051412580304079e-06,
      "loss": 1.63898258,
      "memory(GiB)": 117.38,
      "step": 36340,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.64840808,
      "epoch": 0.9219939117199392,
      "grad_norm": 4.71875,
      "learning_rate": 6.050387380040864e-06,
      "loss": 1.64395428,
      "memory(GiB)": 117.38,
      "step": 36345,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.63976455,
      "epoch": 0.9221207508878742,
      "grad_norm": 5.375,
      "learning_rate": 6.049362133576924e-06,
      "loss": 1.65737228,
      "memory(GiB)": 117.38,
      "step": 36350,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.6784482,
      "epoch": 0.9222475900558093,
      "grad_norm": 6.15625,
      "learning_rate": 6.048336840957351e-06,
      "loss": 1.56237516,
      "memory(GiB)": 117.38,
      "step": 36355,
      "train_speed(iter/s)": 1.636545
    },
    {
      "acc": 0.66122484,
      "epoch": 0.9223744292237442,
      "grad_norm": 5.15625,
      "learning_rate": 6.047311502227245e-06,
      "loss": 1.60077171,
      "memory(GiB)": 117.38,
      "step": 36360,
      "train_speed(iter/s)": 1.636571
    },
    {
      "acc": 0.65276766,
      "epoch": 0.9225012683916793,
      "grad_norm": 5.8125,
      "learning_rate": 6.046286117431703e-06,
      "loss": 1.61099243,
      "memory(GiB)": 117.38,
      "step": 36365,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.65844078,
      "epoch": 0.9226281075596144,
      "grad_norm": 4.40625,
      "learning_rate": 6.0452606866158246e-06,
      "loss": 1.64127617,
      "memory(GiB)": 117.38,
      "step": 36370,
      "train_speed(iter/s)": 1.636618
    },
    {
      "acc": 0.65731187,
      "epoch": 0.9227549467275494,
      "grad_norm": 5.90625,
      "learning_rate": 6.044235209824716e-06,
      "loss": 1.55045872,
      "memory(GiB)": 117.38,
      "step": 36375,
      "train_speed(iter/s)": 1.636644
    },
    {
      "acc": 0.64707279,
      "epoch": 0.9228817858954845,
      "grad_norm": 6.3125,
      "learning_rate": 6.04320968710348e-06,
      "loss": 1.70206375,
      "memory(GiB)": 117.38,
      "step": 36380,
      "train_speed(iter/s)": 1.636668
    },
    {
      "acc": 0.63317547,
      "epoch": 0.9230086250634196,
      "grad_norm": 6.90625,
      "learning_rate": 6.042184118497223e-06,
      "loss": 1.67799072,
      "memory(GiB)": 117.38,
      "step": 36385,
      "train_speed(iter/s)": 1.636694
    },
    {
      "acc": 0.67488623,
      "epoch": 0.9231354642313546,
      "grad_norm": 6.3125,
      "learning_rate": 6.0411585040510576e-06,
      "loss": 1.57867241,
      "memory(GiB)": 117.38,
      "step": 36390,
      "train_speed(iter/s)": 1.636719
    },
    {
      "acc": 0.66463165,
      "epoch": 0.9232623033992897,
      "grad_norm": 5.375,
      "learning_rate": 6.040132843810091e-06,
      "loss": 1.56813745,
      "memory(GiB)": 117.38,
      "step": 36395,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.68164845,
      "epoch": 0.9233891425672247,
      "grad_norm": 6.28125,
      "learning_rate": 6.03910713781944e-06,
      "loss": 1.48613977,
      "memory(GiB)": 117.38,
      "step": 36400,
      "train_speed(iter/s)": 1.636771
    },
    {
      "acc": 0.64825621,
      "epoch": 0.9235159817351598,
      "grad_norm": 6.25,
      "learning_rate": 6.038081386124216e-06,
      "loss": 1.60890808,
      "memory(GiB)": 117.38,
      "step": 36405,
      "train_speed(iter/s)": 1.636797
    },
    {
      "acc": 0.65801048,
      "epoch": 0.9236428209030949,
      "grad_norm": 5.84375,
      "learning_rate": 6.037055588769539e-06,
      "loss": 1.58543015,
      "memory(GiB)": 117.38,
      "step": 36410,
      "train_speed(iter/s)": 1.636821
    },
    {
      "acc": 0.66115756,
      "epoch": 0.9237696600710299,
      "grad_norm": 5.15625,
      "learning_rate": 6.036029745800527e-06,
      "loss": 1.56123562,
      "memory(GiB)": 117.38,
      "step": 36415,
      "train_speed(iter/s)": 1.636846
    },
    {
      "acc": 0.65744977,
      "epoch": 0.923896499238965,
      "grad_norm": 5.40625,
      "learning_rate": 6.0350038572623e-06,
      "loss": 1.66076927,
      "memory(GiB)": 117.38,
      "step": 36420,
      "train_speed(iter/s)": 1.63687
    },
    {
      "acc": 0.64745369,
      "epoch": 0.9240233384069001,
      "grad_norm": 5.15625,
      "learning_rate": 6.033977923199984e-06,
      "loss": 1.60215511,
      "memory(GiB)": 117.38,
      "step": 36425,
      "train_speed(iter/s)": 1.636896
    },
    {
      "acc": 0.67307682,
      "epoch": 0.9241501775748351,
      "grad_norm": 6.375,
      "learning_rate": 6.032951943658702e-06,
      "loss": 1.51451397,
      "memory(GiB)": 117.38,
      "step": 36430,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.65064034,
      "epoch": 0.9242770167427702,
      "grad_norm": 6.0,
      "learning_rate": 6.031925918683582e-06,
      "loss": 1.62221336,
      "memory(GiB)": 117.38,
      "step": 36435,
      "train_speed(iter/s)": 1.636944
    },
    {
      "acc": 0.63989716,
      "epoch": 0.9244038559107052,
      "grad_norm": 6.4375,
      "learning_rate": 6.030899848319754e-06,
      "loss": 1.66184444,
      "memory(GiB)": 117.38,
      "step": 36440,
      "train_speed(iter/s)": 1.636969
    },
    {
      "acc": 0.64444189,
      "epoch": 0.9245306950786403,
      "grad_norm": 5.25,
      "learning_rate": 6.029873732612346e-06,
      "loss": 1.66274128,
      "memory(GiB)": 117.38,
      "step": 36445,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.6538352,
      "epoch": 0.9246575342465754,
      "grad_norm": 6.25,
      "learning_rate": 6.028847571606493e-06,
      "loss": 1.65692596,
      "memory(GiB)": 117.38,
      "step": 36450,
      "train_speed(iter/s)": 1.637017
    },
    {
      "acc": 0.64574213,
      "epoch": 0.9247843734145104,
      "grad_norm": 5.25,
      "learning_rate": 6.0278213653473305e-06,
      "loss": 1.60543823,
      "memory(GiB)": 117.38,
      "step": 36455,
      "train_speed(iter/s)": 1.637042
    },
    {
      "acc": 0.65711288,
      "epoch": 0.9249112125824455,
      "grad_norm": 5.21875,
      "learning_rate": 6.026795113879998e-06,
      "loss": 1.62518845,
      "memory(GiB)": 117.38,
      "step": 36460,
      "train_speed(iter/s)": 1.637068
    },
    {
      "acc": 0.65391207,
      "epoch": 0.9250380517503806,
      "grad_norm": 5.75,
      "learning_rate": 6.025768817249629e-06,
      "loss": 1.63711586,
      "memory(GiB)": 117.38,
      "step": 36465,
      "train_speed(iter/s)": 1.637094
    },
    {
      "acc": 0.62530565,
      "epoch": 0.9251648909183156,
      "grad_norm": 6.15625,
      "learning_rate": 6.024742475501369e-06,
      "loss": 1.71547623,
      "memory(GiB)": 117.38,
      "step": 36470,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.65347185,
      "epoch": 0.9252917300862507,
      "grad_norm": 5.625,
      "learning_rate": 6.023716088680359e-06,
      "loss": 1.5999218,
      "memory(GiB)": 117.38,
      "step": 36475,
      "train_speed(iter/s)": 1.637146
    },
    {
      "acc": 0.64557905,
      "epoch": 0.9254185692541856,
      "grad_norm": 6.0625,
      "learning_rate": 6.022689656831746e-06,
      "loss": 1.54885921,
      "memory(GiB)": 117.38,
      "step": 36480,
      "train_speed(iter/s)": 1.637172
    },
    {
      "acc": 0.65377927,
      "epoch": 0.9255454084221207,
      "grad_norm": 8.0625,
      "learning_rate": 6.021663180000675e-06,
      "loss": 1.58136654,
      "memory(GiB)": 117.38,
      "step": 36485,
      "train_speed(iter/s)": 1.637197
    },
    {
      "acc": 0.65053911,
      "epoch": 0.9256722475900558,
      "grad_norm": 5.75,
      "learning_rate": 6.020636658232297e-06,
      "loss": 1.61182327,
      "memory(GiB)": 117.38,
      "step": 36490,
      "train_speed(iter/s)": 1.637222
    },
    {
      "acc": 0.64730105,
      "epoch": 0.9257990867579908,
      "grad_norm": 6.125,
      "learning_rate": 6.019610091571762e-06,
      "loss": 1.64903641,
      "memory(GiB)": 117.38,
      "step": 36495,
      "train_speed(iter/s)": 1.637246
    },
    {
      "acc": 0.64974604,
      "epoch": 0.9259259259259259,
      "grad_norm": 7.375,
      "learning_rate": 6.018583480064222e-06,
      "loss": 1.69215298,
      "memory(GiB)": 117.38,
      "step": 36500,
      "train_speed(iter/s)": 1.637271
    },
    {
      "acc": 0.65762601,
      "epoch": 0.926052765093861,
      "grad_norm": 5.34375,
      "learning_rate": 6.017556823754833e-06,
      "loss": 1.50422382,
      "memory(GiB)": 117.38,
      "step": 36505,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.66133471,
      "epoch": 0.926179604261796,
      "grad_norm": 7.375,
      "learning_rate": 6.016530122688753e-06,
      "loss": 1.61290493,
      "memory(GiB)": 117.38,
      "step": 36510,
      "train_speed(iter/s)": 1.63732
    },
    {
      "acc": 0.66842136,
      "epoch": 0.9263064434297311,
      "grad_norm": 6.09375,
      "learning_rate": 6.015503376911138e-06,
      "loss": 1.58520184,
      "memory(GiB)": 117.38,
      "step": 36515,
      "train_speed(iter/s)": 1.637346
    },
    {
      "acc": 0.64907427,
      "epoch": 0.9264332825976661,
      "grad_norm": 4.9375,
      "learning_rate": 6.0144765864671515e-06,
      "loss": 1.61483612,
      "memory(GiB)": 117.38,
      "step": 36520,
      "train_speed(iter/s)": 1.637372
    },
    {
      "acc": 0.6322504,
      "epoch": 0.9265601217656012,
      "grad_norm": 5.4375,
      "learning_rate": 6.013449751401954e-06,
      "loss": 1.64845676,
      "memory(GiB)": 117.38,
      "step": 36525,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.65881457,
      "epoch": 0.9266869609335363,
      "grad_norm": 6.15625,
      "learning_rate": 6.012422871760715e-06,
      "loss": 1.59565029,
      "memory(GiB)": 117.38,
      "step": 36530,
      "train_speed(iter/s)": 1.637423
    },
    {
      "acc": 0.64303846,
      "epoch": 0.9268138001014713,
      "grad_norm": 5.25,
      "learning_rate": 6.011395947588594e-06,
      "loss": 1.64643135,
      "memory(GiB)": 117.38,
      "step": 36535,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.65159397,
      "epoch": 0.9269406392694064,
      "grad_norm": 5.46875,
      "learning_rate": 6.010368978930767e-06,
      "loss": 1.60015945,
      "memory(GiB)": 117.38,
      "step": 36540,
      "train_speed(iter/s)": 1.637474
    },
    {
      "acc": 0.65719876,
      "epoch": 0.9270674784373415,
      "grad_norm": 5.9375,
      "learning_rate": 6.0093419658323995e-06,
      "loss": 1.65373898,
      "memory(GiB)": 117.38,
      "step": 36545,
      "train_speed(iter/s)": 1.637499
    },
    {
      "acc": 0.64759283,
      "epoch": 0.9271943176052765,
      "grad_norm": 6.0,
      "learning_rate": 6.0083149083386675e-06,
      "loss": 1.69489746,
      "memory(GiB)": 117.38,
      "step": 36550,
      "train_speed(iter/s)": 1.637525
    },
    {
      "acc": 0.64462337,
      "epoch": 0.9273211567732116,
      "grad_norm": 5.78125,
      "learning_rate": 6.007287806494742e-06,
      "loss": 1.64526329,
      "memory(GiB)": 117.38,
      "step": 36555,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.66012745,
      "epoch": 0.9274479959411466,
      "grad_norm": 5.28125,
      "learning_rate": 6.006260660345802e-06,
      "loss": 1.59140701,
      "memory(GiB)": 117.38,
      "step": 36560,
      "train_speed(iter/s)": 1.637575
    },
    {
      "acc": 0.65189114,
      "epoch": 0.9275748351090817,
      "grad_norm": 5.3125,
      "learning_rate": 6.005233469937027e-06,
      "loss": 1.61465282,
      "memory(GiB)": 117.38,
      "step": 36565,
      "train_speed(iter/s)": 1.637601
    },
    {
      "acc": 0.66557894,
      "epoch": 0.9277016742770168,
      "grad_norm": 6.34375,
      "learning_rate": 6.004206235313594e-06,
      "loss": 1.55650749,
      "memory(GiB)": 117.38,
      "step": 36570,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.65521727,
      "epoch": 0.9278285134449518,
      "grad_norm": 5.21875,
      "learning_rate": 6.003178956520688e-06,
      "loss": 1.59921608,
      "memory(GiB)": 117.38,
      "step": 36575,
      "train_speed(iter/s)": 1.637653
    },
    {
      "acc": 0.65742292,
      "epoch": 0.9279553526128869,
      "grad_norm": 4.4375,
      "learning_rate": 6.002151633603493e-06,
      "loss": 1.62825947,
      "memory(GiB)": 117.38,
      "step": 36580,
      "train_speed(iter/s)": 1.637679
    },
    {
      "acc": 0.64909029,
      "epoch": 0.928082191780822,
      "grad_norm": 5.03125,
      "learning_rate": 6.0011242666071945e-06,
      "loss": 1.62185135,
      "memory(GiB)": 117.38,
      "step": 36585,
      "train_speed(iter/s)": 1.637703
    },
    {
      "acc": 0.65099921,
      "epoch": 0.928209030948757,
      "grad_norm": 5.3125,
      "learning_rate": 6.000096855576982e-06,
      "loss": 1.55952425,
      "memory(GiB)": 117.38,
      "step": 36590,
      "train_speed(iter/s)": 1.637727
    },
    {
      "acc": 0.66165724,
      "epoch": 0.928335870116692,
      "grad_norm": 5.34375,
      "learning_rate": 5.999069400558044e-06,
      "loss": 1.62240829,
      "memory(GiB)": 117.38,
      "step": 36595,
      "train_speed(iter/s)": 1.637749
    },
    {
      "acc": 0.65915008,
      "epoch": 0.928462709284627,
      "grad_norm": 6.09375,
      "learning_rate": 5.998041901595573e-06,
      "loss": 1.58293343,
      "memory(GiB)": 117.38,
      "step": 36600,
      "train_speed(iter/s)": 1.637775
    },
    {
      "acc": 0.63523045,
      "epoch": 0.9285895484525621,
      "grad_norm": 5.78125,
      "learning_rate": 5.997014358734763e-06,
      "loss": 1.6437191,
      "memory(GiB)": 117.38,
      "step": 36605,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.65464315,
      "epoch": 0.9287163876204972,
      "grad_norm": 5.9375,
      "learning_rate": 5.995986772020811e-06,
      "loss": 1.54708357,
      "memory(GiB)": 117.38,
      "step": 36610,
      "train_speed(iter/s)": 1.637825
    },
    {
      "acc": 0.64718585,
      "epoch": 0.9288432267884322,
      "grad_norm": 5.28125,
      "learning_rate": 5.994959141498913e-06,
      "loss": 1.61540813,
      "memory(GiB)": 117.38,
      "step": 36615,
      "train_speed(iter/s)": 1.637849
    },
    {
      "acc": 0.66159501,
      "epoch": 0.9289700659563673,
      "grad_norm": 4.6875,
      "learning_rate": 5.993931467214272e-06,
      "loss": 1.6199379,
      "memory(GiB)": 117.38,
      "step": 36620,
      "train_speed(iter/s)": 1.637874
    },
    {
      "acc": 0.64900413,
      "epoch": 0.9290969051243024,
      "grad_norm": 5.59375,
      "learning_rate": 5.992903749212084e-06,
      "loss": 1.6228775,
      "memory(GiB)": 117.38,
      "step": 36625,
      "train_speed(iter/s)": 1.637899
    },
    {
      "acc": 0.64893112,
      "epoch": 0.9292237442922374,
      "grad_norm": 6.25,
      "learning_rate": 5.991875987537559e-06,
      "loss": 1.6396759,
      "memory(GiB)": 117.38,
      "step": 36630,
      "train_speed(iter/s)": 1.637924
    },
    {
      "acc": 0.65748696,
      "epoch": 0.9293505834601725,
      "grad_norm": 6.84375,
      "learning_rate": 5.990848182235898e-06,
      "loss": 1.63052235,
      "memory(GiB)": 117.38,
      "step": 36635,
      "train_speed(iter/s)": 1.637949
    },
    {
      "acc": 0.64518785,
      "epoch": 0.9294774226281075,
      "grad_norm": 7.0,
      "learning_rate": 5.98982033335231e-06,
      "loss": 1.67615356,
      "memory(GiB)": 117.38,
      "step": 36640,
      "train_speed(iter/s)": 1.637974
    },
    {
      "acc": 0.66198421,
      "epoch": 0.9296042617960426,
      "grad_norm": 5.59375,
      "learning_rate": 5.988792440932006e-06,
      "loss": 1.51006746,
      "memory(GiB)": 117.38,
      "step": 36645,
      "train_speed(iter/s)": 1.637997
    },
    {
      "acc": 0.66837864,
      "epoch": 0.9297311009639777,
      "grad_norm": 5.40625,
      "learning_rate": 5.987764505020195e-06,
      "loss": 1.51374893,
      "memory(GiB)": 117.38,
      "step": 36650,
      "train_speed(iter/s)": 1.638023
    },
    {
      "acc": 0.64073043,
      "epoch": 0.9298579401319127,
      "grad_norm": 5.75,
      "learning_rate": 5.986736525662091e-06,
      "loss": 1.62323952,
      "memory(GiB)": 117.38,
      "step": 36655,
      "train_speed(iter/s)": 1.638048
    },
    {
      "acc": 0.65298071,
      "epoch": 0.9299847792998478,
      "grad_norm": 4.71875,
      "learning_rate": 5.985708502902909e-06,
      "loss": 1.56594706,
      "memory(GiB)": 117.38,
      "step": 36660,
      "train_speed(iter/s)": 1.638072
    },
    {
      "acc": 0.65557222,
      "epoch": 0.9301116184677829,
      "grad_norm": 5.65625,
      "learning_rate": 5.984680436787867e-06,
      "loss": 1.62632198,
      "memory(GiB)": 117.38,
      "step": 36665,
      "train_speed(iter/s)": 1.638096
    },
    {
      "acc": 0.65136414,
      "epoch": 0.9302384576357179,
      "grad_norm": 5.28125,
      "learning_rate": 5.983652327362182e-06,
      "loss": 1.60573158,
      "memory(GiB)": 117.38,
      "step": 36670,
      "train_speed(iter/s)": 1.638119
    },
    {
      "acc": 0.65772572,
      "epoch": 0.930365296803653,
      "grad_norm": 5.09375,
      "learning_rate": 5.982624174671077e-06,
      "loss": 1.64197807,
      "memory(GiB)": 117.38,
      "step": 36675,
      "train_speed(iter/s)": 1.638142
    },
    {
      "acc": 0.65992022,
      "epoch": 0.930492135971588,
      "grad_norm": 6.78125,
      "learning_rate": 5.981595978759773e-06,
      "loss": 1.58844309,
      "memory(GiB)": 117.38,
      "step": 36680,
      "train_speed(iter/s)": 1.638167
    },
    {
      "acc": 0.65233703,
      "epoch": 0.9306189751395231,
      "grad_norm": 4.9375,
      "learning_rate": 5.980567739673495e-06,
      "loss": 1.64766006,
      "memory(GiB)": 117.38,
      "step": 36685,
      "train_speed(iter/s)": 1.638192
    },
    {
      "acc": 0.65973969,
      "epoch": 0.9307458143074582,
      "grad_norm": 5.375,
      "learning_rate": 5.979539457457472e-06,
      "loss": 1.60913048,
      "memory(GiB)": 117.38,
      "step": 36690,
      "train_speed(iter/s)": 1.638218
    },
    {
      "acc": 0.65812969,
      "epoch": 0.9308726534753932,
      "grad_norm": 6.34375,
      "learning_rate": 5.978511132156928e-06,
      "loss": 1.63085709,
      "memory(GiB)": 117.38,
      "step": 36695,
      "train_speed(iter/s)": 1.638241
    },
    {
      "acc": 0.64890075,
      "epoch": 0.9309994926433283,
      "grad_norm": 6.28125,
      "learning_rate": 5.9774827638170965e-06,
      "loss": 1.65883102,
      "memory(GiB)": 117.38,
      "step": 36700,
      "train_speed(iter/s)": 1.638267
    },
    {
      "acc": 0.66292391,
      "epoch": 0.9311263318112634,
      "grad_norm": 5.125,
      "learning_rate": 5.9764543524832085e-06,
      "loss": 1.55409737,
      "memory(GiB)": 117.38,
      "step": 36705,
      "train_speed(iter/s)": 1.638292
    },
    {
      "acc": 0.66930556,
      "epoch": 0.9312531709791984,
      "grad_norm": 6.21875,
      "learning_rate": 5.975425898200499e-06,
      "loss": 1.59013996,
      "memory(GiB)": 117.38,
      "step": 36710,
      "train_speed(iter/s)": 1.638316
    },
    {
      "acc": 0.65103798,
      "epoch": 0.9313800101471335,
      "grad_norm": 6.53125,
      "learning_rate": 5.974397401014202e-06,
      "loss": 1.66796513,
      "memory(GiB)": 117.38,
      "step": 36715,
      "train_speed(iter/s)": 1.63834
    },
    {
      "acc": 0.64828854,
      "epoch": 0.9315068493150684,
      "grad_norm": 5.5,
      "learning_rate": 5.973368860969559e-06,
      "loss": 1.62634964,
      "memory(GiB)": 117.38,
      "step": 36720,
      "train_speed(iter/s)": 1.638364
    },
    {
      "acc": 0.65654998,
      "epoch": 0.9316336884830035,
      "grad_norm": 6.625,
      "learning_rate": 5.972340278111808e-06,
      "loss": 1.62117386,
      "memory(GiB)": 117.38,
      "step": 36725,
      "train_speed(iter/s)": 1.638388
    },
    {
      "acc": 0.66287837,
      "epoch": 0.9317605276509386,
      "grad_norm": 5.1875,
      "learning_rate": 5.9713116524861895e-06,
      "loss": 1.51388626,
      "memory(GiB)": 117.38,
      "step": 36730,
      "train_speed(iter/s)": 1.63841
    },
    {
      "acc": 0.65718532,
      "epoch": 0.9318873668188736,
      "grad_norm": 6.65625,
      "learning_rate": 5.970282984137947e-06,
      "loss": 1.66140022,
      "memory(GiB)": 117.38,
      "step": 36735,
      "train_speed(iter/s)": 1.638434
    },
    {
      "acc": 0.67708817,
      "epoch": 0.9320142059868087,
      "grad_norm": 5.15625,
      "learning_rate": 5.969254273112328e-06,
      "loss": 1.53770084,
      "memory(GiB)": 117.38,
      "step": 36740,
      "train_speed(iter/s)": 1.638459
    },
    {
      "acc": 0.65507145,
      "epoch": 0.9321410451547438,
      "grad_norm": 7.03125,
      "learning_rate": 5.968225519454577e-06,
      "loss": 1.63341446,
      "memory(GiB)": 117.38,
      "step": 36745,
      "train_speed(iter/s)": 1.638481
    },
    {
      "acc": 0.66716251,
      "epoch": 0.9322678843226788,
      "grad_norm": 6.0,
      "learning_rate": 5.967196723209947e-06,
      "loss": 1.57138758,
      "memory(GiB)": 117.38,
      "step": 36750,
      "train_speed(iter/s)": 1.638504
    },
    {
      "acc": 0.64545074,
      "epoch": 0.9323947234906139,
      "grad_norm": 6.40625,
      "learning_rate": 5.966167884423686e-06,
      "loss": 1.69396286,
      "memory(GiB)": 117.38,
      "step": 36755,
      "train_speed(iter/s)": 1.638528
    },
    {
      "acc": 0.67997723,
      "epoch": 0.9325215626585489,
      "grad_norm": 6.78125,
      "learning_rate": 5.965139003141048e-06,
      "loss": 1.50527534,
      "memory(GiB)": 117.38,
      "step": 36760,
      "train_speed(iter/s)": 1.638553
    },
    {
      "acc": 0.65336704,
      "epoch": 0.932648401826484,
      "grad_norm": 5.5,
      "learning_rate": 5.964110079407287e-06,
      "loss": 1.58715115,
      "memory(GiB)": 117.38,
      "step": 36765,
      "train_speed(iter/s)": 1.638579
    },
    {
      "acc": 0.64763618,
      "epoch": 0.9327752409944191,
      "grad_norm": 5.90625,
      "learning_rate": 5.9630811132676625e-06,
      "loss": 1.66229744,
      "memory(GiB)": 117.38,
      "step": 36770,
      "train_speed(iter/s)": 1.638603
    },
    {
      "acc": 0.65673161,
      "epoch": 0.9329020801623541,
      "grad_norm": 5.5,
      "learning_rate": 5.962052104767427e-06,
      "loss": 1.64112015,
      "memory(GiB)": 117.38,
      "step": 36775,
      "train_speed(iter/s)": 1.638627
    },
    {
      "acc": 0.66401796,
      "epoch": 0.9330289193302892,
      "grad_norm": 6.375,
      "learning_rate": 5.961023053951848e-06,
      "loss": 1.59465237,
      "memory(GiB)": 117.38,
      "step": 36780,
      "train_speed(iter/s)": 1.63865
    },
    {
      "acc": 0.66067505,
      "epoch": 0.9331557584982243,
      "grad_norm": 5.96875,
      "learning_rate": 5.9599939608661825e-06,
      "loss": 1.65322723,
      "memory(GiB)": 117.38,
      "step": 36785,
      "train_speed(iter/s)": 1.638672
    },
    {
      "acc": 0.64260502,
      "epoch": 0.9332825976661593,
      "grad_norm": 5.4375,
      "learning_rate": 5.9589648255556975e-06,
      "loss": 1.64575615,
      "memory(GiB)": 117.38,
      "step": 36790,
      "train_speed(iter/s)": 1.638697
    },
    {
      "acc": 0.65490389,
      "epoch": 0.9334094368340944,
      "grad_norm": 6.15625,
      "learning_rate": 5.957935648065658e-06,
      "loss": 1.58748703,
      "memory(GiB)": 117.38,
      "step": 36795,
      "train_speed(iter/s)": 1.638721
    },
    {
      "acc": 0.65766916,
      "epoch": 0.9335362760020294,
      "grad_norm": 6.875,
      "learning_rate": 5.956906428441331e-06,
      "loss": 1.58518772,
      "memory(GiB)": 117.38,
      "step": 36800,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.65439596,
      "epoch": 0.9336631151699645,
      "grad_norm": 5.1875,
      "learning_rate": 5.955877166727988e-06,
      "loss": 1.62941055,
      "memory(GiB)": 117.38,
      "step": 36805,
      "train_speed(iter/s)": 1.63877
    },
    {
      "acc": 0.65952988,
      "epoch": 0.9337899543378996,
      "grad_norm": 6.3125,
      "learning_rate": 5.954847862970898e-06,
      "loss": 1.60132389,
      "memory(GiB)": 117.38,
      "step": 36810,
      "train_speed(iter/s)": 1.638795
    },
    {
      "acc": 0.64738297,
      "epoch": 0.9339167935058346,
      "grad_norm": 4.46875,
      "learning_rate": 5.953818517215338e-06,
      "loss": 1.66351566,
      "memory(GiB)": 117.38,
      "step": 36815,
      "train_speed(iter/s)": 1.638818
    },
    {
      "acc": 0.66023378,
      "epoch": 0.9340436326737697,
      "grad_norm": 5.375,
      "learning_rate": 5.95278912950658e-06,
      "loss": 1.60740948,
      "memory(GiB)": 117.38,
      "step": 36820,
      "train_speed(iter/s)": 1.638841
    },
    {
      "acc": 0.67299509,
      "epoch": 0.9341704718417048,
      "grad_norm": 5.09375,
      "learning_rate": 5.9517596998899e-06,
      "loss": 1.51019211,
      "memory(GiB)": 117.38,
      "step": 36825,
      "train_speed(iter/s)": 1.638867
    },
    {
      "acc": 0.66861639,
      "epoch": 0.9342973110096398,
      "grad_norm": 5.8125,
      "learning_rate": 5.9507302284105836e-06,
      "loss": 1.55800476,
      "memory(GiB)": 117.38,
      "step": 36830,
      "train_speed(iter/s)": 1.638889
    },
    {
      "acc": 0.66935649,
      "epoch": 0.9344241501775749,
      "grad_norm": 5.5,
      "learning_rate": 5.949700715113904e-06,
      "loss": 1.59582157,
      "memory(GiB)": 117.38,
      "step": 36835,
      "train_speed(iter/s)": 1.638913
    },
    {
      "acc": 0.6585845,
      "epoch": 0.9345509893455098,
      "grad_norm": 6.4375,
      "learning_rate": 5.9486711600451484e-06,
      "loss": 1.60249119,
      "memory(GiB)": 117.38,
      "step": 36840,
      "train_speed(iter/s)": 1.638937
    },
    {
      "acc": 0.66531453,
      "epoch": 0.9346778285134449,
      "grad_norm": 5.75,
      "learning_rate": 5.9476415632495974e-06,
      "loss": 1.59580631,
      "memory(GiB)": 117.38,
      "step": 36845,
      "train_speed(iter/s)": 1.638961
    },
    {
      "acc": 0.64385386,
      "epoch": 0.93480466768138,
      "grad_norm": 5.46875,
      "learning_rate": 5.946611924772542e-06,
      "loss": 1.61241417,
      "memory(GiB)": 117.38,
      "step": 36850,
      "train_speed(iter/s)": 1.638984
    },
    {
      "acc": 0.66709561,
      "epoch": 0.934931506849315,
      "grad_norm": 5.21875,
      "learning_rate": 5.945582244659267e-06,
      "loss": 1.55254078,
      "memory(GiB)": 117.38,
      "step": 36855,
      "train_speed(iter/s)": 1.639009
    },
    {
      "acc": 0.65680184,
      "epoch": 0.9350583460172501,
      "grad_norm": 5.90625,
      "learning_rate": 5.944552522955063e-06,
      "loss": 1.63569489,
      "memory(GiB)": 117.38,
      "step": 36860,
      "train_speed(iter/s)": 1.639033
    },
    {
      "acc": 0.67392912,
      "epoch": 0.9351851851851852,
      "grad_norm": 4.4375,
      "learning_rate": 5.943522759705221e-06,
      "loss": 1.58377724,
      "memory(GiB)": 117.38,
      "step": 36865,
      "train_speed(iter/s)": 1.639057
    },
    {
      "acc": 0.64179478,
      "epoch": 0.9353120243531202,
      "grad_norm": 5.0,
      "learning_rate": 5.942492954955037e-06,
      "loss": 1.67848854,
      "memory(GiB)": 117.38,
      "step": 36870,
      "train_speed(iter/s)": 1.639082
    },
    {
      "acc": 0.63717403,
      "epoch": 0.9354388635210553,
      "grad_norm": 5.84375,
      "learning_rate": 5.941463108749804e-06,
      "loss": 1.67264671,
      "memory(GiB)": 117.38,
      "step": 36875,
      "train_speed(iter/s)": 1.639106
    },
    {
      "acc": 0.65555096,
      "epoch": 0.9355657026889903,
      "grad_norm": 6.65625,
      "learning_rate": 5.940433221134821e-06,
      "loss": 1.61928978,
      "memory(GiB)": 117.38,
      "step": 36880,
      "train_speed(iter/s)": 1.639132
    },
    {
      "acc": 0.66600709,
      "epoch": 0.9356925418569254,
      "grad_norm": 5.71875,
      "learning_rate": 5.9394032921553856e-06,
      "loss": 1.61757965,
      "memory(GiB)": 117.38,
      "step": 36885,
      "train_speed(iter/s)": 1.638737
    },
    {
      "acc": 0.6494102,
      "epoch": 0.9358193810248605,
      "grad_norm": 6.625,
      "learning_rate": 5.9383733218568e-06,
      "loss": 1.65413971,
      "memory(GiB)": 117.38,
      "step": 36890,
      "train_speed(iter/s)": 1.638763
    },
    {
      "acc": 0.65740604,
      "epoch": 0.9359462201927955,
      "grad_norm": 5.28125,
      "learning_rate": 5.937343310284365e-06,
      "loss": 1.63521404,
      "memory(GiB)": 117.38,
      "step": 36895,
      "train_speed(iter/s)": 1.638365
    },
    {
      "acc": 0.65558767,
      "epoch": 0.9360730593607306,
      "grad_norm": 5.03125,
      "learning_rate": 5.936313257483387e-06,
      "loss": 1.59242296,
      "memory(GiB)": 117.38,
      "step": 36900,
      "train_speed(iter/s)": 1.638391
    },
    {
      "acc": 0.64981451,
      "epoch": 0.9361998985286657,
      "grad_norm": 5.1875,
      "learning_rate": 5.935283163499171e-06,
      "loss": 1.64229584,
      "memory(GiB)": 117.38,
      "step": 36905,
      "train_speed(iter/s)": 1.638416
    },
    {
      "acc": 0.63950448,
      "epoch": 0.9363267376966007,
      "grad_norm": 5.25,
      "learning_rate": 5.9342530283770274e-06,
      "loss": 1.66219273,
      "memory(GiB)": 117.38,
      "step": 36910,
      "train_speed(iter/s)": 1.638438
    },
    {
      "acc": 0.66492348,
      "epoch": 0.9364535768645358,
      "grad_norm": 4.53125,
      "learning_rate": 5.9332228521622615e-06,
      "loss": 1.51675892,
      "memory(GiB)": 117.38,
      "step": 36915,
      "train_speed(iter/s)": 1.638038
    },
    {
      "acc": 0.66963034,
      "epoch": 0.9365804160324708,
      "grad_norm": 5.6875,
      "learning_rate": 5.93219263490019e-06,
      "loss": 1.614711,
      "memory(GiB)": 117.38,
      "step": 36920,
      "train_speed(iter/s)": 1.638063
    },
    {
      "acc": 0.66557255,
      "epoch": 0.9367072552004059,
      "grad_norm": 5.75,
      "learning_rate": 5.931162376636123e-06,
      "loss": 1.6623661,
      "memory(GiB)": 117.38,
      "step": 36925,
      "train_speed(iter/s)": 1.638087
    },
    {
      "acc": 0.6570353,
      "epoch": 0.936834094368341,
      "grad_norm": 6.5,
      "learning_rate": 5.93013207741538e-06,
      "loss": 1.56609812,
      "memory(GiB)": 117.38,
      "step": 36930,
      "train_speed(iter/s)": 1.63811
    },
    {
      "acc": 0.65826535,
      "epoch": 0.936960933536276,
      "grad_norm": 5.90625,
      "learning_rate": 5.929101737283274e-06,
      "loss": 1.56675377,
      "memory(GiB)": 117.38,
      "step": 36935,
      "train_speed(iter/s)": 1.638135
    },
    {
      "acc": 0.64306622,
      "epoch": 0.9370877727042111,
      "grad_norm": 6.0625,
      "learning_rate": 5.928071356285126e-06,
      "loss": 1.60418987,
      "memory(GiB)": 117.38,
      "step": 36940,
      "train_speed(iter/s)": 1.638159
    },
    {
      "acc": 0.66855669,
      "epoch": 0.9372146118721462,
      "grad_norm": 5.4375,
      "learning_rate": 5.927040934466255e-06,
      "loss": 1.55486479,
      "memory(GiB)": 117.38,
      "step": 36945,
      "train_speed(iter/s)": 1.638185
    },
    {
      "acc": 0.67566752,
      "epoch": 0.9373414510400812,
      "grad_norm": 5.1875,
      "learning_rate": 5.926010471871986e-06,
      "loss": 1.55081129,
      "memory(GiB)": 117.38,
      "step": 36950,
      "train_speed(iter/s)": 1.638209
    },
    {
      "acc": 0.65869427,
      "epoch": 0.9374682902080163,
      "grad_norm": 4.875,
      "learning_rate": 5.924979968547642e-06,
      "loss": 1.5476099,
      "memory(GiB)": 117.38,
      "step": 36955,
      "train_speed(iter/s)": 1.638234
    },
    {
      "acc": 0.63511362,
      "epoch": 0.9375951293759512,
      "grad_norm": 6.53125,
      "learning_rate": 5.9239494245385485e-06,
      "loss": 1.71486073,
      "memory(GiB)": 117.38,
      "step": 36960,
      "train_speed(iter/s)": 1.638258
    },
    {
      "acc": 0.65801692,
      "epoch": 0.9377219685438863,
      "grad_norm": 5.59375,
      "learning_rate": 5.9229188398900325e-06,
      "loss": 1.62753143,
      "memory(GiB)": 117.38,
      "step": 36965,
      "train_speed(iter/s)": 1.638282
    },
    {
      "acc": 0.6414794,
      "epoch": 0.9378488077118214,
      "grad_norm": 5.84375,
      "learning_rate": 5.921888214647429e-06,
      "loss": 1.66198082,
      "memory(GiB)": 117.38,
      "step": 36970,
      "train_speed(iter/s)": 1.638306
    },
    {
      "acc": 0.65282521,
      "epoch": 0.9379756468797564,
      "grad_norm": 5.4375,
      "learning_rate": 5.920857548856064e-06,
      "loss": 1.57444878,
      "memory(GiB)": 117.38,
      "step": 36975,
      "train_speed(iter/s)": 1.63833
    },
    {
      "acc": 0.64693775,
      "epoch": 0.9381024860476915,
      "grad_norm": 7.78125,
      "learning_rate": 5.919826842561274e-06,
      "loss": 1.61087914,
      "memory(GiB)": 117.38,
      "step": 36980,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.65350242,
      "epoch": 0.9382293252156266,
      "grad_norm": 5.6875,
      "learning_rate": 5.91879609580839e-06,
      "loss": 1.64501724,
      "memory(GiB)": 117.38,
      "step": 36985,
      "train_speed(iter/s)": 1.63838
    },
    {
      "acc": 0.64666762,
      "epoch": 0.9383561643835616,
      "grad_norm": 7.78125,
      "learning_rate": 5.917765308642754e-06,
      "loss": 1.66936913,
      "memory(GiB)": 117.38,
      "step": 36990,
      "train_speed(iter/s)": 1.638403
    },
    {
      "acc": 0.65307064,
      "epoch": 0.9384830035514967,
      "grad_norm": 8.25,
      "learning_rate": 5.9167344811097014e-06,
      "loss": 1.66945553,
      "memory(GiB)": 117.38,
      "step": 36995,
      "train_speed(iter/s)": 1.638427
    },
    {
      "acc": 0.66428313,
      "epoch": 0.9386098427194317,
      "grad_norm": 6.40625,
      "learning_rate": 5.9157036132545735e-06,
      "loss": 1.53685236,
      "memory(GiB)": 117.38,
      "step": 37000,
      "train_speed(iter/s)": 1.638451
    },
    {
      "epoch": 0.9386098427194317,
      "eval_acc": 0.6460141545388289,
      "eval_loss": 1.5747417211532593,
      "eval_runtime": 58.4241,
      "eval_samples_per_second": 109.03,
      "eval_steps_per_second": 27.266,
      "step": 37000
    },
    {
      "acc": 0.67087278,
      "epoch": 0.9387366818873668,
      "grad_norm": 7.1875,
      "learning_rate": 5.914672705122713e-06,
      "loss": 1.61551094,
      "memory(GiB)": 117.38,
      "step": 37005,
      "train_speed(iter/s)": 1.633941
    },
    {
      "acc": 0.65566669,
      "epoch": 0.9388635210553019,
      "grad_norm": 6.625,
      "learning_rate": 5.9136417567594615e-06,
      "loss": 1.61629143,
      "memory(GiB)": 117.38,
      "step": 37010,
      "train_speed(iter/s)": 1.633966
    },
    {
      "acc": 0.65871449,
      "epoch": 0.9389903602232369,
      "grad_norm": 6.0625,
      "learning_rate": 5.9126107682101675e-06,
      "loss": 1.65129185,
      "memory(GiB)": 117.38,
      "step": 37015,
      "train_speed(iter/s)": 1.63399
    },
    {
      "acc": 0.63790622,
      "epoch": 0.939117199391172,
      "grad_norm": 4.1875,
      "learning_rate": 5.911579739520178e-06,
      "loss": 1.61802025,
      "memory(GiB)": 117.38,
      "step": 37020,
      "train_speed(iter/s)": 1.634013
    },
    {
      "acc": 0.64074287,
      "epoch": 0.9392440385591071,
      "grad_norm": 6.125,
      "learning_rate": 5.91054867073484e-06,
      "loss": 1.66344681,
      "memory(GiB)": 117.38,
      "step": 37025,
      "train_speed(iter/s)": 1.634037
    },
    {
      "acc": 0.64235535,
      "epoch": 0.9393708777270421,
      "grad_norm": 6.3125,
      "learning_rate": 5.909517561899508e-06,
      "loss": 1.6569458,
      "memory(GiB)": 117.38,
      "step": 37030,
      "train_speed(iter/s)": 1.634062
    },
    {
      "acc": 0.62768097,
      "epoch": 0.9394977168949772,
      "grad_norm": 5.78125,
      "learning_rate": 5.908486413059532e-06,
      "loss": 1.65637264,
      "memory(GiB)": 117.38,
      "step": 37035,
      "train_speed(iter/s)": 1.634088
    },
    {
      "acc": 0.65163279,
      "epoch": 0.9396245560629122,
      "grad_norm": 4.5,
      "learning_rate": 5.907455224260268e-06,
      "loss": 1.6116024,
      "memory(GiB)": 117.38,
      "step": 37040,
      "train_speed(iter/s)": 1.634111
    },
    {
      "acc": 0.64798603,
      "epoch": 0.9397513952308473,
      "grad_norm": 6.46875,
      "learning_rate": 5.9064239955470704e-06,
      "loss": 1.64316425,
      "memory(GiB)": 117.38,
      "step": 37045,
      "train_speed(iter/s)": 1.634136
    },
    {
      "acc": 0.63081484,
      "epoch": 0.9398782343987824,
      "grad_norm": 5.90625,
      "learning_rate": 5.9053927269653e-06,
      "loss": 1.71127014,
      "memory(GiB)": 117.38,
      "step": 37050,
      "train_speed(iter/s)": 1.634158
    },
    {
      "acc": 0.66582217,
      "epoch": 0.9400050735667174,
      "grad_norm": 5.5,
      "learning_rate": 5.904361418560314e-06,
      "loss": 1.59358034,
      "memory(GiB)": 117.38,
      "step": 37055,
      "train_speed(iter/s)": 1.634182
    },
    {
      "acc": 0.6632164,
      "epoch": 0.9401319127346525,
      "grad_norm": 6.84375,
      "learning_rate": 5.903330070377477e-06,
      "loss": 1.62826271,
      "memory(GiB)": 117.38,
      "step": 37060,
      "train_speed(iter/s)": 1.634206
    },
    {
      "acc": 0.66595354,
      "epoch": 0.9402587519025876,
      "grad_norm": 6.21875,
      "learning_rate": 5.902298682462147e-06,
      "loss": 1.5278923,
      "memory(GiB)": 117.38,
      "step": 37065,
      "train_speed(iter/s)": 1.634229
    },
    {
      "acc": 0.66164331,
      "epoch": 0.9403855910705226,
      "grad_norm": 6.6875,
      "learning_rate": 5.901267254859695e-06,
      "loss": 1.6072525,
      "memory(GiB)": 117.38,
      "step": 37070,
      "train_speed(iter/s)": 1.634253
    },
    {
      "acc": 0.65205784,
      "epoch": 0.9405124302384577,
      "grad_norm": 4.875,
      "learning_rate": 5.900235787615485e-06,
      "loss": 1.66901894,
      "memory(GiB)": 117.38,
      "step": 37075,
      "train_speed(iter/s)": 1.634276
    },
    {
      "acc": 0.65549827,
      "epoch": 0.9406392694063926,
      "grad_norm": 6.375,
      "learning_rate": 5.8992042807748866e-06,
      "loss": 1.58327837,
      "memory(GiB)": 117.38,
      "step": 37080,
      "train_speed(iter/s)": 1.634301
    },
    {
      "acc": 0.65217376,
      "epoch": 0.9407661085743277,
      "grad_norm": 5.75,
      "learning_rate": 5.898172734383267e-06,
      "loss": 1.64416962,
      "memory(GiB)": 117.38,
      "step": 37085,
      "train_speed(iter/s)": 1.634325
    },
    {
      "acc": 0.64489574,
      "epoch": 0.9408929477422628,
      "grad_norm": 6.21875,
      "learning_rate": 5.897141148486003e-06,
      "loss": 1.67693863,
      "memory(GiB)": 117.38,
      "step": 37090,
      "train_speed(iter/s)": 1.634349
    },
    {
      "acc": 0.66178365,
      "epoch": 0.9410197869101978,
      "grad_norm": 5.96875,
      "learning_rate": 5.8961095231284645e-06,
      "loss": 1.56035357,
      "memory(GiB)": 117.38,
      "step": 37095,
      "train_speed(iter/s)": 1.634373
    },
    {
      "acc": 0.64687643,
      "epoch": 0.9411466260781329,
      "grad_norm": 5.21875,
      "learning_rate": 5.895077858356029e-06,
      "loss": 1.62201366,
      "memory(GiB)": 117.38,
      "step": 37100,
      "train_speed(iter/s)": 1.634394
    },
    {
      "acc": 0.6479135,
      "epoch": 0.941273465246068,
      "grad_norm": 5.9375,
      "learning_rate": 5.8940461542140725e-06,
      "loss": 1.60642376,
      "memory(GiB)": 117.38,
      "step": 37105,
      "train_speed(iter/s)": 1.634416
    },
    {
      "acc": 0.66524668,
      "epoch": 0.941400304414003,
      "grad_norm": 6.3125,
      "learning_rate": 5.893014410747975e-06,
      "loss": 1.5475399,
      "memory(GiB)": 117.38,
      "step": 37110,
      "train_speed(iter/s)": 1.634441
    },
    {
      "acc": 0.66550741,
      "epoch": 0.9415271435819381,
      "grad_norm": 5.53125,
      "learning_rate": 5.891982628003114e-06,
      "loss": 1.58927517,
      "memory(GiB)": 117.38,
      "step": 37115,
      "train_speed(iter/s)": 1.634466
    },
    {
      "acc": 0.64765778,
      "epoch": 0.9416539827498731,
      "grad_norm": 6.9375,
      "learning_rate": 5.890950806024879e-06,
      "loss": 1.59349022,
      "memory(GiB)": 117.38,
      "step": 37120,
      "train_speed(iter/s)": 1.634491
    },
    {
      "acc": 0.65533237,
      "epoch": 0.9417808219178082,
      "grad_norm": 5.84375,
      "learning_rate": 5.889918944858647e-06,
      "loss": 1.62578106,
      "memory(GiB)": 117.38,
      "step": 37125,
      "train_speed(iter/s)": 1.634516
    },
    {
      "acc": 0.66858497,
      "epoch": 0.9419076610857433,
      "grad_norm": 5.90625,
      "learning_rate": 5.888887044549808e-06,
      "loss": 1.54972477,
      "memory(GiB)": 117.38,
      "step": 37130,
      "train_speed(iter/s)": 1.63454
    },
    {
      "acc": 0.64566202,
      "epoch": 0.9420345002536783,
      "grad_norm": 5.59375,
      "learning_rate": 5.887855105143746e-06,
      "loss": 1.61215782,
      "memory(GiB)": 117.38,
      "step": 37135,
      "train_speed(iter/s)": 1.634564
    },
    {
      "acc": 0.65946627,
      "epoch": 0.9421613394216134,
      "grad_norm": 4.96875,
      "learning_rate": 5.886823126685855e-06,
      "loss": 1.56578712,
      "memory(GiB)": 117.38,
      "step": 37140,
      "train_speed(iter/s)": 1.634589
    },
    {
      "acc": 0.65658712,
      "epoch": 0.9422881785895485,
      "grad_norm": 6.1875,
      "learning_rate": 5.8857911092215214e-06,
      "loss": 1.6559248,
      "memory(GiB)": 117.38,
      "step": 37145,
      "train_speed(iter/s)": 1.634614
    },
    {
      "acc": 0.65072589,
      "epoch": 0.9424150177574835,
      "grad_norm": 5.21875,
      "learning_rate": 5.884759052796142e-06,
      "loss": 1.66819572,
      "memory(GiB)": 117.38,
      "step": 37150,
      "train_speed(iter/s)": 1.634639
    },
    {
      "acc": 0.65706954,
      "epoch": 0.9425418569254186,
      "grad_norm": 5.28125,
      "learning_rate": 5.883726957455108e-06,
      "loss": 1.60351677,
      "memory(GiB)": 117.38,
      "step": 37155,
      "train_speed(iter/s)": 1.634663
    },
    {
      "acc": 0.64331613,
      "epoch": 0.9426686960933536,
      "grad_norm": 5.25,
      "learning_rate": 5.8826948232438176e-06,
      "loss": 1.64139805,
      "memory(GiB)": 117.38,
      "step": 37160,
      "train_speed(iter/s)": 1.634688
    },
    {
      "acc": 0.66055131,
      "epoch": 0.9427955352612887,
      "grad_norm": 4.90625,
      "learning_rate": 5.881662650207667e-06,
      "loss": 1.5450551,
      "memory(GiB)": 117.38,
      "step": 37165,
      "train_speed(iter/s)": 1.634712
    },
    {
      "acc": 0.67571325,
      "epoch": 0.9429223744292238,
      "grad_norm": 6.40625,
      "learning_rate": 5.880630438392057e-06,
      "loss": 1.56790161,
      "memory(GiB)": 117.38,
      "step": 37170,
      "train_speed(iter/s)": 1.634737
    },
    {
      "acc": 0.66040192,
      "epoch": 0.9430492135971588,
      "grad_norm": 5.65625,
      "learning_rate": 5.879598187842389e-06,
      "loss": 1.55055294,
      "memory(GiB)": 117.38,
      "step": 37175,
      "train_speed(iter/s)": 1.634762
    },
    {
      "acc": 0.6687181,
      "epoch": 0.9431760527650939,
      "grad_norm": 9.8125,
      "learning_rate": 5.878565898604066e-06,
      "loss": 1.60814495,
      "memory(GiB)": 117.38,
      "step": 37180,
      "train_speed(iter/s)": 1.634786
    },
    {
      "acc": 0.66111364,
      "epoch": 0.943302891933029,
      "grad_norm": 5.34375,
      "learning_rate": 5.87753357072249e-06,
      "loss": 1.54269657,
      "memory(GiB)": 117.38,
      "step": 37185,
      "train_speed(iter/s)": 1.634809
    },
    {
      "acc": 0.66401286,
      "epoch": 0.943429731100964,
      "grad_norm": 6.6875,
      "learning_rate": 5.876501204243072e-06,
      "loss": 1.59320698,
      "memory(GiB)": 117.38,
      "step": 37190,
      "train_speed(iter/s)": 1.634834
    },
    {
      "acc": 0.66640902,
      "epoch": 0.943556570268899,
      "grad_norm": 8.25,
      "learning_rate": 5.875468799211217e-06,
      "loss": 1.55528622,
      "memory(GiB)": 117.38,
      "step": 37195,
      "train_speed(iter/s)": 1.634859
    },
    {
      "acc": 0.65090442,
      "epoch": 0.943683409436834,
      "grad_norm": 4.6875,
      "learning_rate": 5.874436355672337e-06,
      "loss": 1.55439949,
      "memory(GiB)": 117.38,
      "step": 37200,
      "train_speed(iter/s)": 1.634883
    },
    {
      "acc": 0.65309072,
      "epoch": 0.9438102486047691,
      "grad_norm": 7.78125,
      "learning_rate": 5.873403873671839e-06,
      "loss": 1.62836189,
      "memory(GiB)": 117.38,
      "step": 37205,
      "train_speed(iter/s)": 1.634907
    },
    {
      "acc": 0.65641594,
      "epoch": 0.9439370877727042,
      "grad_norm": 6.09375,
      "learning_rate": 5.872371353255142e-06,
      "loss": 1.61465721,
      "memory(GiB)": 117.38,
      "step": 37210,
      "train_speed(iter/s)": 1.634933
    },
    {
      "acc": 0.67033558,
      "epoch": 0.9440639269406392,
      "grad_norm": 6.15625,
      "learning_rate": 5.871338794467656e-06,
      "loss": 1.58546095,
      "memory(GiB)": 117.38,
      "step": 37215,
      "train_speed(iter/s)": 1.634958
    },
    {
      "acc": 0.65380259,
      "epoch": 0.9441907661085743,
      "grad_norm": 5.4375,
      "learning_rate": 5.8703061973548e-06,
      "loss": 1.57519588,
      "memory(GiB)": 117.38,
      "step": 37220,
      "train_speed(iter/s)": 1.634982
    },
    {
      "acc": 0.6597271,
      "epoch": 0.9443176052765094,
      "grad_norm": 4.3125,
      "learning_rate": 5.869273561961992e-06,
      "loss": 1.56505375,
      "memory(GiB)": 117.38,
      "step": 37225,
      "train_speed(iter/s)": 1.635005
    },
    {
      "acc": 0.67193351,
      "epoch": 0.9444444444444444,
      "grad_norm": 5.75,
      "learning_rate": 5.8682408883346535e-06,
      "loss": 1.59879799,
      "memory(GiB)": 117.38,
      "step": 37230,
      "train_speed(iter/s)": 1.635029
    },
    {
      "acc": 0.66186018,
      "epoch": 0.9445712836123795,
      "grad_norm": 5.90625,
      "learning_rate": 5.867208176518202e-06,
      "loss": 1.58376503,
      "memory(GiB)": 117.38,
      "step": 37235,
      "train_speed(iter/s)": 1.635053
    },
    {
      "acc": 0.6639473,
      "epoch": 0.9446981227803145,
      "grad_norm": 8.0625,
      "learning_rate": 5.866175426558064e-06,
      "loss": 1.56097908,
      "memory(GiB)": 117.38,
      "step": 37240,
      "train_speed(iter/s)": 1.635076
    },
    {
      "acc": 0.65162611,
      "epoch": 0.9448249619482496,
      "grad_norm": 6.0,
      "learning_rate": 5.865142638499664e-06,
      "loss": 1.66612511,
      "memory(GiB)": 117.38,
      "step": 37245,
      "train_speed(iter/s)": 1.6351
    },
    {
      "acc": 0.63316226,
      "epoch": 0.9449518011161847,
      "grad_norm": 5.78125,
      "learning_rate": 5.864109812388426e-06,
      "loss": 1.74475288,
      "memory(GiB)": 117.38,
      "step": 37250,
      "train_speed(iter/s)": 1.635125
    },
    {
      "acc": 0.65003152,
      "epoch": 0.9450786402841197,
      "grad_norm": 5.03125,
      "learning_rate": 5.863076948269782e-06,
      "loss": 1.58734035,
      "memory(GiB)": 117.38,
      "step": 37255,
      "train_speed(iter/s)": 1.635148
    },
    {
      "acc": 0.66514273,
      "epoch": 0.9452054794520548,
      "grad_norm": 5.125,
      "learning_rate": 5.862044046189162e-06,
      "loss": 1.62495632,
      "memory(GiB)": 117.38,
      "step": 37260,
      "train_speed(iter/s)": 1.635173
    },
    {
      "acc": 0.66927495,
      "epoch": 0.9453323186199899,
      "grad_norm": 6.25,
      "learning_rate": 5.8610111061919924e-06,
      "loss": 1.59755268,
      "memory(GiB)": 117.38,
      "step": 37265,
      "train_speed(iter/s)": 1.635197
    },
    {
      "acc": 0.64952631,
      "epoch": 0.9454591577879249,
      "grad_norm": 6.0625,
      "learning_rate": 5.859978128323713e-06,
      "loss": 1.63681412,
      "memory(GiB)": 117.38,
      "step": 37270,
      "train_speed(iter/s)": 1.635222
    },
    {
      "acc": 0.65714588,
      "epoch": 0.94558599695586,
      "grad_norm": 5.75,
      "learning_rate": 5.858945112629755e-06,
      "loss": 1.5896163,
      "memory(GiB)": 117.38,
      "step": 37275,
      "train_speed(iter/s)": 1.635245
    },
    {
      "acc": 0.66810346,
      "epoch": 0.945712836123795,
      "grad_norm": 5.25,
      "learning_rate": 5.857912059155557e-06,
      "loss": 1.58438749,
      "memory(GiB)": 117.38,
      "step": 37280,
      "train_speed(iter/s)": 1.635269
    },
    {
      "acc": 0.66742859,
      "epoch": 0.9458396752917301,
      "grad_norm": 5.625,
      "learning_rate": 5.856878967946555e-06,
      "loss": 1.5479702,
      "memory(GiB)": 117.38,
      "step": 37285,
      "train_speed(iter/s)": 1.635294
    },
    {
      "acc": 0.64885755,
      "epoch": 0.9459665144596652,
      "grad_norm": 6.21875,
      "learning_rate": 5.855845839048191e-06,
      "loss": 1.63455544,
      "memory(GiB)": 117.38,
      "step": 37290,
      "train_speed(iter/s)": 1.635319
    },
    {
      "acc": 0.65881367,
      "epoch": 0.9460933536276002,
      "grad_norm": 5.625,
      "learning_rate": 5.854812672505906e-06,
      "loss": 1.67388725,
      "memory(GiB)": 117.38,
      "step": 37295,
      "train_speed(iter/s)": 1.635342
    },
    {
      "acc": 0.65646925,
      "epoch": 0.9462201927955353,
      "grad_norm": 6.0625,
      "learning_rate": 5.853779468365144e-06,
      "loss": 1.55733929,
      "memory(GiB)": 117.38,
      "step": 37300,
      "train_speed(iter/s)": 1.635367
    },
    {
      "acc": 0.64865289,
      "epoch": 0.9463470319634704,
      "grad_norm": 5.59375,
      "learning_rate": 5.852746226671348e-06,
      "loss": 1.66837769,
      "memory(GiB)": 117.38,
      "step": 37305,
      "train_speed(iter/s)": 1.635392
    },
    {
      "acc": 0.64205294,
      "epoch": 0.9464738711314054,
      "grad_norm": 5.34375,
      "learning_rate": 5.851712947469966e-06,
      "loss": 1.61525612,
      "memory(GiB)": 117.38,
      "step": 37310,
      "train_speed(iter/s)": 1.635417
    },
    {
      "acc": 0.64573021,
      "epoch": 0.9466007102993405,
      "grad_norm": 5.53125,
      "learning_rate": 5.850679630806446e-06,
      "loss": 1.64777679,
      "memory(GiB)": 117.38,
      "step": 37315,
      "train_speed(iter/s)": 1.635441
    },
    {
      "acc": 0.65656967,
      "epoch": 0.9467275494672754,
      "grad_norm": 5.0625,
      "learning_rate": 5.849646276726237e-06,
      "loss": 1.62326584,
      "memory(GiB)": 117.38,
      "step": 37320,
      "train_speed(iter/s)": 1.635464
    },
    {
      "acc": 0.65276423,
      "epoch": 0.9468543886352105,
      "grad_norm": 6.1875,
      "learning_rate": 5.848612885274792e-06,
      "loss": 1.64401741,
      "memory(GiB)": 117.38,
      "step": 37325,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.64332066,
      "epoch": 0.9469812278031456,
      "grad_norm": 5.46875,
      "learning_rate": 5.847579456497564e-06,
      "loss": 1.67527542,
      "memory(GiB)": 117.38,
      "step": 37330,
      "train_speed(iter/s)": 1.635511
    },
    {
      "acc": 0.65072522,
      "epoch": 0.9471080669710806,
      "grad_norm": 5.53125,
      "learning_rate": 5.8465459904400065e-06,
      "loss": 1.5624176,
      "memory(GiB)": 117.38,
      "step": 37335,
      "train_speed(iter/s)": 1.635535
    },
    {
      "acc": 0.66717434,
      "epoch": 0.9472349061390157,
      "grad_norm": 6.15625,
      "learning_rate": 5.845512487147579e-06,
      "loss": 1.611129,
      "memory(GiB)": 117.38,
      "step": 37340,
      "train_speed(iter/s)": 1.635557
    },
    {
      "acc": 0.65510526,
      "epoch": 0.9473617453069508,
      "grad_norm": 6.34375,
      "learning_rate": 5.844478946665733e-06,
      "loss": 1.60988865,
      "memory(GiB)": 117.38,
      "step": 37345,
      "train_speed(iter/s)": 1.635582
    },
    {
      "acc": 0.65682077,
      "epoch": 0.9474885844748858,
      "grad_norm": 5.03125,
      "learning_rate": 5.843445369039937e-06,
      "loss": 1.6545742,
      "memory(GiB)": 117.38,
      "step": 37350,
      "train_speed(iter/s)": 1.635605
    },
    {
      "acc": 0.67353334,
      "epoch": 0.9476154236428209,
      "grad_norm": 7.1875,
      "learning_rate": 5.842411754315645e-06,
      "loss": 1.51080999,
      "memory(GiB)": 117.38,
      "step": 37355,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.651086,
      "epoch": 0.9477422628107559,
      "grad_norm": 5.25,
      "learning_rate": 5.841378102538324e-06,
      "loss": 1.65048065,
      "memory(GiB)": 117.38,
      "step": 37360,
      "train_speed(iter/s)": 1.635651
    },
    {
      "acc": 0.6626194,
      "epoch": 0.947869101978691,
      "grad_norm": 6.3125,
      "learning_rate": 5.840344413753438e-06,
      "loss": 1.55746279,
      "memory(GiB)": 117.38,
      "step": 37365,
      "train_speed(iter/s)": 1.635674
    },
    {
      "acc": 0.64949627,
      "epoch": 0.9479959411466261,
      "grad_norm": 5.03125,
      "learning_rate": 5.8393106880064535e-06,
      "loss": 1.60622749,
      "memory(GiB)": 117.38,
      "step": 37370,
      "train_speed(iter/s)": 1.635699
    },
    {
      "acc": 0.65106373,
      "epoch": 0.9481227803145611,
      "grad_norm": 5.8125,
      "learning_rate": 5.838276925342836e-06,
      "loss": 1.58245268,
      "memory(GiB)": 117.38,
      "step": 37375,
      "train_speed(iter/s)": 1.635721
    },
    {
      "acc": 0.66040907,
      "epoch": 0.9482496194824962,
      "grad_norm": 5.78125,
      "learning_rate": 5.837243125808058e-06,
      "loss": 1.57163219,
      "memory(GiB)": 117.38,
      "step": 37380,
      "train_speed(iter/s)": 1.635743
    },
    {
      "acc": 0.64261847,
      "epoch": 0.9483764586504313,
      "grad_norm": 6.75,
      "learning_rate": 5.8362092894475886e-06,
      "loss": 1.69431362,
      "memory(GiB)": 117.38,
      "step": 37385,
      "train_speed(iter/s)": 1.635767
    },
    {
      "acc": 0.64906178,
      "epoch": 0.9485032978183663,
      "grad_norm": 5.21875,
      "learning_rate": 5.835175416306901e-06,
      "loss": 1.6429142,
      "memory(GiB)": 117.38,
      "step": 37390,
      "train_speed(iter/s)": 1.635791
    },
    {
      "acc": 0.65585403,
      "epoch": 0.9486301369863014,
      "grad_norm": 6.21875,
      "learning_rate": 5.83414150643147e-06,
      "loss": 1.55377531,
      "memory(GiB)": 117.38,
      "step": 37395,
      "train_speed(iter/s)": 1.635814
    },
    {
      "acc": 0.63683319,
      "epoch": 0.9487569761542364,
      "grad_norm": 6.5,
      "learning_rate": 5.833107559866772e-06,
      "loss": 1.55389633,
      "memory(GiB)": 117.38,
      "step": 37400,
      "train_speed(iter/s)": 1.635836
    },
    {
      "acc": 0.65886316,
      "epoch": 0.9488838153221715,
      "grad_norm": 6.4375,
      "learning_rate": 5.832073576658282e-06,
      "loss": 1.60462189,
      "memory(GiB)": 117.38,
      "step": 37405,
      "train_speed(iter/s)": 1.63586
    },
    {
      "acc": 0.66132979,
      "epoch": 0.9490106544901066,
      "grad_norm": 6.25,
      "learning_rate": 5.831039556851485e-06,
      "loss": 1.57812099,
      "memory(GiB)": 117.38,
      "step": 37410,
      "train_speed(iter/s)": 1.635882
    },
    {
      "acc": 0.64156971,
      "epoch": 0.9491374936580416,
      "grad_norm": 5.71875,
      "learning_rate": 5.8300055004918535e-06,
      "loss": 1.60821972,
      "memory(GiB)": 117.38,
      "step": 37415,
      "train_speed(iter/s)": 1.635903
    },
    {
      "acc": 0.64402447,
      "epoch": 0.9492643328259767,
      "grad_norm": 6.25,
      "learning_rate": 5.828971407624877e-06,
      "loss": 1.65598946,
      "memory(GiB)": 117.38,
      "step": 37420,
      "train_speed(iter/s)": 1.635926
    },
    {
      "acc": 0.65985446,
      "epoch": 0.9493911719939118,
      "grad_norm": 5.0625,
      "learning_rate": 5.827937278296037e-06,
      "loss": 1.59278069,
      "memory(GiB)": 117.38,
      "step": 37425,
      "train_speed(iter/s)": 1.635949
    },
    {
      "acc": 0.67123766,
      "epoch": 0.9495180111618468,
      "grad_norm": 6.34375,
      "learning_rate": 5.826903112550819e-06,
      "loss": 1.58408718,
      "memory(GiB)": 117.38,
      "step": 37430,
      "train_speed(iter/s)": 1.635973
    },
    {
      "acc": 0.65343204,
      "epoch": 0.9496448503297819,
      "grad_norm": 6.03125,
      "learning_rate": 5.825868910434708e-06,
      "loss": 1.58439655,
      "memory(GiB)": 117.38,
      "step": 37435,
      "train_speed(iter/s)": 1.635997
    },
    {
      "acc": 0.67348738,
      "epoch": 0.9497716894977168,
      "grad_norm": 5.0625,
      "learning_rate": 5.824834671993197e-06,
      "loss": 1.54148617,
      "memory(GiB)": 117.38,
      "step": 37440,
      "train_speed(iter/s)": 1.63602
    },
    {
      "acc": 0.64992294,
      "epoch": 0.9498985286656519,
      "grad_norm": 6.1875,
      "learning_rate": 5.823800397271774e-06,
      "loss": 1.6309782,
      "memory(GiB)": 117.38,
      "step": 37445,
      "train_speed(iter/s)": 1.636043
    },
    {
      "acc": 0.65153122,
      "epoch": 0.950025367833587,
      "grad_norm": 9.0,
      "learning_rate": 5.822766086315932e-06,
      "loss": 1.62790108,
      "memory(GiB)": 117.38,
      "step": 37450,
      "train_speed(iter/s)": 1.636066
    },
    {
      "acc": 0.65514021,
      "epoch": 0.950152207001522,
      "grad_norm": 5.28125,
      "learning_rate": 5.821731739171164e-06,
      "loss": 1.62272606,
      "memory(GiB)": 117.38,
      "step": 37455,
      "train_speed(iter/s)": 1.63609
    },
    {
      "acc": 0.66500363,
      "epoch": 0.9502790461694571,
      "grad_norm": 5.71875,
      "learning_rate": 5.820697355882965e-06,
      "loss": 1.65374069,
      "memory(GiB)": 117.38,
      "step": 37460,
      "train_speed(iter/s)": 1.636114
    },
    {
      "acc": 0.65905337,
      "epoch": 0.9504058853373922,
      "grad_norm": 6.46875,
      "learning_rate": 5.819662936496833e-06,
      "loss": 1.62111397,
      "memory(GiB)": 117.38,
      "step": 37465,
      "train_speed(iter/s)": 1.636136
    },
    {
      "acc": 0.6510251,
      "epoch": 0.9505327245053272,
      "grad_norm": 4.8125,
      "learning_rate": 5.818628481058265e-06,
      "loss": 1.53246517,
      "memory(GiB)": 117.38,
      "step": 37470,
      "train_speed(iter/s)": 1.63616
    },
    {
      "acc": 0.6442173,
      "epoch": 0.9506595636732623,
      "grad_norm": 5.65625,
      "learning_rate": 5.81759398961276e-06,
      "loss": 1.62716179,
      "memory(GiB)": 117.38,
      "step": 37475,
      "train_speed(iter/s)": 1.636183
    },
    {
      "acc": 0.65647416,
      "epoch": 0.9507864028411973,
      "grad_norm": 5.28125,
      "learning_rate": 5.816559462205824e-06,
      "loss": 1.57571383,
      "memory(GiB)": 117.38,
      "step": 37480,
      "train_speed(iter/s)": 1.636206
    },
    {
      "acc": 0.65683165,
      "epoch": 0.9509132420091324,
      "grad_norm": 6.1875,
      "learning_rate": 5.815524898882954e-06,
      "loss": 1.58756084,
      "memory(GiB)": 117.38,
      "step": 37485,
      "train_speed(iter/s)": 1.63623
    },
    {
      "acc": 0.67337599,
      "epoch": 0.9510400811770675,
      "grad_norm": 6.40625,
      "learning_rate": 5.8144902996896615e-06,
      "loss": 1.58814936,
      "memory(GiB)": 117.38,
      "step": 37490,
      "train_speed(iter/s)": 1.636254
    },
    {
      "acc": 0.6454134,
      "epoch": 0.9511669203450025,
      "grad_norm": 5.0625,
      "learning_rate": 5.813455664671446e-06,
      "loss": 1.65201683,
      "memory(GiB)": 117.38,
      "step": 37495,
      "train_speed(iter/s)": 1.636278
    },
    {
      "acc": 0.67092962,
      "epoch": 0.9512937595129376,
      "grad_norm": 6.03125,
      "learning_rate": 5.812420993873819e-06,
      "loss": 1.4710536,
      "memory(GiB)": 117.38,
      "step": 37500,
      "train_speed(iter/s)": 1.6363
    },
    {
      "acc": 0.6575067,
      "epoch": 0.9514205986808727,
      "grad_norm": 5.90625,
      "learning_rate": 5.81138628734229e-06,
      "loss": 1.59180889,
      "memory(GiB)": 117.38,
      "step": 37505,
      "train_speed(iter/s)": 1.636324
    },
    {
      "acc": 0.65847254,
      "epoch": 0.9515474378488077,
      "grad_norm": 5.8125,
      "learning_rate": 5.81035154512237e-06,
      "loss": 1.56397066,
      "memory(GiB)": 117.38,
      "step": 37510,
      "train_speed(iter/s)": 1.636348
    },
    {
      "acc": 0.64296298,
      "epoch": 0.9516742770167428,
      "grad_norm": 5.90625,
      "learning_rate": 5.809316767259571e-06,
      "loss": 1.65934486,
      "memory(GiB)": 117.38,
      "step": 37515,
      "train_speed(iter/s)": 1.636371
    },
    {
      "acc": 0.64423733,
      "epoch": 0.9518011161846778,
      "grad_norm": 5.5,
      "learning_rate": 5.808281953799408e-06,
      "loss": 1.62972374,
      "memory(GiB)": 117.38,
      "step": 37520,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.63471594,
      "epoch": 0.9519279553526129,
      "grad_norm": 6.5625,
      "learning_rate": 5.807247104787395e-06,
      "loss": 1.63079109,
      "memory(GiB)": 117.38,
      "step": 37525,
      "train_speed(iter/s)": 1.636419
    },
    {
      "acc": 0.66951866,
      "epoch": 0.952054794520548,
      "grad_norm": 6.0,
      "learning_rate": 5.806212220269049e-06,
      "loss": 1.59792833,
      "memory(GiB)": 117.38,
      "step": 37530,
      "train_speed(iter/s)": 1.636444
    },
    {
      "acc": 0.6424468,
      "epoch": 0.952181633688483,
      "grad_norm": 5.59375,
      "learning_rate": 5.805177300289891e-06,
      "loss": 1.69618607,
      "memory(GiB)": 117.38,
      "step": 37535,
      "train_speed(iter/s)": 1.636466
    },
    {
      "acc": 0.64744864,
      "epoch": 0.9523084728564181,
      "grad_norm": 5.53125,
      "learning_rate": 5.804142344895441e-06,
      "loss": 1.60979137,
      "memory(GiB)": 117.38,
      "step": 37540,
      "train_speed(iter/s)": 1.63649
    },
    {
      "acc": 0.66371355,
      "epoch": 0.9524353120243532,
      "grad_norm": 6.53125,
      "learning_rate": 5.803107354131221e-06,
      "loss": 1.56906843,
      "memory(GiB)": 117.38,
      "step": 37545,
      "train_speed(iter/s)": 1.636512
    },
    {
      "acc": 0.64825001,
      "epoch": 0.9525621511922882,
      "grad_norm": 5.96875,
      "learning_rate": 5.802072328042753e-06,
      "loss": 1.64823627,
      "memory(GiB)": 117.38,
      "step": 37550,
      "train_speed(iter/s)": 1.636534
    },
    {
      "acc": 0.64479704,
      "epoch": 0.9526889903602233,
      "grad_norm": 5.03125,
      "learning_rate": 5.8010372666755625e-06,
      "loss": 1.63361969,
      "memory(GiB)": 117.38,
      "step": 37555,
      "train_speed(iter/s)": 1.636557
    },
    {
      "acc": 0.65289497,
      "epoch": 0.9528158295281582,
      "grad_norm": 6.4375,
      "learning_rate": 5.800002170075179e-06,
      "loss": 1.65221481,
      "memory(GiB)": 117.38,
      "step": 37560,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.68047056,
      "epoch": 0.9529426686960933,
      "grad_norm": 5.4375,
      "learning_rate": 5.798967038287125e-06,
      "loss": 1.53734779,
      "memory(GiB)": 117.38,
      "step": 37565,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.66665168,
      "epoch": 0.9530695078640284,
      "grad_norm": 7.21875,
      "learning_rate": 5.797931871356936e-06,
      "loss": 1.54502392,
      "memory(GiB)": 117.38,
      "step": 37570,
      "train_speed(iter/s)": 1.636626
    },
    {
      "acc": 0.65586324,
      "epoch": 0.9531963470319634,
      "grad_norm": 5.5,
      "learning_rate": 5.796896669330139e-06,
      "loss": 1.60042305,
      "memory(GiB)": 117.38,
      "step": 37575,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.64218402,
      "epoch": 0.9533231861998985,
      "grad_norm": 7.0,
      "learning_rate": 5.79586143225227e-06,
      "loss": 1.69684486,
      "memory(GiB)": 117.38,
      "step": 37580,
      "train_speed(iter/s)": 1.636672
    },
    {
      "acc": 0.64952283,
      "epoch": 0.9534500253678336,
      "grad_norm": 6.21875,
      "learning_rate": 5.79482616016886e-06,
      "loss": 1.66387482,
      "memory(GiB)": 117.38,
      "step": 37585,
      "train_speed(iter/s)": 1.636696
    },
    {
      "acc": 0.64950857,
      "epoch": 0.9535768645357686,
      "grad_norm": 7.1875,
      "learning_rate": 5.793790853125449e-06,
      "loss": 1.63628597,
      "memory(GiB)": 117.38,
      "step": 37590,
      "train_speed(iter/s)": 1.636719
    },
    {
      "acc": 0.65667815,
      "epoch": 0.9537037037037037,
      "grad_norm": 6.4375,
      "learning_rate": 5.792755511167572e-06,
      "loss": 1.64018726,
      "memory(GiB)": 117.38,
      "step": 37595,
      "train_speed(iter/s)": 1.636742
    },
    {
      "acc": 0.64910975,
      "epoch": 0.9538305428716387,
      "grad_norm": 6.0,
      "learning_rate": 5.7917201343407685e-06,
      "loss": 1.61706772,
      "memory(GiB)": 117.38,
      "step": 37600,
      "train_speed(iter/s)": 1.636766
    },
    {
      "acc": 0.6598978,
      "epoch": 0.9539573820395738,
      "grad_norm": 5.875,
      "learning_rate": 5.790684722690577e-06,
      "loss": 1.66795273,
      "memory(GiB)": 117.38,
      "step": 37605,
      "train_speed(iter/s)": 1.636789
    },
    {
      "acc": 0.65098424,
      "epoch": 0.9540842212075089,
      "grad_norm": 5.5625,
      "learning_rate": 5.789649276262542e-06,
      "loss": 1.64607563,
      "memory(GiB)": 117.38,
      "step": 37610,
      "train_speed(iter/s)": 1.636809
    },
    {
      "acc": 0.65861449,
      "epoch": 0.9542110603754439,
      "grad_norm": 6.65625,
      "learning_rate": 5.788613795102207e-06,
      "loss": 1.59759712,
      "memory(GiB)": 117.38,
      "step": 37615,
      "train_speed(iter/s)": 1.636833
    },
    {
      "acc": 0.65541153,
      "epoch": 0.954337899543379,
      "grad_norm": 5.1875,
      "learning_rate": 5.787578279255116e-06,
      "loss": 1.5532217,
      "memory(GiB)": 117.38,
      "step": 37620,
      "train_speed(iter/s)": 1.636856
    },
    {
      "acc": 0.66289849,
      "epoch": 0.9544647387113141,
      "grad_norm": 5.15625,
      "learning_rate": 5.786542728766815e-06,
      "loss": 1.57978144,
      "memory(GiB)": 117.38,
      "step": 37625,
      "train_speed(iter/s)": 1.63688
    },
    {
      "acc": 0.65254831,
      "epoch": 0.9545915778792491,
      "grad_norm": 9.125,
      "learning_rate": 5.785507143682856e-06,
      "loss": 1.59954548,
      "memory(GiB)": 117.38,
      "step": 37630,
      "train_speed(iter/s)": 1.636903
    },
    {
      "acc": 0.63040018,
      "epoch": 0.9547184170471842,
      "grad_norm": 5.53125,
      "learning_rate": 5.784471524048782e-06,
      "loss": 1.6585844,
      "memory(GiB)": 117.38,
      "step": 37635,
      "train_speed(iter/s)": 1.636926
    },
    {
      "acc": 0.66040239,
      "epoch": 0.9548452562151192,
      "grad_norm": 6.375,
      "learning_rate": 5.783435869910151e-06,
      "loss": 1.60758381,
      "memory(GiB)": 117.38,
      "step": 37640,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.65827379,
      "epoch": 0.9549720953830543,
      "grad_norm": 4.71875,
      "learning_rate": 5.782400181312511e-06,
      "loss": 1.63642273,
      "memory(GiB)": 117.38,
      "step": 37645,
      "train_speed(iter/s)": 1.636973
    },
    {
      "acc": 0.65990334,
      "epoch": 0.9550989345509894,
      "grad_norm": 5.28125,
      "learning_rate": 5.781364458301419e-06,
      "loss": 1.56688986,
      "memory(GiB)": 117.38,
      "step": 37650,
      "train_speed(iter/s)": 1.636995
    },
    {
      "acc": 0.66363926,
      "epoch": 0.9552257737189244,
      "grad_norm": 5.09375,
      "learning_rate": 5.780328700922427e-06,
      "loss": 1.60596752,
      "memory(GiB)": 117.38,
      "step": 37655,
      "train_speed(iter/s)": 1.637019
    },
    {
      "acc": 0.67590065,
      "epoch": 0.9553526128868595,
      "grad_norm": 5.25,
      "learning_rate": 5.779292909221097e-06,
      "loss": 1.58627148,
      "memory(GiB)": 117.38,
      "step": 37660,
      "train_speed(iter/s)": 1.637042
    },
    {
      "acc": 0.66148157,
      "epoch": 0.9554794520547946,
      "grad_norm": 5.78125,
      "learning_rate": 5.778257083242986e-06,
      "loss": 1.57995892,
      "memory(GiB)": 117.38,
      "step": 37665,
      "train_speed(iter/s)": 1.637066
    },
    {
      "acc": 0.65000353,
      "epoch": 0.9556062912227296,
      "grad_norm": 5.0,
      "learning_rate": 5.777221223033653e-06,
      "loss": 1.62859459,
      "memory(GiB)": 117.38,
      "step": 37670,
      "train_speed(iter/s)": 1.637089
    },
    {
      "acc": 0.64907913,
      "epoch": 0.9557331303906647,
      "grad_norm": 4.6875,
      "learning_rate": 5.77618532863866e-06,
      "loss": 1.6252327,
      "memory(GiB)": 117.38,
      "step": 37675,
      "train_speed(iter/s)": 1.637114
    },
    {
      "acc": 0.67598896,
      "epoch": 0.9558599695585996,
      "grad_norm": 7.53125,
      "learning_rate": 5.775149400103572e-06,
      "loss": 1.49997368,
      "memory(GiB)": 117.38,
      "step": 37680,
      "train_speed(iter/s)": 1.637137
    },
    {
      "acc": 0.65692668,
      "epoch": 0.9559868087265347,
      "grad_norm": 5.6875,
      "learning_rate": 5.774113437473953e-06,
      "loss": 1.5859129,
      "memory(GiB)": 117.38,
      "step": 37685,
      "train_speed(iter/s)": 1.637161
    },
    {
      "acc": 0.64873314,
      "epoch": 0.9561136478944698,
      "grad_norm": 4.78125,
      "learning_rate": 5.7730774407953675e-06,
      "loss": 1.56788731,
      "memory(GiB)": 117.38,
      "step": 37690,
      "train_speed(iter/s)": 1.637185
    },
    {
      "acc": 0.659622,
      "epoch": 0.9562404870624048,
      "grad_norm": 6.5,
      "learning_rate": 5.772041410113384e-06,
      "loss": 1.57235374,
      "memory(GiB)": 117.38,
      "step": 37695,
      "train_speed(iter/s)": 1.637209
    },
    {
      "acc": 0.64488158,
      "epoch": 0.9563673262303399,
      "grad_norm": 5.25,
      "learning_rate": 5.771005345473575e-06,
      "loss": 1.61084862,
      "memory(GiB)": 117.38,
      "step": 37700,
      "train_speed(iter/s)": 1.637235
    },
    {
      "acc": 0.65941591,
      "epoch": 0.956494165398275,
      "grad_norm": 4.96875,
      "learning_rate": 5.769969246921505e-06,
      "loss": 1.55018835,
      "memory(GiB)": 117.38,
      "step": 37705,
      "train_speed(iter/s)": 1.63726
    },
    {
      "acc": 0.64470463,
      "epoch": 0.95662100456621,
      "grad_norm": 5.1875,
      "learning_rate": 5.768933114502753e-06,
      "loss": 1.63372116,
      "memory(GiB)": 117.38,
      "step": 37710,
      "train_speed(iter/s)": 1.637286
    },
    {
      "acc": 0.66603556,
      "epoch": 0.9567478437341451,
      "grad_norm": 6.125,
      "learning_rate": 5.7678969482628875e-06,
      "loss": 1.55573626,
      "memory(GiB)": 117.38,
      "step": 37715,
      "train_speed(iter/s)": 1.637312
    },
    {
      "acc": 0.65465174,
      "epoch": 0.9568746829020801,
      "grad_norm": 8.375,
      "learning_rate": 5.766860748247488e-06,
      "loss": 1.55616789,
      "memory(GiB)": 117.38,
      "step": 37720,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.64066839,
      "epoch": 0.9570015220700152,
      "grad_norm": 5.09375,
      "learning_rate": 5.765824514502126e-06,
      "loss": 1.64387627,
      "memory(GiB)": 117.38,
      "step": 37725,
      "train_speed(iter/s)": 1.637362
    },
    {
      "acc": 0.67350359,
      "epoch": 0.9571283612379503,
      "grad_norm": 9.25,
      "learning_rate": 5.7647882470723846e-06,
      "loss": 1.58552437,
      "memory(GiB)": 117.38,
      "step": 37730,
      "train_speed(iter/s)": 1.637388
    },
    {
      "acc": 0.67741766,
      "epoch": 0.9572552004058853,
      "grad_norm": 8.5,
      "learning_rate": 5.763751946003842e-06,
      "loss": 1.53572369,
      "memory(GiB)": 117.38,
      "step": 37735,
      "train_speed(iter/s)": 1.637414
    },
    {
      "acc": 0.64968634,
      "epoch": 0.9573820395738204,
      "grad_norm": 5.28125,
      "learning_rate": 5.7627156113420775e-06,
      "loss": 1.55772839,
      "memory(GiB)": 117.38,
      "step": 37740,
      "train_speed(iter/s)": 1.637438
    },
    {
      "acc": 0.64974623,
      "epoch": 0.9575088787417555,
      "grad_norm": 6.34375,
      "learning_rate": 5.761679243132677e-06,
      "loss": 1.5931015,
      "memory(GiB)": 117.38,
      "step": 37745,
      "train_speed(iter/s)": 1.637462
    },
    {
      "acc": 0.64106536,
      "epoch": 0.9576357179096905,
      "grad_norm": 5.53125,
      "learning_rate": 5.760642841421222e-06,
      "loss": 1.60904064,
      "memory(GiB)": 117.38,
      "step": 37750,
      "train_speed(iter/s)": 1.637486
    },
    {
      "acc": 0.66763115,
      "epoch": 0.9577625570776256,
      "grad_norm": 6.5,
      "learning_rate": 5.759606406253299e-06,
      "loss": 1.55258694,
      "memory(GiB)": 117.38,
      "step": 37755,
      "train_speed(iter/s)": 1.63751
    },
    {
      "acc": 0.65933352,
      "epoch": 0.9578893962455606,
      "grad_norm": 5.03125,
      "learning_rate": 5.758569937674494e-06,
      "loss": 1.54921799,
      "memory(GiB)": 117.38,
      "step": 37760,
      "train_speed(iter/s)": 1.637534
    },
    {
      "acc": 0.66198144,
      "epoch": 0.9580162354134957,
      "grad_norm": 6.09375,
      "learning_rate": 5.7575334357303954e-06,
      "loss": 1.54654217,
      "memory(GiB)": 117.38,
      "step": 37765,
      "train_speed(iter/s)": 1.637557
    },
    {
      "acc": 0.65941048,
      "epoch": 0.9581430745814308,
      "grad_norm": 5.53125,
      "learning_rate": 5.756496900466596e-06,
      "loss": 1.58932486,
      "memory(GiB)": 117.38,
      "step": 37770,
      "train_speed(iter/s)": 1.637583
    },
    {
      "acc": 0.66474457,
      "epoch": 0.9582699137493658,
      "grad_norm": 5.28125,
      "learning_rate": 5.755460331928684e-06,
      "loss": 1.5787775,
      "memory(GiB)": 117.38,
      "step": 37775,
      "train_speed(iter/s)": 1.637609
    },
    {
      "acc": 0.64596987,
      "epoch": 0.9583967529173009,
      "grad_norm": 6.125,
      "learning_rate": 5.754423730162257e-06,
      "loss": 1.63899269,
      "memory(GiB)": 117.38,
      "step": 37780,
      "train_speed(iter/s)": 1.637633
    },
    {
      "acc": 0.65063429,
      "epoch": 0.958523592085236,
      "grad_norm": 5.40625,
      "learning_rate": 5.753387095212901e-06,
      "loss": 1.61835041,
      "memory(GiB)": 117.38,
      "step": 37785,
      "train_speed(iter/s)": 1.637657
    },
    {
      "acc": 0.65633698,
      "epoch": 0.958650431253171,
      "grad_norm": 7.28125,
      "learning_rate": 5.752350427126221e-06,
      "loss": 1.63100376,
      "memory(GiB)": 117.38,
      "step": 37790,
      "train_speed(iter/s)": 1.637682
    },
    {
      "acc": 0.65996699,
      "epoch": 0.958777270421106,
      "grad_norm": 5.25,
      "learning_rate": 5.751313725947808e-06,
      "loss": 1.54564714,
      "memory(GiB)": 117.38,
      "step": 37795,
      "train_speed(iter/s)": 1.637706
    },
    {
      "acc": 0.6700829,
      "epoch": 0.958904109589041,
      "grad_norm": 7.28125,
      "learning_rate": 5.7502769917232635e-06,
      "loss": 1.57003651,
      "memory(GiB)": 117.38,
      "step": 37800,
      "train_speed(iter/s)": 1.637732
    },
    {
      "acc": 0.64893789,
      "epoch": 0.9590309487569761,
      "grad_norm": 5.0625,
      "learning_rate": 5.7492402244981885e-06,
      "loss": 1.65374413,
      "memory(GiB)": 117.38,
      "step": 37805,
      "train_speed(iter/s)": 1.637755
    },
    {
      "acc": 0.65842857,
      "epoch": 0.9591577879249112,
      "grad_norm": 6.0,
      "learning_rate": 5.748203424318182e-06,
      "loss": 1.60634327,
      "memory(GiB)": 117.38,
      "step": 37810,
      "train_speed(iter/s)": 1.63778
    },
    {
      "acc": 0.67613182,
      "epoch": 0.9592846270928462,
      "grad_norm": 5.6875,
      "learning_rate": 5.747166591228849e-06,
      "loss": 1.54881172,
      "memory(GiB)": 117.38,
      "step": 37815,
      "train_speed(iter/s)": 1.637806
    },
    {
      "acc": 0.66187201,
      "epoch": 0.9594114662607813,
      "grad_norm": 5.71875,
      "learning_rate": 5.746129725275793e-06,
      "loss": 1.6008316,
      "memory(GiB)": 117.38,
      "step": 37820,
      "train_speed(iter/s)": 1.637832
    },
    {
      "acc": 0.65212364,
      "epoch": 0.9595383054287164,
      "grad_norm": 5.5,
      "learning_rate": 5.74509282650462e-06,
      "loss": 1.63671932,
      "memory(GiB)": 117.38,
      "step": 37825,
      "train_speed(iter/s)": 1.637857
    },
    {
      "acc": 0.64497232,
      "epoch": 0.9596651445966514,
      "grad_norm": 7.53125,
      "learning_rate": 5.744055894960938e-06,
      "loss": 1.6811491,
      "memory(GiB)": 117.38,
      "step": 37830,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.64194746,
      "epoch": 0.9597919837645865,
      "grad_norm": 6.03125,
      "learning_rate": 5.743018930690357e-06,
      "loss": 1.70499802,
      "memory(GiB)": 117.38,
      "step": 37835,
      "train_speed(iter/s)": 1.637908
    },
    {
      "acc": 0.65335422,
      "epoch": 0.9599188229325215,
      "grad_norm": 9.0,
      "learning_rate": 5.7419819337384855e-06,
      "loss": 1.60643768,
      "memory(GiB)": 117.38,
      "step": 37840,
      "train_speed(iter/s)": 1.637934
    },
    {
      "acc": 0.64465008,
      "epoch": 0.9600456621004566,
      "grad_norm": 5.9375,
      "learning_rate": 5.740944904150934e-06,
      "loss": 1.61140461,
      "memory(GiB)": 117.38,
      "step": 37845,
      "train_speed(iter/s)": 1.637961
    },
    {
      "acc": 0.65831637,
      "epoch": 0.9601725012683917,
      "grad_norm": 5.59375,
      "learning_rate": 5.739907841973321e-06,
      "loss": 1.56716118,
      "memory(GiB)": 117.38,
      "step": 37850,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.6521193,
      "epoch": 0.9602993404363267,
      "grad_norm": 6.125,
      "learning_rate": 5.738870747251255e-06,
      "loss": 1.63041363,
      "memory(GiB)": 117.38,
      "step": 37855,
      "train_speed(iter/s)": 1.63801
    },
    {
      "acc": 0.64439826,
      "epoch": 0.9604261796042618,
      "grad_norm": 5.71875,
      "learning_rate": 5.737833620030357e-06,
      "loss": 1.61178532,
      "memory(GiB)": 117.38,
      "step": 37860,
      "train_speed(iter/s)": 1.638036
    },
    {
      "acc": 0.68125067,
      "epoch": 0.9605530187721969,
      "grad_norm": 5.90625,
      "learning_rate": 5.7367964603562385e-06,
      "loss": 1.50592403,
      "memory(GiB)": 117.38,
      "step": 37865,
      "train_speed(iter/s)": 1.638062
    },
    {
      "acc": 0.67131348,
      "epoch": 0.9606798579401319,
      "grad_norm": 5.3125,
      "learning_rate": 5.7357592682745245e-06,
      "loss": 1.59699163,
      "memory(GiB)": 117.38,
      "step": 37870,
      "train_speed(iter/s)": 1.638086
    },
    {
      "acc": 0.67061272,
      "epoch": 0.960806697108067,
      "grad_norm": 6.3125,
      "learning_rate": 5.734722043830833e-06,
      "loss": 1.50260658,
      "memory(GiB)": 117.38,
      "step": 37875,
      "train_speed(iter/s)": 1.63811
    },
    {
      "acc": 0.65280647,
      "epoch": 0.960933536276002,
      "grad_norm": 6.28125,
      "learning_rate": 5.7336847870707855e-06,
      "loss": 1.64194069,
      "memory(GiB)": 117.38,
      "step": 37880,
      "train_speed(iter/s)": 1.638137
    },
    {
      "acc": 0.65304947,
      "epoch": 0.9610603754439371,
      "grad_norm": 5.5625,
      "learning_rate": 5.732647498040006e-06,
      "loss": 1.61398125,
      "memory(GiB)": 117.38,
      "step": 37885,
      "train_speed(iter/s)": 1.638162
    },
    {
      "acc": 0.64910769,
      "epoch": 0.9611872146118722,
      "grad_norm": 9.1875,
      "learning_rate": 5.731610176784118e-06,
      "loss": 1.6200182,
      "memory(GiB)": 117.38,
      "step": 37890,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.65103445,
      "epoch": 0.9613140537798072,
      "grad_norm": 5.40625,
      "learning_rate": 5.730572823348748e-06,
      "loss": 1.57979717,
      "memory(GiB)": 117.38,
      "step": 37895,
      "train_speed(iter/s)": 1.638213
    },
    {
      "acc": 0.65187325,
      "epoch": 0.9614408929477423,
      "grad_norm": 5.875,
      "learning_rate": 5.729535437779523e-06,
      "loss": 1.60123997,
      "memory(GiB)": 117.38,
      "step": 37900,
      "train_speed(iter/s)": 1.638236
    },
    {
      "acc": 0.63695345,
      "epoch": 0.9615677321156774,
      "grad_norm": 5.34375,
      "learning_rate": 5.728498020122073e-06,
      "loss": 1.65709114,
      "memory(GiB)": 117.38,
      "step": 37905,
      "train_speed(iter/s)": 1.638261
    },
    {
      "acc": 0.65966816,
      "epoch": 0.9616945712836124,
      "grad_norm": 5.9375,
      "learning_rate": 5.727460570422028e-06,
      "loss": 1.56923771,
      "memory(GiB)": 117.38,
      "step": 37910,
      "train_speed(iter/s)": 1.638285
    },
    {
      "acc": 0.67273316,
      "epoch": 0.9618214104515475,
      "grad_norm": 5.90625,
      "learning_rate": 5.726423088725017e-06,
      "loss": 1.58475723,
      "memory(GiB)": 117.38,
      "step": 37915,
      "train_speed(iter/s)": 1.638308
    },
    {
      "acc": 0.64787149,
      "epoch": 0.9619482496194824,
      "grad_norm": 4.75,
      "learning_rate": 5.725385575076677e-06,
      "loss": 1.59393101,
      "memory(GiB)": 117.38,
      "step": 37920,
      "train_speed(iter/s)": 1.638334
    },
    {
      "acc": 0.63942108,
      "epoch": 0.9620750887874175,
      "grad_norm": 6.09375,
      "learning_rate": 5.7243480295226405e-06,
      "loss": 1.64283791,
      "memory(GiB)": 117.38,
      "step": 37925,
      "train_speed(iter/s)": 1.638359
    },
    {
      "acc": 0.65549374,
      "epoch": 0.9622019279553526,
      "grad_norm": 5.46875,
      "learning_rate": 5.723310452108545e-06,
      "loss": 1.6001955,
      "memory(GiB)": 117.38,
      "step": 37930,
      "train_speed(iter/s)": 1.638385
    },
    {
      "acc": 0.63950868,
      "epoch": 0.9623287671232876,
      "grad_norm": 5.0625,
      "learning_rate": 5.722272842880023e-06,
      "loss": 1.70025711,
      "memory(GiB)": 117.38,
      "step": 37935,
      "train_speed(iter/s)": 1.638408
    },
    {
      "acc": 0.65101233,
      "epoch": 0.9624556062912227,
      "grad_norm": 5.375,
      "learning_rate": 5.7212352018827215e-06,
      "loss": 1.60794716,
      "memory(GiB)": 117.38,
      "step": 37940,
      "train_speed(iter/s)": 1.63843
    },
    {
      "acc": 0.66388993,
      "epoch": 0.9625824454591578,
      "grad_norm": 5.25,
      "learning_rate": 5.720197529162272e-06,
      "loss": 1.58975258,
      "memory(GiB)": 117.38,
      "step": 37945,
      "train_speed(iter/s)": 1.638456
    },
    {
      "acc": 0.66025572,
      "epoch": 0.9627092846270928,
      "grad_norm": 4.65625,
      "learning_rate": 5.719159824764321e-06,
      "loss": 1.53744736,
      "memory(GiB)": 117.38,
      "step": 37950,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.66164737,
      "epoch": 0.9628361237950279,
      "grad_norm": 5.65625,
      "learning_rate": 5.71812208873451e-06,
      "loss": 1.64038773,
      "memory(GiB)": 117.38,
      "step": 37955,
      "train_speed(iter/s)": 1.638503
    },
    {
      "acc": 0.65875645,
      "epoch": 0.9629629629629629,
      "grad_norm": 5.625,
      "learning_rate": 5.717084321118482e-06,
      "loss": 1.55650082,
      "memory(GiB)": 117.38,
      "step": 37960,
      "train_speed(iter/s)": 1.638527
    },
    {
      "acc": 0.6570097,
      "epoch": 0.963089802130898,
      "grad_norm": 6.46875,
      "learning_rate": 5.716046521961887e-06,
      "loss": 1.66541557,
      "memory(GiB)": 117.38,
      "step": 37965,
      "train_speed(iter/s)": 1.638551
    },
    {
      "acc": 0.66001363,
      "epoch": 0.9632166412988331,
      "grad_norm": 6.6875,
      "learning_rate": 5.715008691310366e-06,
      "loss": 1.61850758,
      "memory(GiB)": 117.38,
      "step": 37970,
      "train_speed(iter/s)": 1.638576
    },
    {
      "acc": 0.64226384,
      "epoch": 0.9633434804667681,
      "grad_norm": 7.15625,
      "learning_rate": 5.713970829209573e-06,
      "loss": 1.69995575,
      "memory(GiB)": 117.38,
      "step": 37975,
      "train_speed(iter/s)": 1.6386
    },
    {
      "acc": 0.65770597,
      "epoch": 0.9634703196347032,
      "grad_norm": 5.21875,
      "learning_rate": 5.712932935705153e-06,
      "loss": 1.64444675,
      "memory(GiB)": 117.38,
      "step": 37980,
      "train_speed(iter/s)": 1.638624
    },
    {
      "acc": 0.63616838,
      "epoch": 0.9635971588026383,
      "grad_norm": 5.5,
      "learning_rate": 5.711895010842762e-06,
      "loss": 1.69516068,
      "memory(GiB)": 117.38,
      "step": 37985,
      "train_speed(iter/s)": 1.638648
    },
    {
      "acc": 0.66105471,
      "epoch": 0.9637239979705733,
      "grad_norm": 6.1875,
      "learning_rate": 5.710857054668048e-06,
      "loss": 1.60404091,
      "memory(GiB)": 117.38,
      "step": 37990,
      "train_speed(iter/s)": 1.638673
    },
    {
      "acc": 0.66310992,
      "epoch": 0.9638508371385084,
      "grad_norm": 5.875,
      "learning_rate": 5.7098190672266675e-06,
      "loss": 1.60103741,
      "memory(GiB)": 117.38,
      "step": 37995,
      "train_speed(iter/s)": 1.638697
    },
    {
      "acc": 0.65080233,
      "epoch": 0.9639776763064434,
      "grad_norm": 5.8125,
      "learning_rate": 5.708781048564276e-06,
      "loss": 1.59080944,
      "memory(GiB)": 117.38,
      "step": 38000,
      "train_speed(iter/s)": 1.638721
    },
    {
      "epoch": 0.9639776763064434,
      "eval_acc": 0.6460947631547623,
      "eval_loss": 1.5742748975753784,
      "eval_runtime": 58.1661,
      "eval_samples_per_second": 109.514,
      "eval_steps_per_second": 27.387,
      "step": 38000
    },
    {
      "acc": 0.66002145,
      "epoch": 0.9641045154743785,
      "grad_norm": 8.875,
      "learning_rate": 5.707742998726527e-06,
      "loss": 1.56301279,
      "memory(GiB)": 117.38,
      "step": 38005,
      "train_speed(iter/s)": 1.634345
    },
    {
      "acc": 0.6494401,
      "epoch": 0.9642313546423136,
      "grad_norm": 6.4375,
      "learning_rate": 5.706704917759085e-06,
      "loss": 1.55074511,
      "memory(GiB)": 117.38,
      "step": 38010,
      "train_speed(iter/s)": 1.634372
    },
    {
      "acc": 0.67064743,
      "epoch": 0.9643581938102486,
      "grad_norm": 6.21875,
      "learning_rate": 5.705666805707603e-06,
      "loss": 1.55384264,
      "memory(GiB)": 117.38,
      "step": 38015,
      "train_speed(iter/s)": 1.634398
    },
    {
      "acc": 0.6467577,
      "epoch": 0.9644850329781837,
      "grad_norm": 4.8125,
      "learning_rate": 5.704628662617744e-06,
      "loss": 1.57891865,
      "memory(GiB)": 117.38,
      "step": 38020,
      "train_speed(iter/s)": 1.634424
    },
    {
      "acc": 0.66551604,
      "epoch": 0.9646118721461188,
      "grad_norm": 5.84375,
      "learning_rate": 5.703590488535171e-06,
      "loss": 1.59015503,
      "memory(GiB)": 117.38,
      "step": 38025,
      "train_speed(iter/s)": 1.634449
    },
    {
      "acc": 0.66058064,
      "epoch": 0.9647387113140538,
      "grad_norm": 6.03125,
      "learning_rate": 5.702552283505548e-06,
      "loss": 1.60008774,
      "memory(GiB)": 117.38,
      "step": 38030,
      "train_speed(iter/s)": 1.634475
    },
    {
      "acc": 0.66553555,
      "epoch": 0.9648655504819889,
      "grad_norm": 5.46875,
      "learning_rate": 5.7015140475745376e-06,
      "loss": 1.55555792,
      "memory(GiB)": 117.38,
      "step": 38035,
      "train_speed(iter/s)": 1.634501
    },
    {
      "acc": 0.66134729,
      "epoch": 0.9649923896499238,
      "grad_norm": 5.84375,
      "learning_rate": 5.700475780787809e-06,
      "loss": 1.55473156,
      "memory(GiB)": 117.38,
      "step": 38040,
      "train_speed(iter/s)": 1.634528
    },
    {
      "acc": 0.65012608,
      "epoch": 0.9651192288178589,
      "grad_norm": 5.6875,
      "learning_rate": 5.699437483191027e-06,
      "loss": 1.55532284,
      "memory(GiB)": 117.38,
      "step": 38045,
      "train_speed(iter/s)": 1.634553
    },
    {
      "acc": 0.6550499,
      "epoch": 0.965246067985794,
      "grad_norm": 5.34375,
      "learning_rate": 5.6983991548298615e-06,
      "loss": 1.60887337,
      "memory(GiB)": 117.38,
      "step": 38050,
      "train_speed(iter/s)": 1.634578
    },
    {
      "acc": 0.66220093,
      "epoch": 0.965372907153729,
      "grad_norm": 7.0,
      "learning_rate": 5.697360795749983e-06,
      "loss": 1.5825057,
      "memory(GiB)": 117.38,
      "step": 38055,
      "train_speed(iter/s)": 1.634603
    },
    {
      "acc": 0.65022602,
      "epoch": 0.9654997463216641,
      "grad_norm": 5.25,
      "learning_rate": 5.696322405997064e-06,
      "loss": 1.64629097,
      "memory(GiB)": 117.38,
      "step": 38060,
      "train_speed(iter/s)": 1.634629
    },
    {
      "acc": 0.6586452,
      "epoch": 0.9656265854895992,
      "grad_norm": 8.0625,
      "learning_rate": 5.695283985616775e-06,
      "loss": 1.64529858,
      "memory(GiB)": 117.38,
      "step": 38065,
      "train_speed(iter/s)": 1.634655
    },
    {
      "acc": 0.64954076,
      "epoch": 0.9657534246575342,
      "grad_norm": 4.9375,
      "learning_rate": 5.694245534654795e-06,
      "loss": 1.64456291,
      "memory(GiB)": 117.38,
      "step": 38070,
      "train_speed(iter/s)": 1.634681
    },
    {
      "acc": 0.65863709,
      "epoch": 0.9658802638254693,
      "grad_norm": 6.46875,
      "learning_rate": 5.693207053156794e-06,
      "loss": 1.46597719,
      "memory(GiB)": 117.38,
      "step": 38075,
      "train_speed(iter/s)": 1.634707
    },
    {
      "acc": 0.65898852,
      "epoch": 0.9660071029934043,
      "grad_norm": 4.5625,
      "learning_rate": 5.692168541168455e-06,
      "loss": 1.59460192,
      "memory(GiB)": 117.38,
      "step": 38080,
      "train_speed(iter/s)": 1.63473
    },
    {
      "acc": 0.66738076,
      "epoch": 0.9661339421613394,
      "grad_norm": 5.125,
      "learning_rate": 5.691129998735449e-06,
      "loss": 1.50150843,
      "memory(GiB)": 117.38,
      "step": 38085,
      "train_speed(iter/s)": 1.634756
    },
    {
      "acc": 0.65731931,
      "epoch": 0.9662607813292745,
      "grad_norm": 5.9375,
      "learning_rate": 5.690091425903464e-06,
      "loss": 1.66143837,
      "memory(GiB)": 117.38,
      "step": 38090,
      "train_speed(iter/s)": 1.634779
    },
    {
      "acc": 0.65085773,
      "epoch": 0.9663876204972095,
      "grad_norm": 6.375,
      "learning_rate": 5.689052822718175e-06,
      "loss": 1.63089962,
      "memory(GiB)": 117.38,
      "step": 38095,
      "train_speed(iter/s)": 1.634806
    },
    {
      "acc": 0.66536517,
      "epoch": 0.9665144596651446,
      "grad_norm": 6.40625,
      "learning_rate": 5.688014189225266e-06,
      "loss": 1.52673798,
      "memory(GiB)": 117.38,
      "step": 38100,
      "train_speed(iter/s)": 1.634832
    },
    {
      "acc": 0.65155015,
      "epoch": 0.9666412988330797,
      "grad_norm": 6.0,
      "learning_rate": 5.686975525470423e-06,
      "loss": 1.63932495,
      "memory(GiB)": 117.38,
      "step": 38105,
      "train_speed(iter/s)": 1.634857
    },
    {
      "acc": 0.66317453,
      "epoch": 0.9667681380010147,
      "grad_norm": 6.5625,
      "learning_rate": 5.685936831499328e-06,
      "loss": 1.63441505,
      "memory(GiB)": 117.38,
      "step": 38110,
      "train_speed(iter/s)": 1.634881
    },
    {
      "acc": 0.65226994,
      "epoch": 0.9668949771689498,
      "grad_norm": 5.71875,
      "learning_rate": 5.684898107357669e-06,
      "loss": 1.62173424,
      "memory(GiB)": 117.38,
      "step": 38115,
      "train_speed(iter/s)": 1.634904
    },
    {
      "acc": 0.64400449,
      "epoch": 0.9670218163368848,
      "grad_norm": 6.5,
      "learning_rate": 5.683859353091133e-06,
      "loss": 1.68070927,
      "memory(GiB)": 117.38,
      "step": 38120,
      "train_speed(iter/s)": 1.634931
    },
    {
      "acc": 0.64637203,
      "epoch": 0.9671486555048199,
      "grad_norm": 5.53125,
      "learning_rate": 5.6828205687454094e-06,
      "loss": 1.62900791,
      "memory(GiB)": 117.38,
      "step": 38125,
      "train_speed(iter/s)": 1.634957
    },
    {
      "acc": 0.64605885,
      "epoch": 0.967275494672755,
      "grad_norm": 5.3125,
      "learning_rate": 5.68178175436619e-06,
      "loss": 1.67542953,
      "memory(GiB)": 117.38,
      "step": 38130,
      "train_speed(iter/s)": 1.634983
    },
    {
      "acc": 0.65680084,
      "epoch": 0.96740233384069,
      "grad_norm": 5.5,
      "learning_rate": 5.680742909999163e-06,
      "loss": 1.62871933,
      "memory(GiB)": 117.38,
      "step": 38135,
      "train_speed(iter/s)": 1.635009
    },
    {
      "acc": 0.63650732,
      "epoch": 0.9675291730086251,
      "grad_norm": 6.6875,
      "learning_rate": 5.679704035690026e-06,
      "loss": 1.73096752,
      "memory(GiB)": 117.38,
      "step": 38140,
      "train_speed(iter/s)": 1.635035
    },
    {
      "acc": 0.65398383,
      "epoch": 0.9676560121765602,
      "grad_norm": 5.09375,
      "learning_rate": 5.6786651314844675e-06,
      "loss": 1.53638954,
      "memory(GiB)": 117.38,
      "step": 38145,
      "train_speed(iter/s)": 1.635061
    },
    {
      "acc": 0.64991064,
      "epoch": 0.9677828513444952,
      "grad_norm": 5.90625,
      "learning_rate": 5.67762619742819e-06,
      "loss": 1.62337036,
      "memory(GiB)": 117.38,
      "step": 38150,
      "train_speed(iter/s)": 1.635087
    },
    {
      "acc": 0.64448156,
      "epoch": 0.9679096905124303,
      "grad_norm": 6.15625,
      "learning_rate": 5.676587233566885e-06,
      "loss": 1.60961685,
      "memory(GiB)": 117.38,
      "step": 38155,
      "train_speed(iter/s)": 1.635112
    },
    {
      "acc": 0.65426359,
      "epoch": 0.9680365296803652,
      "grad_norm": 5.15625,
      "learning_rate": 5.675548239946254e-06,
      "loss": 1.59950018,
      "memory(GiB)": 117.38,
      "step": 38160,
      "train_speed(iter/s)": 1.63514
    },
    {
      "acc": 0.6396699,
      "epoch": 0.9681633688483003,
      "grad_norm": 6.25,
      "learning_rate": 5.674509216611993e-06,
      "loss": 1.68408718,
      "memory(GiB)": 117.38,
      "step": 38165,
      "train_speed(iter/s)": 1.635165
    },
    {
      "acc": 0.65698185,
      "epoch": 0.9682902080162354,
      "grad_norm": 5.625,
      "learning_rate": 5.673470163609806e-06,
      "loss": 1.57145653,
      "memory(GiB)": 117.38,
      "step": 38170,
      "train_speed(iter/s)": 1.635191
    },
    {
      "acc": 0.67125373,
      "epoch": 0.9684170471841704,
      "grad_norm": 5.125,
      "learning_rate": 5.672431080985395e-06,
      "loss": 1.51870546,
      "memory(GiB)": 117.38,
      "step": 38175,
      "train_speed(iter/s)": 1.635216
    },
    {
      "acc": 0.65947361,
      "epoch": 0.9685438863521055,
      "grad_norm": 6.5,
      "learning_rate": 5.671391968784464e-06,
      "loss": 1.58719416,
      "memory(GiB)": 117.38,
      "step": 38180,
      "train_speed(iter/s)": 1.635242
    },
    {
      "acc": 0.66260357,
      "epoch": 0.9686707255200406,
      "grad_norm": 6.09375,
      "learning_rate": 5.670352827052715e-06,
      "loss": 1.59185772,
      "memory(GiB)": 117.38,
      "step": 38185,
      "train_speed(iter/s)": 1.635268
    },
    {
      "acc": 0.66881456,
      "epoch": 0.9687975646879756,
      "grad_norm": 5.625,
      "learning_rate": 5.6693136558358565e-06,
      "loss": 1.62469482,
      "memory(GiB)": 117.38,
      "step": 38190,
      "train_speed(iter/s)": 1.635294
    },
    {
      "acc": 0.64528675,
      "epoch": 0.9689244038559107,
      "grad_norm": 5.53125,
      "learning_rate": 5.668274455179595e-06,
      "loss": 1.64711704,
      "memory(GiB)": 117.38,
      "step": 38195,
      "train_speed(iter/s)": 1.635317
    },
    {
      "acc": 0.63186703,
      "epoch": 0.9690512430238457,
      "grad_norm": 6.28125,
      "learning_rate": 5.667235225129639e-06,
      "loss": 1.67505989,
      "memory(GiB)": 117.38,
      "step": 38200,
      "train_speed(iter/s)": 1.635342
    },
    {
      "acc": 0.63843536,
      "epoch": 0.9691780821917808,
      "grad_norm": 6.96875,
      "learning_rate": 5.6661959657317e-06,
      "loss": 1.61682205,
      "memory(GiB)": 117.38,
      "step": 38205,
      "train_speed(iter/s)": 1.635366
    },
    {
      "acc": 0.67373304,
      "epoch": 0.9693049213597159,
      "grad_norm": 7.15625,
      "learning_rate": 5.665156677031487e-06,
      "loss": 1.55660601,
      "memory(GiB)": 117.38,
      "step": 38210,
      "train_speed(iter/s)": 1.635393
    },
    {
      "acc": 0.65169334,
      "epoch": 0.9694317605276509,
      "grad_norm": 6.84375,
      "learning_rate": 5.664117359074712e-06,
      "loss": 1.63847847,
      "memory(GiB)": 117.38,
      "step": 38215,
      "train_speed(iter/s)": 1.63542
    },
    {
      "acc": 0.64914036,
      "epoch": 0.969558599695586,
      "grad_norm": 6.1875,
      "learning_rate": 5.6630780119070935e-06,
      "loss": 1.67669601,
      "memory(GiB)": 117.38,
      "step": 38220,
      "train_speed(iter/s)": 1.635446
    },
    {
      "acc": 0.6539856,
      "epoch": 0.9696854388635211,
      "grad_norm": 5.59375,
      "learning_rate": 5.6620386355743415e-06,
      "loss": 1.63298473,
      "memory(GiB)": 117.38,
      "step": 38225,
      "train_speed(iter/s)": 1.635471
    },
    {
      "acc": 0.65535269,
      "epoch": 0.9698122780314561,
      "grad_norm": 5.9375,
      "learning_rate": 5.660999230122177e-06,
      "loss": 1.58836575,
      "memory(GiB)": 117.38,
      "step": 38230,
      "train_speed(iter/s)": 1.635496
    },
    {
      "acc": 0.65222006,
      "epoch": 0.9699391171993912,
      "grad_norm": 5.78125,
      "learning_rate": 5.659959795596313e-06,
      "loss": 1.59134541,
      "memory(GiB)": 117.38,
      "step": 38235,
      "train_speed(iter/s)": 1.635517
    },
    {
      "acc": 0.64766917,
      "epoch": 0.9700659563673262,
      "grad_norm": 6.40625,
      "learning_rate": 5.65892033204247e-06,
      "loss": 1.64647732,
      "memory(GiB)": 117.38,
      "step": 38240,
      "train_speed(iter/s)": 1.635542
    },
    {
      "acc": 0.67077255,
      "epoch": 0.9701927955352613,
      "grad_norm": 5.125,
      "learning_rate": 5.657880839506371e-06,
      "loss": 1.54590244,
      "memory(GiB)": 117.38,
      "step": 38245,
      "train_speed(iter/s)": 1.635567
    },
    {
      "acc": 0.65406132,
      "epoch": 0.9703196347031964,
      "grad_norm": 7.625,
      "learning_rate": 5.656841318033735e-06,
      "loss": 1.58277416,
      "memory(GiB)": 117.38,
      "step": 38250,
      "train_speed(iter/s)": 1.635593
    },
    {
      "acc": 0.65494175,
      "epoch": 0.9704464738711314,
      "grad_norm": 7.8125,
      "learning_rate": 5.6558017676702846e-06,
      "loss": 1.58465919,
      "memory(GiB)": 117.38,
      "step": 38255,
      "train_speed(iter/s)": 1.635617
    },
    {
      "acc": 0.65172529,
      "epoch": 0.9705733130390665,
      "grad_norm": 5.59375,
      "learning_rate": 5.654762188461744e-06,
      "loss": 1.64318771,
      "memory(GiB)": 117.38,
      "step": 38260,
      "train_speed(iter/s)": 1.635643
    },
    {
      "acc": 0.6476625,
      "epoch": 0.9707001522070016,
      "grad_norm": 5.90625,
      "learning_rate": 5.653722580453841e-06,
      "loss": 1.65056038,
      "memory(GiB)": 117.38,
      "step": 38265,
      "train_speed(iter/s)": 1.635669
    },
    {
      "acc": 0.64132814,
      "epoch": 0.9708269913749366,
      "grad_norm": 5.59375,
      "learning_rate": 5.652682943692299e-06,
      "loss": 1.6684557,
      "memory(GiB)": 117.38,
      "step": 38270,
      "train_speed(iter/s)": 1.635695
    },
    {
      "acc": 0.66135368,
      "epoch": 0.9709538305428717,
      "grad_norm": 7.3125,
      "learning_rate": 5.651643278222847e-06,
      "loss": 1.60470467,
      "memory(GiB)": 117.38,
      "step": 38275,
      "train_speed(iter/s)": 1.635723
    },
    {
      "acc": 0.67604713,
      "epoch": 0.9710806697108066,
      "grad_norm": 6.21875,
      "learning_rate": 5.6506035840912145e-06,
      "loss": 1.55927334,
      "memory(GiB)": 117.38,
      "step": 38280,
      "train_speed(iter/s)": 1.635748
    },
    {
      "acc": 0.6676239,
      "epoch": 0.9712075088787417,
      "grad_norm": 6.0625,
      "learning_rate": 5.649563861343131e-06,
      "loss": 1.59370117,
      "memory(GiB)": 117.38,
      "step": 38285,
      "train_speed(iter/s)": 1.635773
    },
    {
      "acc": 0.63996372,
      "epoch": 0.9713343480466768,
      "grad_norm": 6.96875,
      "learning_rate": 5.648524110024331e-06,
      "loss": 1.62207546,
      "memory(GiB)": 117.38,
      "step": 38290,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.66518064,
      "epoch": 0.9714611872146118,
      "grad_norm": 5.71875,
      "learning_rate": 5.647484330180542e-06,
      "loss": 1.60813046,
      "memory(GiB)": 117.38,
      "step": 38295,
      "train_speed(iter/s)": 1.635822
    },
    {
      "acc": 0.6587152,
      "epoch": 0.9715880263825469,
      "grad_norm": 7.53125,
      "learning_rate": 5.646444521857504e-06,
      "loss": 1.64002476,
      "memory(GiB)": 117.38,
      "step": 38300,
      "train_speed(iter/s)": 1.635847
    },
    {
      "acc": 0.67270718,
      "epoch": 0.971714865550482,
      "grad_norm": 5.8125,
      "learning_rate": 5.645404685100948e-06,
      "loss": 1.53261986,
      "memory(GiB)": 117.38,
      "step": 38305,
      "train_speed(iter/s)": 1.635873
    },
    {
      "acc": 0.65908842,
      "epoch": 0.971841704718417,
      "grad_norm": 5.65625,
      "learning_rate": 5.644364819956613e-06,
      "loss": 1.62424812,
      "memory(GiB)": 117.38,
      "step": 38310,
      "train_speed(iter/s)": 1.635898
    },
    {
      "acc": 0.65067244,
      "epoch": 0.9719685438863521,
      "grad_norm": 6.125,
      "learning_rate": 5.643324926470236e-06,
      "loss": 1.6156662,
      "memory(GiB)": 117.38,
      "step": 38315,
      "train_speed(iter/s)": 1.635923
    },
    {
      "acc": 0.6559042,
      "epoch": 0.9720953830542871,
      "grad_norm": 7.21875,
      "learning_rate": 5.642285004687557e-06,
      "loss": 1.60668678,
      "memory(GiB)": 117.38,
      "step": 38320,
      "train_speed(iter/s)": 1.635949
    },
    {
      "acc": 0.66721106,
      "epoch": 0.9722222222222222,
      "grad_norm": 6.0,
      "learning_rate": 5.6412450546543165e-06,
      "loss": 1.58803902,
      "memory(GiB)": 117.38,
      "step": 38325,
      "train_speed(iter/s)": 1.635973
    },
    {
      "acc": 0.64878178,
      "epoch": 0.9723490613901573,
      "grad_norm": 5.90625,
      "learning_rate": 5.640205076416254e-06,
      "loss": 1.71433201,
      "memory(GiB)": 117.38,
      "step": 38330,
      "train_speed(iter/s)": 1.635998
    },
    {
      "acc": 0.65180159,
      "epoch": 0.9724759005580923,
      "grad_norm": 7.3125,
      "learning_rate": 5.639165070019116e-06,
      "loss": 1.62308025,
      "memory(GiB)": 117.38,
      "step": 38335,
      "train_speed(iter/s)": 1.636021
    },
    {
      "acc": 0.65096407,
      "epoch": 0.9726027397260274,
      "grad_norm": 6.53125,
      "learning_rate": 5.638125035508642e-06,
      "loss": 1.60337181,
      "memory(GiB)": 117.38,
      "step": 38340,
      "train_speed(iter/s)": 1.636045
    },
    {
      "acc": 0.66166754,
      "epoch": 0.9727295788939625,
      "grad_norm": 7.625,
      "learning_rate": 5.6370849729305825e-06,
      "loss": 1.60065327,
      "memory(GiB)": 117.38,
      "step": 38345,
      "train_speed(iter/s)": 1.636068
    },
    {
      "acc": 0.63902817,
      "epoch": 0.9728564180618975,
      "grad_norm": 5.4375,
      "learning_rate": 5.63604488233068e-06,
      "loss": 1.66976547,
      "memory(GiB)": 117.38,
      "step": 38350,
      "train_speed(iter/s)": 1.636092
    },
    {
      "acc": 0.65062933,
      "epoch": 0.9729832572298326,
      "grad_norm": 6.09375,
      "learning_rate": 5.635004763754683e-06,
      "loss": 1.58455353,
      "memory(GiB)": 117.38,
      "step": 38355,
      "train_speed(iter/s)": 1.636115
    },
    {
      "acc": 0.64914169,
      "epoch": 0.9731100963977676,
      "grad_norm": 6.375,
      "learning_rate": 5.633964617248345e-06,
      "loss": 1.62896042,
      "memory(GiB)": 117.38,
      "step": 38360,
      "train_speed(iter/s)": 1.636138
    },
    {
      "acc": 0.66763501,
      "epoch": 0.9732369355657027,
      "grad_norm": 5.3125,
      "learning_rate": 5.6329244428574085e-06,
      "loss": 1.5472496,
      "memory(GiB)": 117.38,
      "step": 38365,
      "train_speed(iter/s)": 1.636163
    },
    {
      "acc": 0.65448613,
      "epoch": 0.9733637747336378,
      "grad_norm": 5.5625,
      "learning_rate": 5.631884240627632e-06,
      "loss": 1.62341328,
      "memory(GiB)": 117.38,
      "step": 38370,
      "train_speed(iter/s)": 1.636187
    },
    {
      "acc": 0.6505312,
      "epoch": 0.9734906139015728,
      "grad_norm": 4.75,
      "learning_rate": 5.6308440106047634e-06,
      "loss": 1.65536308,
      "memory(GiB)": 117.38,
      "step": 38375,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.66176796,
      "epoch": 0.9736174530695079,
      "grad_norm": 5.875,
      "learning_rate": 5.62980375283456e-06,
      "loss": 1.59309978,
      "memory(GiB)": 117.38,
      "step": 38380,
      "train_speed(iter/s)": 1.636234
    },
    {
      "acc": 0.65792465,
      "epoch": 0.973744292237443,
      "grad_norm": 5.8125,
      "learning_rate": 5.628763467362775e-06,
      "loss": 1.60247307,
      "memory(GiB)": 117.38,
      "step": 38385,
      "train_speed(iter/s)": 1.636256
    },
    {
      "acc": 0.64849577,
      "epoch": 0.973871131405378,
      "grad_norm": 5.375,
      "learning_rate": 5.627723154235165e-06,
      "loss": 1.61164494,
      "memory(GiB)": 117.38,
      "step": 38390,
      "train_speed(iter/s)": 1.63628
    },
    {
      "acc": 0.6562932,
      "epoch": 0.973997970573313,
      "grad_norm": 5.9375,
      "learning_rate": 5.62668281349749e-06,
      "loss": 1.66902771,
      "memory(GiB)": 117.38,
      "step": 38395,
      "train_speed(iter/s)": 1.636304
    },
    {
      "acc": 0.64309597,
      "epoch": 0.974124809741248,
      "grad_norm": 6.03125,
      "learning_rate": 5.625642445195505e-06,
      "loss": 1.69052143,
      "memory(GiB)": 117.38,
      "step": 38400,
      "train_speed(iter/s)": 1.636327
    },
    {
      "acc": 0.6652966,
      "epoch": 0.9742516489091831,
      "grad_norm": 6.09375,
      "learning_rate": 5.6246020493749735e-06,
      "loss": 1.59762754,
      "memory(GiB)": 117.38,
      "step": 38405,
      "train_speed(iter/s)": 1.63635
    },
    {
      "acc": 0.65778608,
      "epoch": 0.9743784880771182,
      "grad_norm": 6.375,
      "learning_rate": 5.623561626081654e-06,
      "loss": 1.62377644,
      "memory(GiB)": 117.38,
      "step": 38410,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.65515709,
      "epoch": 0.9745053272450532,
      "grad_norm": 7.375,
      "learning_rate": 5.622521175361311e-06,
      "loss": 1.57583199,
      "memory(GiB)": 117.38,
      "step": 38415,
      "train_speed(iter/s)": 1.636396
    },
    {
      "acc": 0.66052942,
      "epoch": 0.9746321664129883,
      "grad_norm": 5.71875,
      "learning_rate": 5.621480697259707e-06,
      "loss": 1.68279095,
      "memory(GiB)": 117.38,
      "step": 38420,
      "train_speed(iter/s)": 1.636421
    },
    {
      "acc": 0.64012756,
      "epoch": 0.9747590055809234,
      "grad_norm": 5.5,
      "learning_rate": 5.620440191822607e-06,
      "loss": 1.6234848,
      "memory(GiB)": 117.38,
      "step": 38425,
      "train_speed(iter/s)": 1.636445
    },
    {
      "acc": 0.63879981,
      "epoch": 0.9748858447488584,
      "grad_norm": 5.15625,
      "learning_rate": 5.619399659095778e-06,
      "loss": 1.63880997,
      "memory(GiB)": 117.38,
      "step": 38430,
      "train_speed(iter/s)": 1.636469
    },
    {
      "acc": 0.65664611,
      "epoch": 0.9750126839167935,
      "grad_norm": 6.375,
      "learning_rate": 5.618359099124985e-06,
      "loss": 1.62167263,
      "memory(GiB)": 117.38,
      "step": 38435,
      "train_speed(iter/s)": 1.636494
    },
    {
      "acc": 0.6515183,
      "epoch": 0.9751395230847285,
      "grad_norm": 5.09375,
      "learning_rate": 5.617318511956001e-06,
      "loss": 1.65804596,
      "memory(GiB)": 117.38,
      "step": 38440,
      "train_speed(iter/s)": 1.636518
    },
    {
      "acc": 0.6528573,
      "epoch": 0.9752663622526636,
      "grad_norm": 6.1875,
      "learning_rate": 5.61627789763459e-06,
      "loss": 1.64230652,
      "memory(GiB)": 117.38,
      "step": 38445,
      "train_speed(iter/s)": 1.636541
    },
    {
      "acc": 0.65744667,
      "epoch": 0.9753932014205987,
      "grad_norm": 5.53125,
      "learning_rate": 5.6152372562065275e-06,
      "loss": 1.56383057,
      "memory(GiB)": 117.38,
      "step": 38450,
      "train_speed(iter/s)": 1.636566
    },
    {
      "acc": 0.64883323,
      "epoch": 0.9755200405885337,
      "grad_norm": 5.6875,
      "learning_rate": 5.614196587717581e-06,
      "loss": 1.62848892,
      "memory(GiB)": 117.38,
      "step": 38455,
      "train_speed(iter/s)": 1.636591
    },
    {
      "acc": 0.66938925,
      "epoch": 0.9756468797564688,
      "grad_norm": 4.75,
      "learning_rate": 5.613155892213529e-06,
      "loss": 1.55475492,
      "memory(GiB)": 117.38,
      "step": 38460,
      "train_speed(iter/s)": 1.636615
    },
    {
      "acc": 0.67106466,
      "epoch": 0.9757737189244039,
      "grad_norm": 5.1875,
      "learning_rate": 5.612115169740142e-06,
      "loss": 1.55789709,
      "memory(GiB)": 117.38,
      "step": 38465,
      "train_speed(iter/s)": 1.636639
    },
    {
      "acc": 0.66363754,
      "epoch": 0.9759005580923389,
      "grad_norm": 5.96875,
      "learning_rate": 5.611074420343197e-06,
      "loss": 1.58945103,
      "memory(GiB)": 117.38,
      "step": 38470,
      "train_speed(iter/s)": 1.636664
    },
    {
      "acc": 0.66767759,
      "epoch": 0.976027397260274,
      "grad_norm": 5.40625,
      "learning_rate": 5.610033644068471e-06,
      "loss": 1.5798296,
      "memory(GiB)": 117.38,
      "step": 38475,
      "train_speed(iter/s)": 1.63669
    },
    {
      "acc": 0.66720638,
      "epoch": 0.976154236428209,
      "grad_norm": 4.78125,
      "learning_rate": 5.608992840961742e-06,
      "loss": 1.55297451,
      "memory(GiB)": 117.38,
      "step": 38480,
      "train_speed(iter/s)": 1.636714
    },
    {
      "acc": 0.65064545,
      "epoch": 0.9762810755961441,
      "grad_norm": 5.78125,
      "learning_rate": 5.6079520110687876e-06,
      "loss": 1.59366951,
      "memory(GiB)": 117.38,
      "step": 38485,
      "train_speed(iter/s)": 1.63674
    },
    {
      "acc": 0.66323409,
      "epoch": 0.9764079147640792,
      "grad_norm": 5.375,
      "learning_rate": 5.606911154435392e-06,
      "loss": 1.56340561,
      "memory(GiB)": 117.38,
      "step": 38490,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.65641155,
      "epoch": 0.9765347539320142,
      "grad_norm": 7.4375,
      "learning_rate": 5.605870271107332e-06,
      "loss": 1.61000671,
      "memory(GiB)": 117.38,
      "step": 38495,
      "train_speed(iter/s)": 1.636789
    },
    {
      "acc": 0.65363274,
      "epoch": 0.9766615930999493,
      "grad_norm": 5.625,
      "learning_rate": 5.6048293611303925e-06,
      "loss": 1.61809254,
      "memory(GiB)": 117.38,
      "step": 38500,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.6769455,
      "epoch": 0.9767884322678844,
      "grad_norm": 4.75,
      "learning_rate": 5.603788424550357e-06,
      "loss": 1.57775764,
      "memory(GiB)": 117.38,
      "step": 38505,
      "train_speed(iter/s)": 1.636837
    },
    {
      "acc": 0.65813751,
      "epoch": 0.9769152714358194,
      "grad_norm": 4.8125,
      "learning_rate": 5.602747461413014e-06,
      "loss": 1.56889706,
      "memory(GiB)": 117.38,
      "step": 38510,
      "train_speed(iter/s)": 1.63686
    },
    {
      "acc": 0.66425519,
      "epoch": 0.9770421106037545,
      "grad_norm": 7.8125,
      "learning_rate": 5.6017064717641435e-06,
      "loss": 1.53046474,
      "memory(GiB)": 117.38,
      "step": 38515,
      "train_speed(iter/s)": 1.636884
    },
    {
      "acc": 0.65539885,
      "epoch": 0.9771689497716894,
      "grad_norm": 5.15625,
      "learning_rate": 5.600665455649538e-06,
      "loss": 1.609202,
      "memory(GiB)": 117.38,
      "step": 38520,
      "train_speed(iter/s)": 1.636909
    },
    {
      "acc": 0.65283327,
      "epoch": 0.9772957889396245,
      "grad_norm": 5.84375,
      "learning_rate": 5.599624413114981e-06,
      "loss": 1.60686741,
      "memory(GiB)": 117.38,
      "step": 38525,
      "train_speed(iter/s)": 1.636932
    },
    {
      "acc": 0.657969,
      "epoch": 0.9774226281075596,
      "grad_norm": 5.28125,
      "learning_rate": 5.5985833442062676e-06,
      "loss": 1.59414825,
      "memory(GiB)": 117.38,
      "step": 38530,
      "train_speed(iter/s)": 1.636957
    },
    {
      "acc": 0.66860781,
      "epoch": 0.9775494672754946,
      "grad_norm": 5.34375,
      "learning_rate": 5.597542248969185e-06,
      "loss": 1.51761341,
      "memory(GiB)": 117.38,
      "step": 38535,
      "train_speed(iter/s)": 1.636982
    },
    {
      "acc": 0.64384117,
      "epoch": 0.9776763064434297,
      "grad_norm": 7.28125,
      "learning_rate": 5.596501127449527e-06,
      "loss": 1.62396908,
      "memory(GiB)": 117.38,
      "step": 38540,
      "train_speed(iter/s)": 1.637006
    },
    {
      "acc": 0.65910268,
      "epoch": 0.9778031456113648,
      "grad_norm": 5.46875,
      "learning_rate": 5.595459979693086e-06,
      "loss": 1.57738466,
      "memory(GiB)": 117.38,
      "step": 38545,
      "train_speed(iter/s)": 1.63703
    },
    {
      "acc": 0.6607317,
      "epoch": 0.9779299847792998,
      "grad_norm": 5.875,
      "learning_rate": 5.594418805745657e-06,
      "loss": 1.58972359,
      "memory(GiB)": 117.38,
      "step": 38550,
      "train_speed(iter/s)": 1.637055
    },
    {
      "acc": 0.6520052,
      "epoch": 0.9780568239472349,
      "grad_norm": 6.1875,
      "learning_rate": 5.593377605653035e-06,
      "loss": 1.58321934,
      "memory(GiB)": 117.38,
      "step": 38555,
      "train_speed(iter/s)": 1.63708
    },
    {
      "acc": 0.6600708,
      "epoch": 0.9781836631151699,
      "grad_norm": 5.625,
      "learning_rate": 5.592336379461018e-06,
      "loss": 1.57247629,
      "memory(GiB)": 117.38,
      "step": 38560,
      "train_speed(iter/s)": 1.637106
    },
    {
      "acc": 0.64563799,
      "epoch": 0.978310502283105,
      "grad_norm": 7.46875,
      "learning_rate": 5.5912951272154004e-06,
      "loss": 1.55649872,
      "memory(GiB)": 117.38,
      "step": 38565,
      "train_speed(iter/s)": 1.63713
    },
    {
      "acc": 0.65249586,
      "epoch": 0.9784373414510401,
      "grad_norm": 5.0,
      "learning_rate": 5.590253848961984e-06,
      "loss": 1.65953255,
      "memory(GiB)": 117.38,
      "step": 38570,
      "train_speed(iter/s)": 1.637154
    },
    {
      "acc": 0.64530449,
      "epoch": 0.9785641806189751,
      "grad_norm": 5.03125,
      "learning_rate": 5.589212544746566e-06,
      "loss": 1.63760529,
      "memory(GiB)": 117.38,
      "step": 38575,
      "train_speed(iter/s)": 1.637178
    },
    {
      "acc": 0.6618928,
      "epoch": 0.9786910197869102,
      "grad_norm": 7.09375,
      "learning_rate": 5.588171214614953e-06,
      "loss": 1.63993301,
      "memory(GiB)": 117.38,
      "step": 38580,
      "train_speed(iter/s)": 1.637202
    },
    {
      "acc": 0.64911199,
      "epoch": 0.9788178589548453,
      "grad_norm": 5.28125,
      "learning_rate": 5.587129858612941e-06,
      "loss": 1.60417976,
      "memory(GiB)": 117.38,
      "step": 38585,
      "train_speed(iter/s)": 1.637226
    },
    {
      "acc": 0.6474401,
      "epoch": 0.9789446981227803,
      "grad_norm": 4.65625,
      "learning_rate": 5.586088476786339e-06,
      "loss": 1.62720528,
      "memory(GiB)": 117.38,
      "step": 38590,
      "train_speed(iter/s)": 1.637249
    },
    {
      "acc": 0.64482594,
      "epoch": 0.9790715372907154,
      "grad_norm": 5.4375,
      "learning_rate": 5.585047069180947e-06,
      "loss": 1.67413311,
      "memory(GiB)": 117.38,
      "step": 38595,
      "train_speed(iter/s)": 1.637274
    },
    {
      "acc": 0.65761099,
      "epoch": 0.9791983764586504,
      "grad_norm": 6.375,
      "learning_rate": 5.5840056358425755e-06,
      "loss": 1.55436783,
      "memory(GiB)": 117.38,
      "step": 38600,
      "train_speed(iter/s)": 1.637298
    },
    {
      "acc": 0.6373436,
      "epoch": 0.9793252156265855,
      "grad_norm": 5.71875,
      "learning_rate": 5.582964176817025e-06,
      "loss": 1.6881958,
      "memory(GiB)": 117.38,
      "step": 38605,
      "train_speed(iter/s)": 1.637321
    },
    {
      "acc": 0.66941342,
      "epoch": 0.9794520547945206,
      "grad_norm": 7.0,
      "learning_rate": 5.58192269215011e-06,
      "loss": 1.53719635,
      "memory(GiB)": 117.38,
      "step": 38610,
      "train_speed(iter/s)": 1.637344
    },
    {
      "acc": 0.66720614,
      "epoch": 0.9795788939624556,
      "grad_norm": 5.84375,
      "learning_rate": 5.580881181887636e-06,
      "loss": 1.60927563,
      "memory(GiB)": 117.38,
      "step": 38615,
      "train_speed(iter/s)": 1.637369
    },
    {
      "acc": 0.65170488,
      "epoch": 0.9797057331303907,
      "grad_norm": 4.84375,
      "learning_rate": 5.579839646075414e-06,
      "loss": 1.60060921,
      "memory(GiB)": 117.38,
      "step": 38620,
      "train_speed(iter/s)": 1.637392
    },
    {
      "acc": 0.64748268,
      "epoch": 0.9798325722983258,
      "grad_norm": 4.5,
      "learning_rate": 5.578798084759257e-06,
      "loss": 1.65848389,
      "memory(GiB)": 117.38,
      "step": 38625,
      "train_speed(iter/s)": 1.637413
    },
    {
      "acc": 0.66331711,
      "epoch": 0.9799594114662608,
      "grad_norm": 6.53125,
      "learning_rate": 5.577756497984975e-06,
      "loss": 1.54005833,
      "memory(GiB)": 117.38,
      "step": 38630,
      "train_speed(iter/s)": 1.637436
    },
    {
      "acc": 0.65858107,
      "epoch": 0.9800862506341959,
      "grad_norm": 5.28125,
      "learning_rate": 5.576714885798382e-06,
      "loss": 1.53903885,
      "memory(GiB)": 117.38,
      "step": 38635,
      "train_speed(iter/s)": 1.63746
    },
    {
      "acc": 0.66127539,
      "epoch": 0.9802130898021308,
      "grad_norm": 6.625,
      "learning_rate": 5.575673248245295e-06,
      "loss": 1.57987604,
      "memory(GiB)": 117.38,
      "step": 38640,
      "train_speed(iter/s)": 1.637484
    },
    {
      "acc": 0.64577646,
      "epoch": 0.9803399289700659,
      "grad_norm": 8.375,
      "learning_rate": 5.574631585371527e-06,
      "loss": 1.71133423,
      "memory(GiB)": 117.38,
      "step": 38645,
      "train_speed(iter/s)": 1.637508
    },
    {
      "acc": 0.6480269,
      "epoch": 0.980466768138001,
      "grad_norm": 6.09375,
      "learning_rate": 5.573589897222897e-06,
      "loss": 1.59905272,
      "memory(GiB)": 117.38,
      "step": 38650,
      "train_speed(iter/s)": 1.637531
    },
    {
      "acc": 0.66004782,
      "epoch": 0.980593607305936,
      "grad_norm": 5.0,
      "learning_rate": 5.572548183845222e-06,
      "loss": 1.54437485,
      "memory(GiB)": 117.38,
      "step": 38655,
      "train_speed(iter/s)": 1.637556
    },
    {
      "acc": 0.65010457,
      "epoch": 0.9807204464738711,
      "grad_norm": 4.90625,
      "learning_rate": 5.571506445284322e-06,
      "loss": 1.61843472,
      "memory(GiB)": 117.38,
      "step": 38660,
      "train_speed(iter/s)": 1.637578
    },
    {
      "acc": 0.6674324,
      "epoch": 0.9808472856418062,
      "grad_norm": 4.5,
      "learning_rate": 5.570464681586017e-06,
      "loss": 1.57486477,
      "memory(GiB)": 117.38,
      "step": 38665,
      "train_speed(iter/s)": 1.6376
    },
    {
      "acc": 0.64909735,
      "epoch": 0.9809741248097412,
      "grad_norm": 5.34375,
      "learning_rate": 5.569422892796129e-06,
      "loss": 1.59526539,
      "memory(GiB)": 117.38,
      "step": 38670,
      "train_speed(iter/s)": 1.637623
    },
    {
      "acc": 0.63954821,
      "epoch": 0.9811009639776763,
      "grad_norm": 5.71875,
      "learning_rate": 5.568381078960479e-06,
      "loss": 1.75403938,
      "memory(GiB)": 117.38,
      "step": 38675,
      "train_speed(iter/s)": 1.637647
    },
    {
      "acc": 0.65852995,
      "epoch": 0.9812278031456113,
      "grad_norm": 6.9375,
      "learning_rate": 5.567339240124892e-06,
      "loss": 1.53508778,
      "memory(GiB)": 117.38,
      "step": 38680,
      "train_speed(iter/s)": 1.637671
    },
    {
      "acc": 0.65534935,
      "epoch": 0.9813546423135464,
      "grad_norm": 5.40625,
      "learning_rate": 5.5662973763351915e-06,
      "loss": 1.64453754,
      "memory(GiB)": 117.38,
      "step": 38685,
      "train_speed(iter/s)": 1.637693
    },
    {
      "acc": 0.6488111,
      "epoch": 0.9814814814814815,
      "grad_norm": 6.625,
      "learning_rate": 5.565255487637204e-06,
      "loss": 1.62392483,
      "memory(GiB)": 117.38,
      "step": 38690,
      "train_speed(iter/s)": 1.637718
    },
    {
      "acc": 0.65491405,
      "epoch": 0.9816083206494165,
      "grad_norm": 6.03125,
      "learning_rate": 5.564213574076757e-06,
      "loss": 1.58939095,
      "memory(GiB)": 117.38,
      "step": 38695,
      "train_speed(iter/s)": 1.637741
    },
    {
      "acc": 0.64531393,
      "epoch": 0.9817351598173516,
      "grad_norm": 6.53125,
      "learning_rate": 5.563171635699678e-06,
      "loss": 1.65395947,
      "memory(GiB)": 117.38,
      "step": 38700,
      "train_speed(iter/s)": 1.637765
    },
    {
      "acc": 0.67010503,
      "epoch": 0.9818619989852867,
      "grad_norm": 6.09375,
      "learning_rate": 5.562129672551796e-06,
      "loss": 1.54790859,
      "memory(GiB)": 117.38,
      "step": 38705,
      "train_speed(iter/s)": 1.637789
    },
    {
      "acc": 0.64942441,
      "epoch": 0.9819888381532217,
      "grad_norm": 5.71875,
      "learning_rate": 5.561087684678941e-06,
      "loss": 1.61868114,
      "memory(GiB)": 117.38,
      "step": 38710,
      "train_speed(iter/s)": 1.637812
    },
    {
      "acc": 0.64999676,
      "epoch": 0.9821156773211568,
      "grad_norm": 5.3125,
      "learning_rate": 5.560045672126945e-06,
      "loss": 1.62361717,
      "memory(GiB)": 117.38,
      "step": 38715,
      "train_speed(iter/s)": 1.637835
    },
    {
      "acc": 0.65503788,
      "epoch": 0.9822425164890918,
      "grad_norm": 5.875,
      "learning_rate": 5.55900363494164e-06,
      "loss": 1.61794205,
      "memory(GiB)": 117.38,
      "step": 38720,
      "train_speed(iter/s)": 1.637858
    },
    {
      "acc": 0.66599765,
      "epoch": 0.9823693556570269,
      "grad_norm": 5.96875,
      "learning_rate": 5.557961573168857e-06,
      "loss": 1.63279839,
      "memory(GiB)": 117.38,
      "step": 38725,
      "train_speed(iter/s)": 1.637882
    },
    {
      "acc": 0.66272669,
      "epoch": 0.982496194824962,
      "grad_norm": 6.125,
      "learning_rate": 5.5569194868544376e-06,
      "loss": 1.58562908,
      "memory(GiB)": 117.38,
      "step": 38730,
      "train_speed(iter/s)": 1.637906
    },
    {
      "acc": 0.65216403,
      "epoch": 0.982623033992897,
      "grad_norm": 5.96875,
      "learning_rate": 5.555877376044209e-06,
      "loss": 1.57129736,
      "memory(GiB)": 117.38,
      "step": 38735,
      "train_speed(iter/s)": 1.637929
    },
    {
      "acc": 0.67389421,
      "epoch": 0.9827498731608321,
      "grad_norm": 5.6875,
      "learning_rate": 5.554835240784013e-06,
      "loss": 1.59081326,
      "memory(GiB)": 117.38,
      "step": 38740,
      "train_speed(iter/s)": 1.637952
    },
    {
      "acc": 0.65340695,
      "epoch": 0.9828767123287672,
      "grad_norm": 5.8125,
      "learning_rate": 5.553793081119685e-06,
      "loss": 1.62839966,
      "memory(GiB)": 117.38,
      "step": 38745,
      "train_speed(iter/s)": 1.637973
    },
    {
      "acc": 0.65438223,
      "epoch": 0.9830035514967022,
      "grad_norm": 4.78125,
      "learning_rate": 5.552750897097065e-06,
      "loss": 1.61699829,
      "memory(GiB)": 117.38,
      "step": 38750,
      "train_speed(iter/s)": 1.637996
    },
    {
      "acc": 0.64969902,
      "epoch": 0.9831303906646373,
      "grad_norm": 6.75,
      "learning_rate": 5.551708688761993e-06,
      "loss": 1.61930351,
      "memory(GiB)": 117.38,
      "step": 38755,
      "train_speed(iter/s)": 1.63802
    },
    {
      "acc": 0.65622864,
      "epoch": 0.9832572298325722,
      "grad_norm": 5.6875,
      "learning_rate": 5.550666456160311e-06,
      "loss": 1.59590492,
      "memory(GiB)": 117.38,
      "step": 38760,
      "train_speed(iter/s)": 1.638043
    },
    {
      "acc": 0.65961337,
      "epoch": 0.9833840690005073,
      "grad_norm": 5.03125,
      "learning_rate": 5.549624199337857e-06,
      "loss": 1.60344238,
      "memory(GiB)": 117.38,
      "step": 38765,
      "train_speed(iter/s)": 1.638068
    },
    {
      "acc": 0.65280104,
      "epoch": 0.9835109081684424,
      "grad_norm": 6.9375,
      "learning_rate": 5.548581918340479e-06,
      "loss": 1.63526535,
      "memory(GiB)": 117.38,
      "step": 38770,
      "train_speed(iter/s)": 1.638092
    },
    {
      "acc": 0.66507111,
      "epoch": 0.9836377473363774,
      "grad_norm": 5.9375,
      "learning_rate": 5.547539613214019e-06,
      "loss": 1.56163425,
      "memory(GiB)": 117.38,
      "step": 38775,
      "train_speed(iter/s)": 1.638115
    },
    {
      "acc": 0.65190334,
      "epoch": 0.9837645865043125,
      "grad_norm": 6.8125,
      "learning_rate": 5.546497284004321e-06,
      "loss": 1.62859459,
      "memory(GiB)": 117.38,
      "step": 38780,
      "train_speed(iter/s)": 1.638141
    },
    {
      "acc": 0.66125822,
      "epoch": 0.9838914256722476,
      "grad_norm": 5.90625,
      "learning_rate": 5.545454930757233e-06,
      "loss": 1.60100193,
      "memory(GiB)": 117.38,
      "step": 38785,
      "train_speed(iter/s)": 1.638164
    },
    {
      "acc": 0.67107859,
      "epoch": 0.9840182648401826,
      "grad_norm": 5.125,
      "learning_rate": 5.544412553518602e-06,
      "loss": 1.52109604,
      "memory(GiB)": 117.38,
      "step": 38790,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.65286479,
      "epoch": 0.9841451040081177,
      "grad_norm": 4.71875,
      "learning_rate": 5.543370152334275e-06,
      "loss": 1.57824802,
      "memory(GiB)": 117.38,
      "step": 38795,
      "train_speed(iter/s)": 1.638212
    },
    {
      "acc": 0.65725117,
      "epoch": 0.9842719431760527,
      "grad_norm": 4.9375,
      "learning_rate": 5.542327727250105e-06,
      "loss": 1.59497948,
      "memory(GiB)": 117.38,
      "step": 38800,
      "train_speed(iter/s)": 1.638233
    },
    {
      "acc": 0.64453201,
      "epoch": 0.9843987823439878,
      "grad_norm": 6.09375,
      "learning_rate": 5.5412852783119385e-06,
      "loss": 1.62583885,
      "memory(GiB)": 117.38,
      "step": 38805,
      "train_speed(iter/s)": 1.638255
    },
    {
      "acc": 0.65223112,
      "epoch": 0.9845256215119229,
      "grad_norm": 4.78125,
      "learning_rate": 5.54024280556563e-06,
      "loss": 1.58456421,
      "memory(GiB)": 117.38,
      "step": 38810,
      "train_speed(iter/s)": 1.638279
    },
    {
      "acc": 0.64567976,
      "epoch": 0.9846524606798579,
      "grad_norm": 8.625,
      "learning_rate": 5.53920030905703e-06,
      "loss": 1.67348175,
      "memory(GiB)": 117.38,
      "step": 38815,
      "train_speed(iter/s)": 1.638303
    },
    {
      "acc": 0.65117941,
      "epoch": 0.984779299847793,
      "grad_norm": 5.15625,
      "learning_rate": 5.538157788831993e-06,
      "loss": 1.57780151,
      "memory(GiB)": 117.38,
      "step": 38820,
      "train_speed(iter/s)": 1.638328
    },
    {
      "acc": 0.65188837,
      "epoch": 0.9849061390157281,
      "grad_norm": 9.9375,
      "learning_rate": 5.537115244936374e-06,
      "loss": 1.61107025,
      "memory(GiB)": 117.38,
      "step": 38825,
      "train_speed(iter/s)": 1.638351
    },
    {
      "acc": 0.64141674,
      "epoch": 0.9850329781836631,
      "grad_norm": 6.6875,
      "learning_rate": 5.536072677416029e-06,
      "loss": 1.68645287,
      "memory(GiB)": 117.38,
      "step": 38830,
      "train_speed(iter/s)": 1.638374
    },
    {
      "acc": 0.66752872,
      "epoch": 0.9851598173515982,
      "grad_norm": 5.375,
      "learning_rate": 5.535030086316814e-06,
      "loss": 1.56166592,
      "memory(GiB)": 117.38,
      "step": 38835,
      "train_speed(iter/s)": 1.638397
    },
    {
      "acc": 0.66933012,
      "epoch": 0.9852866565195332,
      "grad_norm": 4.84375,
      "learning_rate": 5.533987471684586e-06,
      "loss": 1.5428009,
      "memory(GiB)": 117.38,
      "step": 38840,
      "train_speed(iter/s)": 1.638418
    },
    {
      "acc": 0.65214624,
      "epoch": 0.9854134956874683,
      "grad_norm": 5.03125,
      "learning_rate": 5.532944833565207e-06,
      "loss": 1.58116951,
      "memory(GiB)": 117.38,
      "step": 38845,
      "train_speed(iter/s)": 1.63844
    },
    {
      "acc": 0.64970388,
      "epoch": 0.9855403348554034,
      "grad_norm": 5.40625,
      "learning_rate": 5.531902172004533e-06,
      "loss": 1.61301765,
      "memory(GiB)": 117.38,
      "step": 38850,
      "train_speed(iter/s)": 1.638463
    },
    {
      "acc": 0.65473394,
      "epoch": 0.9856671740233384,
      "grad_norm": 5.625,
      "learning_rate": 5.530859487048427e-06,
      "loss": 1.62891655,
      "memory(GiB)": 117.38,
      "step": 38855,
      "train_speed(iter/s)": 1.638487
    },
    {
      "acc": 0.63526421,
      "epoch": 0.9857940131912735,
      "grad_norm": 4.75,
      "learning_rate": 5.529816778742752e-06,
      "loss": 1.70370007,
      "memory(GiB)": 117.38,
      "step": 38860,
      "train_speed(iter/s)": 1.638508
    },
    {
      "acc": 0.68229003,
      "epoch": 0.9859208523592086,
      "grad_norm": 6.0,
      "learning_rate": 5.528774047133369e-06,
      "loss": 1.50216904,
      "memory(GiB)": 117.38,
      "step": 38865,
      "train_speed(iter/s)": 1.638532
    },
    {
      "acc": 0.65571666,
      "epoch": 0.9860476915271436,
      "grad_norm": 5.0625,
      "learning_rate": 5.527731292266142e-06,
      "loss": 1.60538349,
      "memory(GiB)": 117.38,
      "step": 38870,
      "train_speed(iter/s)": 1.638556
    },
    {
      "acc": 0.66178646,
      "epoch": 0.9861745306950787,
      "grad_norm": 6.09375,
      "learning_rate": 5.5266885141869355e-06,
      "loss": 1.60124855,
      "memory(GiB)": 117.38,
      "step": 38875,
      "train_speed(iter/s)": 1.63858
    },
    {
      "acc": 0.65636034,
      "epoch": 0.9863013698630136,
      "grad_norm": 6.5,
      "learning_rate": 5.5256457129416185e-06,
      "loss": 1.63822937,
      "memory(GiB)": 117.38,
      "step": 38880,
      "train_speed(iter/s)": 1.638605
    },
    {
      "acc": 0.67386751,
      "epoch": 0.9864282090309487,
      "grad_norm": 5.5625,
      "learning_rate": 5.524602888576055e-06,
      "loss": 1.57072525,
      "memory(GiB)": 117.38,
      "step": 38885,
      "train_speed(iter/s)": 1.638628
    },
    {
      "acc": 0.63869648,
      "epoch": 0.9865550481988838,
      "grad_norm": 5.4375,
      "learning_rate": 5.523560041136116e-06,
      "loss": 1.65355835,
      "memory(GiB)": 117.38,
      "step": 38890,
      "train_speed(iter/s)": 1.638651
    },
    {
      "acc": 0.63306437,
      "epoch": 0.9866818873668188,
      "grad_norm": 5.4375,
      "learning_rate": 5.522517170667667e-06,
      "loss": 1.72260876,
      "memory(GiB)": 117.38,
      "step": 38895,
      "train_speed(iter/s)": 1.638676
    },
    {
      "acc": 0.66508541,
      "epoch": 0.9868087265347539,
      "grad_norm": 6.6875,
      "learning_rate": 5.5214742772165806e-06,
      "loss": 1.55520439,
      "memory(GiB)": 117.38,
      "step": 38900,
      "train_speed(iter/s)": 1.6387
    },
    {
      "acc": 0.63914924,
      "epoch": 0.986935565702689,
      "grad_norm": 6.84375,
      "learning_rate": 5.520431360828728e-06,
      "loss": 1.68224697,
      "memory(GiB)": 117.38,
      "step": 38905,
      "train_speed(iter/s)": 1.638723
    },
    {
      "acc": 0.64293098,
      "epoch": 0.987062404870624,
      "grad_norm": 6.1875,
      "learning_rate": 5.51938842154998e-06,
      "loss": 1.65292435,
      "memory(GiB)": 117.38,
      "step": 38910,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.65842671,
      "epoch": 0.9871892440385591,
      "grad_norm": 8.0,
      "learning_rate": 5.51834545942621e-06,
      "loss": 1.57294483,
      "memory(GiB)": 117.38,
      "step": 38915,
      "train_speed(iter/s)": 1.638769
    },
    {
      "acc": 0.65121107,
      "epoch": 0.9873160832064941,
      "grad_norm": 7.28125,
      "learning_rate": 5.5173024745032925e-06,
      "loss": 1.57889328,
      "memory(GiB)": 117.38,
      "step": 38920,
      "train_speed(iter/s)": 1.638794
    },
    {
      "acc": 0.65043373,
      "epoch": 0.9874429223744292,
      "grad_norm": 6.15625,
      "learning_rate": 5.516259466827103e-06,
      "loss": 1.64338379,
      "memory(GiB)": 117.38,
      "step": 38925,
      "train_speed(iter/s)": 1.638817
    },
    {
      "acc": 0.65469093,
      "epoch": 0.9875697615423643,
      "grad_norm": 6.375,
      "learning_rate": 5.515216436443517e-06,
      "loss": 1.58907661,
      "memory(GiB)": 117.38,
      "step": 38930,
      "train_speed(iter/s)": 1.638841
    },
    {
      "acc": 0.65526352,
      "epoch": 0.9876966007102993,
      "grad_norm": 7.15625,
      "learning_rate": 5.514173383398412e-06,
      "loss": 1.66379585,
      "memory(GiB)": 117.38,
      "step": 38935,
      "train_speed(iter/s)": 1.638865
    },
    {
      "acc": 0.65677443,
      "epoch": 0.9878234398782344,
      "grad_norm": 5.8125,
      "learning_rate": 5.513130307737666e-06,
      "loss": 1.62588825,
      "memory(GiB)": 117.38,
      "step": 38940,
      "train_speed(iter/s)": 1.638888
    },
    {
      "acc": 0.65021114,
      "epoch": 0.9879502790461695,
      "grad_norm": 4.8125,
      "learning_rate": 5.512087209507157e-06,
      "loss": 1.65093613,
      "memory(GiB)": 117.38,
      "step": 38945,
      "train_speed(iter/s)": 1.63891
    },
    {
      "acc": 0.66078768,
      "epoch": 0.9880771182141045,
      "grad_norm": 5.125,
      "learning_rate": 5.5110440887527684e-06,
      "loss": 1.52087193,
      "memory(GiB)": 117.38,
      "step": 38950,
      "train_speed(iter/s)": 1.638935
    },
    {
      "acc": 0.66037054,
      "epoch": 0.9882039573820396,
      "grad_norm": 4.5625,
      "learning_rate": 5.510000945520377e-06,
      "loss": 1.57690353,
      "memory(GiB)": 117.38,
      "step": 38955,
      "train_speed(iter/s)": 1.638957
    },
    {
      "acc": 0.65718465,
      "epoch": 0.9883307965499746,
      "grad_norm": 5.84375,
      "learning_rate": 5.508957779855869e-06,
      "loss": 1.61658516,
      "memory(GiB)": 117.38,
      "step": 38960,
      "train_speed(iter/s)": 1.63898
    },
    {
      "acc": 0.64686213,
      "epoch": 0.9884576357179097,
      "grad_norm": 6.15625,
      "learning_rate": 5.507914591805124e-06,
      "loss": 1.61408405,
      "memory(GiB)": 117.38,
      "step": 38965,
      "train_speed(iter/s)": 1.639003
    },
    {
      "acc": 0.65653453,
      "epoch": 0.9885844748858448,
      "grad_norm": 5.3125,
      "learning_rate": 5.506871381414027e-06,
      "loss": 1.57099218,
      "memory(GiB)": 117.38,
      "step": 38970,
      "train_speed(iter/s)": 1.639027
    },
    {
      "acc": 0.65301905,
      "epoch": 0.9887113140537798,
      "grad_norm": 6.625,
      "learning_rate": 5.505828148728465e-06,
      "loss": 1.63272476,
      "memory(GiB)": 117.38,
      "step": 38975,
      "train_speed(iter/s)": 1.639051
    },
    {
      "acc": 0.66440511,
      "epoch": 0.9888381532217149,
      "grad_norm": 5.71875,
      "learning_rate": 5.5047848937943225e-06,
      "loss": 1.59567413,
      "memory(GiB)": 117.38,
      "step": 38980,
      "train_speed(iter/s)": 1.639075
    },
    {
      "acc": 0.6642467,
      "epoch": 0.98896499238965,
      "grad_norm": 5.65625,
      "learning_rate": 5.503741616657486e-06,
      "loss": 1.57433796,
      "memory(GiB)": 117.38,
      "step": 38985,
      "train_speed(iter/s)": 1.639098
    },
    {
      "acc": 0.65406647,
      "epoch": 0.989091831557585,
      "grad_norm": 8.125,
      "learning_rate": 5.502698317363846e-06,
      "loss": 1.66263123,
      "memory(GiB)": 117.38,
      "step": 38990,
      "train_speed(iter/s)": 1.63912
    },
    {
      "acc": 0.64882717,
      "epoch": 0.98921867072552,
      "grad_norm": 6.21875,
      "learning_rate": 5.501654995959288e-06,
      "loss": 1.65302734,
      "memory(GiB)": 117.38,
      "step": 38995,
      "train_speed(iter/s)": 1.639144
    },
    {
      "acc": 0.64091539,
      "epoch": 0.989345509893455,
      "grad_norm": 6.5,
      "learning_rate": 5.5006116524897034e-06,
      "loss": 1.63139229,
      "memory(GiB)": 117.38,
      "step": 39000,
      "train_speed(iter/s)": 1.639167
    },
    {
      "epoch": 0.989345509893455,
      "eval_acc": 0.6460910042037602,
      "eval_loss": 1.574126958847046,
      "eval_runtime": 58.4403,
      "eval_samples_per_second": 109.0,
      "eval_steps_per_second": 27.259,
      "step": 39000
    },
    {
      "acc": 0.65300798,
      "epoch": 0.9894723490613901,
      "grad_norm": 6.40625,
      "learning_rate": 5.499568287000984e-06,
      "loss": 1.63923283,
      "memory(GiB)": 117.38,
      "step": 39005,
      "train_speed(iter/s)": 1.634883
    },
    {
      "acc": 0.66818128,
      "epoch": 0.9895991882293252,
      "grad_norm": 5.71875,
      "learning_rate": 5.49852489953902e-06,
      "loss": 1.6062746,
      "memory(GiB)": 117.38,
      "step": 39010,
      "train_speed(iter/s)": 1.634905
    },
    {
      "acc": 0.66122351,
      "epoch": 0.9897260273972602,
      "grad_norm": 6.6875,
      "learning_rate": 5.497481490149705e-06,
      "loss": 1.57987423,
      "memory(GiB)": 117.38,
      "step": 39015,
      "train_speed(iter/s)": 1.634928
    },
    {
      "acc": 0.65389156,
      "epoch": 0.9898528665651953,
      "grad_norm": 5.4375,
      "learning_rate": 5.496438058878936e-06,
      "loss": 1.62811966,
      "memory(GiB)": 117.38,
      "step": 39020,
      "train_speed(iter/s)": 1.63495
    },
    {
      "acc": 0.65526714,
      "epoch": 0.9899797057331304,
      "grad_norm": 5.40625,
      "learning_rate": 5.4953946057726005e-06,
      "loss": 1.63807335,
      "memory(GiB)": 117.38,
      "step": 39025,
      "train_speed(iter/s)": 1.634974
    },
    {
      "acc": 0.66559639,
      "epoch": 0.9901065449010654,
      "grad_norm": 5.1875,
      "learning_rate": 5.494351130876602e-06,
      "loss": 1.56376524,
      "memory(GiB)": 117.38,
      "step": 39030,
      "train_speed(iter/s)": 1.634997
    },
    {
      "acc": 0.67267427,
      "epoch": 0.9902333840690005,
      "grad_norm": 5.34375,
      "learning_rate": 5.493307634236831e-06,
      "loss": 1.56079607,
      "memory(GiB)": 117.38,
      "step": 39035,
      "train_speed(iter/s)": 1.635021
    },
    {
      "acc": 0.64773064,
      "epoch": 0.9903602232369355,
      "grad_norm": 5.59375,
      "learning_rate": 5.492264115899189e-06,
      "loss": 1.62359657,
      "memory(GiB)": 117.38,
      "step": 39040,
      "train_speed(iter/s)": 1.635043
    },
    {
      "acc": 0.65202203,
      "epoch": 0.9904870624048706,
      "grad_norm": 12.625,
      "learning_rate": 5.491220575909573e-06,
      "loss": 1.62284679,
      "memory(GiB)": 117.38,
      "step": 39045,
      "train_speed(iter/s)": 1.635066
    },
    {
      "acc": 0.65474219,
      "epoch": 0.9906139015728057,
      "grad_norm": 4.71875,
      "learning_rate": 5.4901770143138835e-06,
      "loss": 1.64360008,
      "memory(GiB)": 117.38,
      "step": 39050,
      "train_speed(iter/s)": 1.63509
    },
    {
      "acc": 0.65606775,
      "epoch": 0.9907407407407407,
      "grad_norm": 4.75,
      "learning_rate": 5.48913343115802e-06,
      "loss": 1.58373165,
      "memory(GiB)": 117.38,
      "step": 39055,
      "train_speed(iter/s)": 1.635112
    },
    {
      "acc": 0.64923234,
      "epoch": 0.9908675799086758,
      "grad_norm": 4.84375,
      "learning_rate": 5.488089826487884e-06,
      "loss": 1.58500471,
      "memory(GiB)": 117.38,
      "step": 39060,
      "train_speed(iter/s)": 1.635135
    },
    {
      "acc": 0.65942621,
      "epoch": 0.9909944190766109,
      "grad_norm": 4.9375,
      "learning_rate": 5.48704620034938e-06,
      "loss": 1.57913656,
      "memory(GiB)": 117.38,
      "step": 39065,
      "train_speed(iter/s)": 1.635147
    },
    {
      "acc": 0.67092609,
      "epoch": 0.9911212582445459,
      "grad_norm": 7.4375,
      "learning_rate": 5.486002552788408e-06,
      "loss": 1.60568542,
      "memory(GiB)": 117.38,
      "step": 39070,
      "train_speed(iter/s)": 1.63517
    },
    {
      "acc": 0.650278,
      "epoch": 0.991248097412481,
      "grad_norm": 6.0,
      "learning_rate": 5.4849588838508734e-06,
      "loss": 1.59628487,
      "memory(GiB)": 117.38,
      "step": 39075,
      "train_speed(iter/s)": 1.635193
    },
    {
      "acc": 0.6701498,
      "epoch": 0.991374936580416,
      "grad_norm": 6.0,
      "learning_rate": 5.483915193582684e-06,
      "loss": 1.57542067,
      "memory(GiB)": 117.38,
      "step": 39080,
      "train_speed(iter/s)": 1.635216
    },
    {
      "acc": 0.6399384,
      "epoch": 0.9915017757483511,
      "grad_norm": 5.21875,
      "learning_rate": 5.482871482029742e-06,
      "loss": 1.57752762,
      "memory(GiB)": 117.38,
      "step": 39085,
      "train_speed(iter/s)": 1.635238
    },
    {
      "acc": 0.66096268,
      "epoch": 0.9916286149162862,
      "grad_norm": 5.09375,
      "learning_rate": 5.4818277492379565e-06,
      "loss": 1.61035233,
      "memory(GiB)": 117.38,
      "step": 39090,
      "train_speed(iter/s)": 1.635261
    },
    {
      "acc": 0.66812725,
      "epoch": 0.9917554540842212,
      "grad_norm": 6.59375,
      "learning_rate": 5.480783995253236e-06,
      "loss": 1.58545609,
      "memory(GiB)": 117.38,
      "step": 39095,
      "train_speed(iter/s)": 1.635284
    },
    {
      "acc": 0.66740046,
      "epoch": 0.9918822932521563,
      "grad_norm": 6.5,
      "learning_rate": 5.47974022012149e-06,
      "loss": 1.58404446,
      "memory(GiB)": 117.38,
      "step": 39100,
      "train_speed(iter/s)": 1.635306
    },
    {
      "acc": 0.66551247,
      "epoch": 0.9920091324200914,
      "grad_norm": 6.65625,
      "learning_rate": 5.478696423888624e-06,
      "loss": 1.55421324,
      "memory(GiB)": 117.38,
      "step": 39105,
      "train_speed(iter/s)": 1.635329
    },
    {
      "acc": 0.65430903,
      "epoch": 0.9921359715880264,
      "grad_norm": 6.40625,
      "learning_rate": 5.477652606600555e-06,
      "loss": 1.5904953,
      "memory(GiB)": 117.38,
      "step": 39110,
      "train_speed(iter/s)": 1.635353
    },
    {
      "acc": 0.66596785,
      "epoch": 0.9922628107559615,
      "grad_norm": 6.0,
      "learning_rate": 5.47660876830319e-06,
      "loss": 1.52695141,
      "memory(GiB)": 117.38,
      "step": 39115,
      "train_speed(iter/s)": 1.635374
    },
    {
      "acc": 0.66659899,
      "epoch": 0.9923896499238964,
      "grad_norm": 6.3125,
      "learning_rate": 5.475564909042444e-06,
      "loss": 1.5241662,
      "memory(GiB)": 117.38,
      "step": 39120,
      "train_speed(iter/s)": 1.635393
    },
    {
      "acc": 0.65699735,
      "epoch": 0.9925164890918315,
      "grad_norm": 6.21875,
      "learning_rate": 5.4745210288642306e-06,
      "loss": 1.59540148,
      "memory(GiB)": 117.38,
      "step": 39125,
      "train_speed(iter/s)": 1.635416
    },
    {
      "acc": 0.63995919,
      "epoch": 0.9926433282597666,
      "grad_norm": 5.125,
      "learning_rate": 5.473477127814464e-06,
      "loss": 1.59047737,
      "memory(GiB)": 117.38,
      "step": 39130,
      "train_speed(iter/s)": 1.635439
    },
    {
      "acc": 0.63322458,
      "epoch": 0.9927701674277016,
      "grad_norm": 6.375,
      "learning_rate": 5.472433205939058e-06,
      "loss": 1.72567482,
      "memory(GiB)": 117.38,
      "step": 39135,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.67278895,
      "epoch": 0.9928970065956367,
      "grad_norm": 6.46875,
      "learning_rate": 5.471389263283932e-06,
      "loss": 1.55444756,
      "memory(GiB)": 117.38,
      "step": 39140,
      "train_speed(iter/s)": 1.635486
    },
    {
      "acc": 0.67533879,
      "epoch": 0.9930238457635718,
      "grad_norm": 6.53125,
      "learning_rate": 5.4703452998950005e-06,
      "loss": 1.52089148,
      "memory(GiB)": 117.38,
      "step": 39145,
      "train_speed(iter/s)": 1.635507
    },
    {
      "acc": 0.65287189,
      "epoch": 0.9931506849315068,
      "grad_norm": 5.40625,
      "learning_rate": 5.469301315818183e-06,
      "loss": 1.63992882,
      "memory(GiB)": 117.38,
      "step": 39150,
      "train_speed(iter/s)": 1.635529
    },
    {
      "acc": 0.65749435,
      "epoch": 0.9932775240994419,
      "grad_norm": 5.09375,
      "learning_rate": 5.468257311099399e-06,
      "loss": 1.49845467,
      "memory(GiB)": 117.38,
      "step": 39155,
      "train_speed(iter/s)": 1.635552
    },
    {
      "acc": 0.67615004,
      "epoch": 0.9934043632673769,
      "grad_norm": 7.03125,
      "learning_rate": 5.467213285784567e-06,
      "loss": 1.54736443,
      "memory(GiB)": 117.38,
      "step": 39160,
      "train_speed(iter/s)": 1.635576
    },
    {
      "acc": 0.65445733,
      "epoch": 0.993531202435312,
      "grad_norm": 5.8125,
      "learning_rate": 5.466169239919608e-06,
      "loss": 1.67408276,
      "memory(GiB)": 117.38,
      "step": 39165,
      "train_speed(iter/s)": 1.6356
    },
    {
      "acc": 0.65126104,
      "epoch": 0.9936580416032471,
      "grad_norm": 6.125,
      "learning_rate": 5.465125173550446e-06,
      "loss": 1.65996876,
      "memory(GiB)": 117.38,
      "step": 39170,
      "train_speed(iter/s)": 1.635623
    },
    {
      "acc": 0.65474653,
      "epoch": 0.9937848807711821,
      "grad_norm": 6.375,
      "learning_rate": 5.464081086723001e-06,
      "loss": 1.64905415,
      "memory(GiB)": 117.38,
      "step": 39175,
      "train_speed(iter/s)": 1.635647
    },
    {
      "acc": 0.64159436,
      "epoch": 0.9939117199391172,
      "grad_norm": 5.90625,
      "learning_rate": 5.4630369794832006e-06,
      "loss": 1.66727028,
      "memory(GiB)": 117.38,
      "step": 39180,
      "train_speed(iter/s)": 1.635669
    },
    {
      "acc": 0.66229115,
      "epoch": 0.9940385591070523,
      "grad_norm": 5.9375,
      "learning_rate": 5.461992851876963e-06,
      "loss": 1.6150219,
      "memory(GiB)": 117.38,
      "step": 39185,
      "train_speed(iter/s)": 1.63569
    },
    {
      "acc": 0.65428801,
      "epoch": 0.9941653982749873,
      "grad_norm": 8.4375,
      "learning_rate": 5.460948703950218e-06,
      "loss": 1.68495903,
      "memory(GiB)": 117.38,
      "step": 39190,
      "train_speed(iter/s)": 1.635712
    },
    {
      "acc": 0.65292454,
      "epoch": 0.9942922374429224,
      "grad_norm": 6.375,
      "learning_rate": 5.459904535748892e-06,
      "loss": 1.64405212,
      "memory(GiB)": 117.38,
      "step": 39195,
      "train_speed(iter/s)": 1.635736
    },
    {
      "acc": 0.68249865,
      "epoch": 0.9944190766108574,
      "grad_norm": 5.90625,
      "learning_rate": 5.458860347318912e-06,
      "loss": 1.50441771,
      "memory(GiB)": 117.38,
      "step": 39200,
      "train_speed(iter/s)": 1.635759
    },
    {
      "acc": 0.65553164,
      "epoch": 0.9945459157787925,
      "grad_norm": 6.15625,
      "learning_rate": 5.457816138706203e-06,
      "loss": 1.65315666,
      "memory(GiB)": 117.38,
      "step": 39205,
      "train_speed(iter/s)": 1.635783
    },
    {
      "acc": 0.66222887,
      "epoch": 0.9946727549467276,
      "grad_norm": 8.8125,
      "learning_rate": 5.456771909956697e-06,
      "loss": 1.60525055,
      "memory(GiB)": 117.38,
      "step": 39210,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.6699338,
      "epoch": 0.9947995941146626,
      "grad_norm": 6.90625,
      "learning_rate": 5.455727661116324e-06,
      "loss": 1.56743908,
      "memory(GiB)": 117.38,
      "step": 39215,
      "train_speed(iter/s)": 1.635831
    },
    {
      "acc": 0.66158605,
      "epoch": 0.9949264332825977,
      "grad_norm": 15.1875,
      "learning_rate": 5.454683392231014e-06,
      "loss": 1.63190041,
      "memory(GiB)": 117.38,
      "step": 39220,
      "train_speed(iter/s)": 1.635854
    },
    {
      "acc": 0.66914868,
      "epoch": 0.9950532724505328,
      "grad_norm": 5.5,
      "learning_rate": 5.453639103346697e-06,
      "loss": 1.55317698,
      "memory(GiB)": 117.38,
      "step": 39225,
      "train_speed(iter/s)": 1.635877
    },
    {
      "acc": 0.6589571,
      "epoch": 0.9951801116184678,
      "grad_norm": 7.375,
      "learning_rate": 5.452594794509307e-06,
      "loss": 1.53998489,
      "memory(GiB)": 117.38,
      "step": 39230,
      "train_speed(iter/s)": 1.6359
    },
    {
      "acc": 0.66344757,
      "epoch": 0.9953069507864029,
      "grad_norm": 5.625,
      "learning_rate": 5.4515504657647765e-06,
      "loss": 1.62008419,
      "memory(GiB)": 117.38,
      "step": 39235,
      "train_speed(iter/s)": 1.635923
    },
    {
      "acc": 0.65448818,
      "epoch": 0.9954337899543378,
      "grad_norm": 5.53125,
      "learning_rate": 5.450506117159044e-06,
      "loss": 1.68340836,
      "memory(GiB)": 117.38,
      "step": 39240,
      "train_speed(iter/s)": 1.635946
    },
    {
      "acc": 0.65918422,
      "epoch": 0.9955606291222729,
      "grad_norm": 7.1875,
      "learning_rate": 5.449461748738037e-06,
      "loss": 1.62783661,
      "memory(GiB)": 117.38,
      "step": 39245,
      "train_speed(iter/s)": 1.635969
    },
    {
      "acc": 0.66453824,
      "epoch": 0.995687468290208,
      "grad_norm": 6.09375,
      "learning_rate": 5.448417360547699e-06,
      "loss": 1.50588951,
      "memory(GiB)": 117.38,
      "step": 39250,
      "train_speed(iter/s)": 1.635991
    },
    {
      "acc": 0.66352844,
      "epoch": 0.995814307458143,
      "grad_norm": 5.28125,
      "learning_rate": 5.44737295263396e-06,
      "loss": 1.53560629,
      "memory(GiB)": 117.38,
      "step": 39255,
      "train_speed(iter/s)": 1.636014
    },
    {
      "acc": 0.64515104,
      "epoch": 0.9959411466260781,
      "grad_norm": 5.84375,
      "learning_rate": 5.446328525042764e-06,
      "loss": 1.67365685,
      "memory(GiB)": 117.38,
      "step": 39260,
      "train_speed(iter/s)": 1.636036
    },
    {
      "acc": 0.65106459,
      "epoch": 0.9960679857940132,
      "grad_norm": 6.40625,
      "learning_rate": 5.4452840778200456e-06,
      "loss": 1.6498579,
      "memory(GiB)": 117.38,
      "step": 39265,
      "train_speed(iter/s)": 1.63606
    },
    {
      "acc": 0.65146074,
      "epoch": 0.9961948249619482,
      "grad_norm": 6.59375,
      "learning_rate": 5.444239611011746e-06,
      "loss": 1.59827147,
      "memory(GiB)": 117.38,
      "step": 39270,
      "train_speed(iter/s)": 1.636083
    },
    {
      "acc": 0.65056009,
      "epoch": 0.9963216641298833,
      "grad_norm": 4.875,
      "learning_rate": 5.443195124663804e-06,
      "loss": 1.65174198,
      "memory(GiB)": 117.38,
      "step": 39275,
      "train_speed(iter/s)": 1.636106
    },
    {
      "acc": 0.65349693,
      "epoch": 0.9964485032978183,
      "grad_norm": 6.8125,
      "learning_rate": 5.442150618822162e-06,
      "loss": 1.61036835,
      "memory(GiB)": 117.38,
      "step": 39280,
      "train_speed(iter/s)": 1.63613
    },
    {
      "acc": 0.66394897,
      "epoch": 0.9965753424657534,
      "grad_norm": 5.15625,
      "learning_rate": 5.441106093532762e-06,
      "loss": 1.64434776,
      "memory(GiB)": 117.38,
      "step": 39285,
      "train_speed(iter/s)": 1.636152
    },
    {
      "acc": 0.66421161,
      "epoch": 0.9967021816336885,
      "grad_norm": 4.71875,
      "learning_rate": 5.440061548841546e-06,
      "loss": 1.57276306,
      "memory(GiB)": 117.38,
      "step": 39290,
      "train_speed(iter/s)": 1.636177
    },
    {
      "acc": 0.66076994,
      "epoch": 0.9968290208016235,
      "grad_norm": 6.90625,
      "learning_rate": 5.43901698479446e-06,
      "loss": 1.60429821,
      "memory(GiB)": 117.38,
      "step": 39295,
      "train_speed(iter/s)": 1.636201
    },
    {
      "acc": 0.63934999,
      "epoch": 0.9969558599695586,
      "grad_norm": 5.75,
      "learning_rate": 5.4379724014374455e-06,
      "loss": 1.65393124,
      "memory(GiB)": 117.38,
      "step": 39300,
      "train_speed(iter/s)": 1.636227
    },
    {
      "acc": 0.64348688,
      "epoch": 0.9970826991374937,
      "grad_norm": 4.90625,
      "learning_rate": 5.436927798816448e-06,
      "loss": 1.63104057,
      "memory(GiB)": 117.38,
      "step": 39305,
      "train_speed(iter/s)": 1.63625
    },
    {
      "acc": 0.67017736,
      "epoch": 0.9972095383054287,
      "grad_norm": 11.5,
      "learning_rate": 5.4358831769774174e-06,
      "loss": 1.54743223,
      "memory(GiB)": 117.38,
      "step": 39310,
      "train_speed(iter/s)": 1.636273
    },
    {
      "acc": 0.65907726,
      "epoch": 0.9973363774733638,
      "grad_norm": 5.15625,
      "learning_rate": 5.434838535966298e-06,
      "loss": 1.65316467,
      "memory(GiB)": 117.38,
      "step": 39315,
      "train_speed(iter/s)": 1.636295
    },
    {
      "acc": 0.64373393,
      "epoch": 0.9974632166412988,
      "grad_norm": 6.34375,
      "learning_rate": 5.43379387582904e-06,
      "loss": 1.61678085,
      "memory(GiB)": 117.38,
      "step": 39320,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.65938292,
      "epoch": 0.9975900558092339,
      "grad_norm": 5.3125,
      "learning_rate": 5.432749196611587e-06,
      "loss": 1.5760704,
      "memory(GiB)": 117.38,
      "step": 39325,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.65438557,
      "epoch": 0.997716894977169,
      "grad_norm": 5.71875,
      "learning_rate": 5.431704498359896e-06,
      "loss": 1.58403225,
      "memory(GiB)": 117.38,
      "step": 39330,
      "train_speed(iter/s)": 1.636369
    },
    {
      "acc": 0.66184835,
      "epoch": 0.997843734145104,
      "grad_norm": 6.78125,
      "learning_rate": 5.43065978111991e-06,
      "loss": 1.60483894,
      "memory(GiB)": 117.38,
      "step": 39335,
      "train_speed(iter/s)": 1.636391
    },
    {
      "acc": 0.65278807,
      "epoch": 0.9979705733130391,
      "grad_norm": 7.15625,
      "learning_rate": 5.429615044937586e-06,
      "loss": 1.57684555,
      "memory(GiB)": 117.38,
      "step": 39340,
      "train_speed(iter/s)": 1.636416
    },
    {
      "acc": 0.65417552,
      "epoch": 0.9980974124809742,
      "grad_norm": 5.5625,
      "learning_rate": 5.4285702898588754e-06,
      "loss": 1.6590395,
      "memory(GiB)": 117.38,
      "step": 39345,
      "train_speed(iter/s)": 1.636438
    },
    {
      "acc": 0.65630341,
      "epoch": 0.9982242516489092,
      "grad_norm": 5.1875,
      "learning_rate": 5.427525515929729e-06,
      "loss": 1.5860218,
      "memory(GiB)": 117.38,
      "step": 39350,
      "train_speed(iter/s)": 1.636463
    },
    {
      "acc": 0.67345562,
      "epoch": 0.9983510908168443,
      "grad_norm": 6.625,
      "learning_rate": 5.426480723196102e-06,
      "loss": 1.49641132,
      "memory(GiB)": 117.38,
      "step": 39355,
      "train_speed(iter/s)": 1.636487
    },
    {
      "acc": 0.6565259,
      "epoch": 0.9984779299847792,
      "grad_norm": 5.5625,
      "learning_rate": 5.425435911703948e-06,
      "loss": 1.60635338,
      "memory(GiB)": 117.38,
      "step": 39360,
      "train_speed(iter/s)": 1.636509
    },
    {
      "acc": 0.64514771,
      "epoch": 0.9986047691527143,
      "grad_norm": 6.125,
      "learning_rate": 5.424391081499223e-06,
      "loss": 1.61983109,
      "memory(GiB)": 117.38,
      "step": 39365,
      "train_speed(iter/s)": 1.636532
    },
    {
      "acc": 0.65590596,
      "epoch": 0.9987316083206494,
      "grad_norm": 5.0,
      "learning_rate": 5.423346232627884e-06,
      "loss": 1.60533066,
      "memory(GiB)": 117.38,
      "step": 39370,
      "train_speed(iter/s)": 1.636556
    },
    {
      "acc": 0.64388051,
      "epoch": 0.9988584474885844,
      "grad_norm": 6.71875,
      "learning_rate": 5.422301365135887e-06,
      "loss": 1.66110821,
      "memory(GiB)": 117.38,
      "step": 39375,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.63931942,
      "epoch": 0.9989852866565195,
      "grad_norm": 6.21875,
      "learning_rate": 5.421256479069191e-06,
      "loss": 1.62837257,
      "memory(GiB)": 117.38,
      "step": 39380,
      "train_speed(iter/s)": 1.636604
    },
    {
      "acc": 0.66496649,
      "epoch": 0.9991121258244546,
      "grad_norm": 6.5,
      "learning_rate": 5.420211574473754e-06,
      "loss": 1.54188709,
      "memory(GiB)": 117.38,
      "step": 39385,
      "train_speed(iter/s)": 1.636629
    },
    {
      "acc": 0.65972557,
      "epoch": 0.9992389649923896,
      "grad_norm": 5.5,
      "learning_rate": 5.419166651395536e-06,
      "loss": 1.57559299,
      "memory(GiB)": 117.38,
      "step": 39390,
      "train_speed(iter/s)": 1.636652
    },
    {
      "acc": 0.66274433,
      "epoch": 0.9993658041603247,
      "grad_norm": 6.34375,
      "learning_rate": 5.418121709880497e-06,
      "loss": 1.57407169,
      "memory(GiB)": 117.38,
      "step": 39395,
      "train_speed(iter/s)": 1.636676
    },
    {
      "acc": 0.65294142,
      "epoch": 0.9994926433282597,
      "grad_norm": 6.375,
      "learning_rate": 5.4170767499746e-06,
      "loss": 1.59775505,
      "memory(GiB)": 117.38,
      "step": 39400,
      "train_speed(iter/s)": 1.636699
    },
    {
      "acc": 0.65430641,
      "epoch": 0.9996194824961948,
      "grad_norm": 6.34375,
      "learning_rate": 5.416031771723803e-06,
      "loss": 1.63736858,
      "memory(GiB)": 117.38,
      "step": 39405,
      "train_speed(iter/s)": 1.636723
    },
    {
      "acc": 0.65762796,
      "epoch": 0.9997463216641299,
      "grad_norm": 7.1875,
      "learning_rate": 5.414986775174073e-06,
      "loss": 1.57327948,
      "memory(GiB)": 117.38,
      "step": 39410,
      "train_speed(iter/s)": 1.636748
    },
    {
      "acc": 0.66443744,
      "epoch": 0.9998731608320649,
      "grad_norm": 5.0625,
      "learning_rate": 5.41394176037137e-06,
      "loss": 1.57263412,
      "memory(GiB)": 117.38,
      "step": 39415,
      "train_speed(iter/s)": 1.636771
    },
    {
      "acc": 0.64962502,
      "epoch": 1.0,
      "grad_norm": 6.375,
      "learning_rate": 5.412896727361663e-06,
      "loss": 1.66451626,
      "memory(GiB)": 117.38,
      "step": 39420,
      "train_speed(iter/s)": 1.636743
    },
    {
      "acc": 0.66950126,
      "epoch": 1.000126839167935,
      "grad_norm": 5.90625,
      "learning_rate": 5.411851676190912e-06,
      "loss": 1.56461639,
      "memory(GiB)": 117.38,
      "step": 39425,
      "train_speed(iter/s)": 1.636739
    },
    {
      "acc": 0.64897671,
      "epoch": 1.0002536783358702,
      "grad_norm": 5.4375,
      "learning_rate": 5.4108066069050864e-06,
      "loss": 1.63193035,
      "memory(GiB)": 117.38,
      "step": 39430,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.62981844,
      "epoch": 1.0003805175038052,
      "grad_norm": 5.34375,
      "learning_rate": 5.409761519550153e-06,
      "loss": 1.65145054,
      "memory(GiB)": 117.38,
      "step": 39435,
      "train_speed(iter/s)": 1.636783
    },
    {
      "acc": 0.6696702,
      "epoch": 1.0005073566717402,
      "grad_norm": 6.125,
      "learning_rate": 5.408716414172077e-06,
      "loss": 1.55201426,
      "memory(GiB)": 117.38,
      "step": 39440,
      "train_speed(iter/s)": 1.636804
    },
    {
      "acc": 0.66762285,
      "epoch": 1.0006341958396754,
      "grad_norm": 7.0625,
      "learning_rate": 5.407671290816829e-06,
      "loss": 1.60498619,
      "memory(GiB)": 117.38,
      "step": 39445,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.63307271,
      "epoch": 1.0007610350076104,
      "grad_norm": 5.84375,
      "learning_rate": 5.406626149530378e-06,
      "loss": 1.72459278,
      "memory(GiB)": 117.38,
      "step": 39450,
      "train_speed(iter/s)": 1.636847
    },
    {
      "acc": 0.67057791,
      "epoch": 1.0008878741755454,
      "grad_norm": 4.96875,
      "learning_rate": 5.405580990358692e-06,
      "loss": 1.51603107,
      "memory(GiB)": 117.38,
      "step": 39455,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.66682401,
      "epoch": 1.0010147133434804,
      "grad_norm": 7.0625,
      "learning_rate": 5.404535813347746e-06,
      "loss": 1.57820301,
      "memory(GiB)": 117.38,
      "step": 39460,
      "train_speed(iter/s)": 1.636891
    },
    {
      "acc": 0.66099606,
      "epoch": 1.0011415525114156,
      "grad_norm": 8.1875,
      "learning_rate": 5.403490618543505e-06,
      "loss": 1.63314533,
      "memory(GiB)": 117.38,
      "step": 39465,
      "train_speed(iter/s)": 1.636913
    },
    {
      "acc": 0.67097068,
      "epoch": 1.0012683916793506,
      "grad_norm": 5.96875,
      "learning_rate": 5.40244540599195e-06,
      "loss": 1.56456203,
      "memory(GiB)": 117.38,
      "step": 39470,
      "train_speed(iter/s)": 1.636936
    },
    {
      "acc": 0.66049309,
      "epoch": 1.0013952308472855,
      "grad_norm": 5.0,
      "learning_rate": 5.401400175739045e-06,
      "loss": 1.63894768,
      "memory(GiB)": 117.38,
      "step": 39475,
      "train_speed(iter/s)": 1.636959
    },
    {
      "acc": 0.64392538,
      "epoch": 1.0015220700152208,
      "grad_norm": 6.8125,
      "learning_rate": 5.400354927830769e-06,
      "loss": 1.64915848,
      "memory(GiB)": 117.38,
      "step": 39480,
      "train_speed(iter/s)": 1.636985
    },
    {
      "acc": 0.66596642,
      "epoch": 1.0016489091831557,
      "grad_norm": 6.46875,
      "learning_rate": 5.399309662313097e-06,
      "loss": 1.56436806,
      "memory(GiB)": 117.38,
      "step": 39485,
      "train_speed(iter/s)": 1.637008
    },
    {
      "acc": 0.64307508,
      "epoch": 1.0017757483510907,
      "grad_norm": 5.6875,
      "learning_rate": 5.3982643792320024e-06,
      "loss": 1.61942768,
      "memory(GiB)": 117.38,
      "step": 39490,
      "train_speed(iter/s)": 1.63703
    },
    {
      "acc": 0.65330992,
      "epoch": 1.001902587519026,
      "grad_norm": 5.09375,
      "learning_rate": 5.397219078633462e-06,
      "loss": 1.59372082,
      "memory(GiB)": 117.38,
      "step": 39495,
      "train_speed(iter/s)": 1.637052
    },
    {
      "acc": 0.65937791,
      "epoch": 1.002029426686961,
      "grad_norm": 6.6875,
      "learning_rate": 5.3961737605634546e-06,
      "loss": 1.61160278,
      "memory(GiB)": 117.38,
      "step": 39500,
      "train_speed(iter/s)": 1.637075
    },
    {
      "acc": 0.66893864,
      "epoch": 1.002156265854896,
      "grad_norm": 5.90625,
      "learning_rate": 5.395128425067954e-06,
      "loss": 1.60031319,
      "memory(GiB)": 117.38,
      "step": 39505,
      "train_speed(iter/s)": 1.637098
    },
    {
      "acc": 0.68094864,
      "epoch": 1.0022831050228311,
      "grad_norm": 5.0625,
      "learning_rate": 5.394083072192944e-06,
      "loss": 1.51652832,
      "memory(GiB)": 117.38,
      "step": 39510,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.67578306,
      "epoch": 1.0024099441907661,
      "grad_norm": 6.3125,
      "learning_rate": 5.393037701984399e-06,
      "loss": 1.51201839,
      "memory(GiB)": 117.38,
      "step": 39515,
      "train_speed(iter/s)": 1.637143
    },
    {
      "acc": 0.64685206,
      "epoch": 1.0025367833587011,
      "grad_norm": 5.21875,
      "learning_rate": 5.391992314488303e-06,
      "loss": 1.58944092,
      "memory(GiB)": 117.38,
      "step": 39520,
      "train_speed(iter/s)": 1.637166
    },
    {
      "acc": 0.64986658,
      "epoch": 1.0026636225266363,
      "grad_norm": 5.0625,
      "learning_rate": 5.3909469097506314e-06,
      "loss": 1.59459209,
      "memory(GiB)": 117.38,
      "step": 39525,
      "train_speed(iter/s)": 1.637188
    },
    {
      "acc": 0.63995323,
      "epoch": 1.0027904616945713,
      "grad_norm": 5.96875,
      "learning_rate": 5.389901487817373e-06,
      "loss": 1.62200928,
      "memory(GiB)": 117.38,
      "step": 39530,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.6651732,
      "epoch": 1.0029173008625063,
      "grad_norm": 6.25,
      "learning_rate": 5.388856048734505e-06,
      "loss": 1.60574818,
      "memory(GiB)": 117.38,
      "step": 39535,
      "train_speed(iter/s)": 1.637235
    },
    {
      "acc": 0.6459837,
      "epoch": 1.0030441400304415,
      "grad_norm": 5.5625,
      "learning_rate": 5.3878105925480115e-06,
      "loss": 1.70576916,
      "memory(GiB)": 117.38,
      "step": 39540,
      "train_speed(iter/s)": 1.637258
    },
    {
      "acc": 0.66164217,
      "epoch": 1.0031709791983765,
      "grad_norm": 5.03125,
      "learning_rate": 5.3867651193038765e-06,
      "loss": 1.60276623,
      "memory(GiB)": 117.38,
      "step": 39545,
      "train_speed(iter/s)": 1.63728
    },
    {
      "acc": 0.65812364,
      "epoch": 1.0032978183663115,
      "grad_norm": 5.625,
      "learning_rate": 5.385719629048086e-06,
      "loss": 1.5767518,
      "memory(GiB)": 117.38,
      "step": 39550,
      "train_speed(iter/s)": 1.637302
    },
    {
      "acc": 0.64759388,
      "epoch": 1.0034246575342465,
      "grad_norm": 4.96875,
      "learning_rate": 5.384674121826622e-06,
      "loss": 1.64803886,
      "memory(GiB)": 117.38,
      "step": 39555,
      "train_speed(iter/s)": 1.637324
    },
    {
      "acc": 0.6585988,
      "epoch": 1.0035514967021817,
      "grad_norm": 6.28125,
      "learning_rate": 5.383628597685474e-06,
      "loss": 1.58061104,
      "memory(GiB)": 117.38,
      "step": 39560,
      "train_speed(iter/s)": 1.637348
    },
    {
      "acc": 0.67884254,
      "epoch": 1.0036783358701167,
      "grad_norm": 5.75,
      "learning_rate": 5.382583056670627e-06,
      "loss": 1.52536678,
      "memory(GiB)": 117.38,
      "step": 39565,
      "train_speed(iter/s)": 1.63737
    },
    {
      "acc": 0.65091953,
      "epoch": 1.0038051750380517,
      "grad_norm": 6.6875,
      "learning_rate": 5.38153749882807e-06,
      "loss": 1.63830948,
      "memory(GiB)": 117.38,
      "step": 39570,
      "train_speed(iter/s)": 1.637394
    },
    {
      "acc": 0.65617323,
      "epoch": 1.0039320142059869,
      "grad_norm": 6.15625,
      "learning_rate": 5.38049192420379e-06,
      "loss": 1.58975906,
      "memory(GiB)": 117.38,
      "step": 39575,
      "train_speed(iter/s)": 1.637416
    },
    {
      "acc": 0.65104179,
      "epoch": 1.0040588533739219,
      "grad_norm": 4.53125,
      "learning_rate": 5.3794463328437766e-06,
      "loss": 1.60249252,
      "memory(GiB)": 117.38,
      "step": 39580,
      "train_speed(iter/s)": 1.637439
    },
    {
      "acc": 0.66326785,
      "epoch": 1.0041856925418569,
      "grad_norm": 6.6875,
      "learning_rate": 5.3784007247940185e-06,
      "loss": 1.52351313,
      "memory(GiB)": 117.38,
      "step": 39585,
      "train_speed(iter/s)": 1.637459
    },
    {
      "acc": 0.65451446,
      "epoch": 1.004312531709792,
      "grad_norm": 9.0625,
      "learning_rate": 5.377355100100508e-06,
      "loss": 1.53261662,
      "memory(GiB)": 117.38,
      "step": 39590,
      "train_speed(iter/s)": 1.63748
    },
    {
      "acc": 0.66253719,
      "epoch": 1.004439370877727,
      "grad_norm": 6.625,
      "learning_rate": 5.376309458809235e-06,
      "loss": 1.5595974,
      "memory(GiB)": 117.38,
      "step": 39595,
      "train_speed(iter/s)": 1.637501
    },
    {
      "acc": 0.64328623,
      "epoch": 1.004566210045662,
      "grad_norm": 6.03125,
      "learning_rate": 5.375263800966192e-06,
      "loss": 1.62471447,
      "memory(GiB)": 117.38,
      "step": 39600,
      "train_speed(iter/s)": 1.637521
    },
    {
      "acc": 0.64843526,
      "epoch": 1.0046930492135973,
      "grad_norm": 7.34375,
      "learning_rate": 5.374218126617371e-06,
      "loss": 1.57931347,
      "memory(GiB)": 117.38,
      "step": 39605,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.6540575,
      "epoch": 1.0048198883815322,
      "grad_norm": 5.21875,
      "learning_rate": 5.373172435808768e-06,
      "loss": 1.62933846,
      "memory(GiB)": 117.38,
      "step": 39610,
      "train_speed(iter/s)": 1.637562
    },
    {
      "acc": 0.65167828,
      "epoch": 1.0049467275494672,
      "grad_norm": 6.625,
      "learning_rate": 5.372126728586372e-06,
      "loss": 1.64481163,
      "memory(GiB)": 117.38,
      "step": 39615,
      "train_speed(iter/s)": 1.637583
    },
    {
      "acc": 0.6512394,
      "epoch": 1.0050735667174022,
      "grad_norm": 5.375,
      "learning_rate": 5.371081004996184e-06,
      "loss": 1.57376308,
      "memory(GiB)": 117.38,
      "step": 39620,
      "train_speed(iter/s)": 1.637602
    },
    {
      "acc": 0.64877143,
      "epoch": 1.0052004058853374,
      "grad_norm": 5.34375,
      "learning_rate": 5.370035265084195e-06,
      "loss": 1.59633064,
      "memory(GiB)": 117.38,
      "step": 39625,
      "train_speed(iter/s)": 1.637623
    },
    {
      "acc": 0.65403261,
      "epoch": 1.0053272450532724,
      "grad_norm": 6.71875,
      "learning_rate": 5.3689895088964025e-06,
      "loss": 1.63066387,
      "memory(GiB)": 117.38,
      "step": 39630,
      "train_speed(iter/s)": 1.637644
    },
    {
      "acc": 0.64735985,
      "epoch": 1.0054540842212074,
      "grad_norm": 6.0,
      "learning_rate": 5.367943736478806e-06,
      "loss": 1.62418671,
      "memory(GiB)": 117.38,
      "step": 39635,
      "train_speed(iter/s)": 1.637664
    },
    {
      "acc": 0.64825654,
      "epoch": 1.0055809233891426,
      "grad_norm": 5.875,
      "learning_rate": 5.3668979478774e-06,
      "loss": 1.67106895,
      "memory(GiB)": 117.38,
      "step": 39640,
      "train_speed(iter/s)": 1.637685
    },
    {
      "acc": 0.66372509,
      "epoch": 1.0057077625570776,
      "grad_norm": 5.71875,
      "learning_rate": 5.3658521431381836e-06,
      "loss": 1.56859446,
      "memory(GiB)": 117.38,
      "step": 39645,
      "train_speed(iter/s)": 1.637708
    },
    {
      "acc": 0.65510969,
      "epoch": 1.0058346017250126,
      "grad_norm": 5.875,
      "learning_rate": 5.364806322307158e-06,
      "loss": 1.58570452,
      "memory(GiB)": 117.38,
      "step": 39650,
      "train_speed(iter/s)": 1.637731
    },
    {
      "acc": 0.64684038,
      "epoch": 1.0059614408929478,
      "grad_norm": 5.9375,
      "learning_rate": 5.363760485430321e-06,
      "loss": 1.6566288,
      "memory(GiB)": 117.38,
      "step": 39655,
      "train_speed(iter/s)": 1.63775
    },
    {
      "acc": 0.66637049,
      "epoch": 1.0060882800608828,
      "grad_norm": 5.8125,
      "learning_rate": 5.3627146325536725e-06,
      "loss": 1.62865295,
      "memory(GiB)": 117.38,
      "step": 39660,
      "train_speed(iter/s)": 1.637772
    },
    {
      "acc": 0.67953014,
      "epoch": 1.0062151192288178,
      "grad_norm": 5.96875,
      "learning_rate": 5.361668763723216e-06,
      "loss": 1.46089296,
      "memory(GiB)": 117.38,
      "step": 39665,
      "train_speed(iter/s)": 1.637793
    },
    {
      "acc": 0.65385904,
      "epoch": 1.006341958396753,
      "grad_norm": 5.46875,
      "learning_rate": 5.360622878984954e-06,
      "loss": 1.57221041,
      "memory(GiB)": 117.38,
      "step": 39670,
      "train_speed(iter/s)": 1.637814
    },
    {
      "acc": 0.65011883,
      "epoch": 1.006468797564688,
      "grad_norm": 5.625,
      "learning_rate": 5.359576978384885e-06,
      "loss": 1.62987633,
      "memory(GiB)": 117.38,
      "step": 39675,
      "train_speed(iter/s)": 1.637836
    },
    {
      "acc": 0.66177073,
      "epoch": 1.006595636732623,
      "grad_norm": 5.375,
      "learning_rate": 5.358531061969018e-06,
      "loss": 1.56385326,
      "memory(GiB)": 117.38,
      "step": 39680,
      "train_speed(iter/s)": 1.637858
    },
    {
      "acc": 0.65169616,
      "epoch": 1.0067224759005582,
      "grad_norm": 6.1875,
      "learning_rate": 5.357485129783351e-06,
      "loss": 1.67378426,
      "memory(GiB)": 117.38,
      "step": 39685,
      "train_speed(iter/s)": 1.637879
    },
    {
      "acc": 0.65342159,
      "epoch": 1.0068493150684932,
      "grad_norm": 5.9375,
      "learning_rate": 5.356439181873895e-06,
      "loss": 1.60303345,
      "memory(GiB)": 117.38,
      "step": 39690,
      "train_speed(iter/s)": 1.637901
    },
    {
      "acc": 0.66733513,
      "epoch": 1.0069761542364282,
      "grad_norm": 5.625,
      "learning_rate": 5.35539321828665e-06,
      "loss": 1.52794981,
      "memory(GiB)": 117.38,
      "step": 39695,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.65667996,
      "epoch": 1.0071029934043634,
      "grad_norm": 5.1875,
      "learning_rate": 5.354347239067625e-06,
      "loss": 1.57421608,
      "memory(GiB)": 117.38,
      "step": 39700,
      "train_speed(iter/s)": 1.637942
    },
    {
      "acc": 0.66278677,
      "epoch": 1.0072298325722984,
      "grad_norm": 6.0,
      "learning_rate": 5.3533012442628275e-06,
      "loss": 1.52947922,
      "memory(GiB)": 117.38,
      "step": 39705,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.66822367,
      "epoch": 1.0073566717402334,
      "grad_norm": 8.125,
      "learning_rate": 5.3522552339182635e-06,
      "loss": 1.51784086,
      "memory(GiB)": 117.38,
      "step": 39710,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.65919166,
      "epoch": 1.0074835109081683,
      "grad_norm": 5.875,
      "learning_rate": 5.351209208079941e-06,
      "loss": 1.62209034,
      "memory(GiB)": 117.38,
      "step": 39715,
      "train_speed(iter/s)": 1.638008
    },
    {
      "acc": 0.65563951,
      "epoch": 1.0076103500761036,
      "grad_norm": 6.4375,
      "learning_rate": 5.35016316679387e-06,
      "loss": 1.60887737,
      "memory(GiB)": 117.38,
      "step": 39720,
      "train_speed(iter/s)": 1.63803
    },
    {
      "acc": 0.66711435,
      "epoch": 1.0077371892440385,
      "grad_norm": 7.0,
      "learning_rate": 5.349117110106059e-06,
      "loss": 1.49012108,
      "memory(GiB)": 117.38,
      "step": 39725,
      "train_speed(iter/s)": 1.638054
    },
    {
      "acc": 0.65878925,
      "epoch": 1.0078640284119735,
      "grad_norm": 6.375,
      "learning_rate": 5.34807103806252e-06,
      "loss": 1.59575462,
      "memory(GiB)": 117.38,
      "step": 39730,
      "train_speed(iter/s)": 1.638076
    },
    {
      "acc": 0.64732299,
      "epoch": 1.0079908675799087,
      "grad_norm": 5.71875,
      "learning_rate": 5.347024950709262e-06,
      "loss": 1.60237503,
      "memory(GiB)": 117.38,
      "step": 39735,
      "train_speed(iter/s)": 1.638098
    },
    {
      "acc": 0.64321766,
      "epoch": 1.0081177067478437,
      "grad_norm": 5.53125,
      "learning_rate": 5.345978848092297e-06,
      "loss": 1.66265335,
      "memory(GiB)": 117.38,
      "step": 39740,
      "train_speed(iter/s)": 1.638121
    },
    {
      "acc": 0.65173054,
      "epoch": 1.0082445459157787,
      "grad_norm": 7.625,
      "learning_rate": 5.344932730257637e-06,
      "loss": 1.58735352,
      "memory(GiB)": 117.38,
      "step": 39745,
      "train_speed(iter/s)": 1.638143
    },
    {
      "acc": 0.6556314,
      "epoch": 1.008371385083714,
      "grad_norm": 6.875,
      "learning_rate": 5.343886597251298e-06,
      "loss": 1.60826073,
      "memory(GiB)": 117.38,
      "step": 39750,
      "train_speed(iter/s)": 1.638165
    },
    {
      "acc": 0.66044836,
      "epoch": 1.008498224251649,
      "grad_norm": 5.375,
      "learning_rate": 5.342840449119287e-06,
      "loss": 1.64020424,
      "memory(GiB)": 117.38,
      "step": 39755,
      "train_speed(iter/s)": 1.638187
    },
    {
      "acc": 0.65782261,
      "epoch": 1.008625063419584,
      "grad_norm": 6.03125,
      "learning_rate": 5.341794285907627e-06,
      "loss": 1.60111179,
      "memory(GiB)": 117.38,
      "step": 39760,
      "train_speed(iter/s)": 1.638209
    },
    {
      "acc": 0.64857674,
      "epoch": 1.0087519025875191,
      "grad_norm": 5.84375,
      "learning_rate": 5.340748107662324e-06,
      "loss": 1.66032734,
      "memory(GiB)": 117.38,
      "step": 39765,
      "train_speed(iter/s)": 1.638232
    },
    {
      "acc": 0.65215044,
      "epoch": 1.0088787417554541,
      "grad_norm": 5.46875,
      "learning_rate": 5.339701914429402e-06,
      "loss": 1.62041969,
      "memory(GiB)": 117.38,
      "step": 39770,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.647154,
      "epoch": 1.009005580923389,
      "grad_norm": 5.96875,
      "learning_rate": 5.338655706254871e-06,
      "loss": 1.72091618,
      "memory(GiB)": 117.38,
      "step": 39775,
      "train_speed(iter/s)": 1.638276
    },
    {
      "acc": 0.6473125,
      "epoch": 1.009132420091324,
      "grad_norm": 4.625,
      "learning_rate": 5.33760948318475e-06,
      "loss": 1.61755104,
      "memory(GiB)": 117.38,
      "step": 39780,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.6669251,
      "epoch": 1.0092592592592593,
      "grad_norm": 6.09375,
      "learning_rate": 5.336563245265056e-06,
      "loss": 1.63113251,
      "memory(GiB)": 117.38,
      "step": 39785,
      "train_speed(iter/s)": 1.63832
    },
    {
      "acc": 0.64419541,
      "epoch": 1.0093860984271943,
      "grad_norm": 6.75,
      "learning_rate": 5.3355169925418095e-06,
      "loss": 1.69696808,
      "memory(GiB)": 117.38,
      "step": 39790,
      "train_speed(iter/s)": 1.638342
    },
    {
      "acc": 0.65545344,
      "epoch": 1.0095129375951293,
      "grad_norm": 5.1875,
      "learning_rate": 5.334470725061027e-06,
      "loss": 1.64371738,
      "memory(GiB)": 117.38,
      "step": 39795,
      "train_speed(iter/s)": 1.638364
    },
    {
      "acc": 0.65667014,
      "epoch": 1.0096397767630645,
      "grad_norm": 5.625,
      "learning_rate": 5.333424442868729e-06,
      "loss": 1.63895397,
      "memory(GiB)": 117.38,
      "step": 39800,
      "train_speed(iter/s)": 1.638386
    },
    {
      "acc": 0.65159626,
      "epoch": 1.0097666159309995,
      "grad_norm": 6.5625,
      "learning_rate": 5.3323781460109345e-06,
      "loss": 1.65605927,
      "memory(GiB)": 117.38,
      "step": 39805,
      "train_speed(iter/s)": 1.638407
    },
    {
      "acc": 0.65893574,
      "epoch": 1.0098934550989345,
      "grad_norm": 5.9375,
      "learning_rate": 5.3313318345336665e-06,
      "loss": 1.59443979,
      "memory(GiB)": 117.38,
      "step": 39810,
      "train_speed(iter/s)": 1.63843
    },
    {
      "acc": 0.66861,
      "epoch": 1.0100202942668697,
      "grad_norm": 6.1875,
      "learning_rate": 5.330285508482944e-06,
      "loss": 1.58127804,
      "memory(GiB)": 117.38,
      "step": 39815,
      "train_speed(iter/s)": 1.638453
    },
    {
      "acc": 0.6593791,
      "epoch": 1.0101471334348047,
      "grad_norm": 9.125,
      "learning_rate": 5.3292391679047905e-06,
      "loss": 1.58990421,
      "memory(GiB)": 117.38,
      "step": 39820,
      "train_speed(iter/s)": 1.638476
    },
    {
      "acc": 0.65348225,
      "epoch": 1.0102739726027397,
      "grad_norm": 5.09375,
      "learning_rate": 5.328192812845228e-06,
      "loss": 1.64288521,
      "memory(GiB)": 117.38,
      "step": 39825,
      "train_speed(iter/s)": 1.638498
    },
    {
      "acc": 0.65901661,
      "epoch": 1.0104008117706749,
      "grad_norm": 6.5625,
      "learning_rate": 5.3271464433502805e-06,
      "loss": 1.57470474,
      "memory(GiB)": 117.38,
      "step": 39830,
      "train_speed(iter/s)": 1.63852
    },
    {
      "acc": 0.65531597,
      "epoch": 1.0105276509386099,
      "grad_norm": 6.75,
      "learning_rate": 5.3261000594659715e-06,
      "loss": 1.65476475,
      "memory(GiB)": 117.38,
      "step": 39835,
      "train_speed(iter/s)": 1.638542
    },
    {
      "acc": 0.63626199,
      "epoch": 1.0106544901065448,
      "grad_norm": 5.59375,
      "learning_rate": 5.3250536612383275e-06,
      "loss": 1.70565567,
      "memory(GiB)": 117.38,
      "step": 39840,
      "train_speed(iter/s)": 1.638563
    },
    {
      "acc": 0.67136898,
      "epoch": 1.01078132927448,
      "grad_norm": 4.625,
      "learning_rate": 5.32400724871337e-06,
      "loss": 1.52749996,
      "memory(GiB)": 117.38,
      "step": 39845,
      "train_speed(iter/s)": 1.638584
    },
    {
      "acc": 0.66342649,
      "epoch": 1.010908168442415,
      "grad_norm": 5.375,
      "learning_rate": 5.322960821937129e-06,
      "loss": 1.57162724,
      "memory(GiB)": 117.38,
      "step": 39850,
      "train_speed(iter/s)": 1.638608
    },
    {
      "acc": 0.64148188,
      "epoch": 1.01103500761035,
      "grad_norm": 6.40625,
      "learning_rate": 5.321914380955628e-06,
      "loss": 1.65033703,
      "memory(GiB)": 117.38,
      "step": 39855,
      "train_speed(iter/s)": 1.638631
    },
    {
      "acc": 0.65975623,
      "epoch": 1.0111618467782852,
      "grad_norm": 5.75,
      "learning_rate": 5.320867925814896e-06,
      "loss": 1.60131836,
      "memory(GiB)": 117.38,
      "step": 39860,
      "train_speed(iter/s)": 1.638652
    },
    {
      "acc": 0.65837464,
      "epoch": 1.0112886859462202,
      "grad_norm": 5.5625,
      "learning_rate": 5.31982145656096e-06,
      "loss": 1.6029356,
      "memory(GiB)": 117.38,
      "step": 39865,
      "train_speed(iter/s)": 1.638673
    },
    {
      "acc": 0.65174217,
      "epoch": 1.0114155251141552,
      "grad_norm": 5.25,
      "learning_rate": 5.318774973239849e-06,
      "loss": 1.55748167,
      "memory(GiB)": 117.38,
      "step": 39870,
      "train_speed(iter/s)": 1.638694
    },
    {
      "acc": 0.65062418,
      "epoch": 1.0115423642820902,
      "grad_norm": 5.5,
      "learning_rate": 5.31772847589759e-06,
      "loss": 1.62412872,
      "memory(GiB)": 117.38,
      "step": 39875,
      "train_speed(iter/s)": 1.638716
    },
    {
      "acc": 0.66752129,
      "epoch": 1.0116692034500254,
      "grad_norm": 6.25,
      "learning_rate": 5.316681964580215e-06,
      "loss": 1.56336918,
      "memory(GiB)": 117.38,
      "step": 39880,
      "train_speed(iter/s)": 1.638739
    },
    {
      "acc": 0.66314564,
      "epoch": 1.0117960426179604,
      "grad_norm": 6.0625,
      "learning_rate": 5.315635439333753e-06,
      "loss": 1.59444866,
      "memory(GiB)": 117.38,
      "step": 39885,
      "train_speed(iter/s)": 1.63876
    },
    {
      "acc": 0.64957018,
      "epoch": 1.0119228817858954,
      "grad_norm": 5.65625,
      "learning_rate": 5.314588900204235e-06,
      "loss": 1.61485176,
      "memory(GiB)": 117.38,
      "step": 39890,
      "train_speed(iter/s)": 1.638781
    },
    {
      "acc": 0.65460482,
      "epoch": 1.0120497209538306,
      "grad_norm": 6.46875,
      "learning_rate": 5.313542347237692e-06,
      "loss": 1.65141449,
      "memory(GiB)": 117.38,
      "step": 39895,
      "train_speed(iter/s)": 1.638803
    },
    {
      "acc": 0.63795977,
      "epoch": 1.0121765601217656,
      "grad_norm": 6.59375,
      "learning_rate": 5.312495780480159e-06,
      "loss": 1.63843136,
      "memory(GiB)": 117.38,
      "step": 39900,
      "train_speed(iter/s)": 1.638825
    },
    {
      "acc": 0.65736651,
      "epoch": 1.0123033992897006,
      "grad_norm": 5.40625,
      "learning_rate": 5.311449199977664e-06,
      "loss": 1.59072075,
      "memory(GiB)": 117.38,
      "step": 39905,
      "train_speed(iter/s)": 1.638846
    },
    {
      "acc": 0.66679993,
      "epoch": 1.0124302384576358,
      "grad_norm": 6.4375,
      "learning_rate": 5.310402605776245e-06,
      "loss": 1.56476583,
      "memory(GiB)": 117.38,
      "step": 39910,
      "train_speed(iter/s)": 1.638867
    },
    {
      "acc": 0.66197143,
      "epoch": 1.0125570776255708,
      "grad_norm": 5.25,
      "learning_rate": 5.309355997921931e-06,
      "loss": 1.57712107,
      "memory(GiB)": 117.38,
      "step": 39915,
      "train_speed(iter/s)": 1.638888
    },
    {
      "acc": 0.67339616,
      "epoch": 1.0126839167935058,
      "grad_norm": 5.59375,
      "learning_rate": 5.308309376460761e-06,
      "loss": 1.59358978,
      "memory(GiB)": 117.38,
      "step": 39920,
      "train_speed(iter/s)": 1.638909
    },
    {
      "acc": 0.66828194,
      "epoch": 1.012810755961441,
      "grad_norm": 4.9375,
      "learning_rate": 5.307262741438767e-06,
      "loss": 1.53220406,
      "memory(GiB)": 117.38,
      "step": 39925,
      "train_speed(iter/s)": 1.63893
    },
    {
      "acc": 0.65965991,
      "epoch": 1.012937595129376,
      "grad_norm": 6.03125,
      "learning_rate": 5.3062160929019855e-06,
      "loss": 1.6212965,
      "memory(GiB)": 117.38,
      "step": 39930,
      "train_speed(iter/s)": 1.638953
    },
    {
      "acc": 0.67123747,
      "epoch": 1.013064434297311,
      "grad_norm": 5.25,
      "learning_rate": 5.305169430896454e-06,
      "loss": 1.56313429,
      "memory(GiB)": 117.38,
      "step": 39935,
      "train_speed(iter/s)": 1.638975
    },
    {
      "acc": 0.66302409,
      "epoch": 1.013191273465246,
      "grad_norm": 4.875,
      "learning_rate": 5.304122755468209e-06,
      "loss": 1.55104485,
      "memory(GiB)": 117.38,
      "step": 39940,
      "train_speed(iter/s)": 1.638996
    },
    {
      "acc": 0.65456266,
      "epoch": 1.0133181126331812,
      "grad_norm": 6.53125,
      "learning_rate": 5.303076066663286e-06,
      "loss": 1.5185524,
      "memory(GiB)": 117.38,
      "step": 39945,
      "train_speed(iter/s)": 1.639018
    },
    {
      "acc": 0.64203253,
      "epoch": 1.0134449518011162,
      "grad_norm": 5.5,
      "learning_rate": 5.302029364527726e-06,
      "loss": 1.644701,
      "memory(GiB)": 117.38,
      "step": 39950,
      "train_speed(iter/s)": 1.639039
    },
    {
      "acc": 0.65256262,
      "epoch": 1.0135717909690511,
      "grad_norm": 6.46875,
      "learning_rate": 5.3009826491075645e-06,
      "loss": 1.6785635,
      "memory(GiB)": 117.38,
      "step": 39955,
      "train_speed(iter/s)": 1.63906
    },
    {
      "acc": 0.6583322,
      "epoch": 1.0136986301369864,
      "grad_norm": 8.125,
      "learning_rate": 5.299935920448843e-06,
      "loss": 1.53910131,
      "memory(GiB)": 117.38,
      "step": 39960,
      "train_speed(iter/s)": 1.639079
    },
    {
      "acc": 0.66376405,
      "epoch": 1.0138254693049213,
      "grad_norm": 5.3125,
      "learning_rate": 5.298889178597599e-06,
      "loss": 1.56935625,
      "memory(GiB)": 117.38,
      "step": 39965,
      "train_speed(iter/s)": 1.639099
    },
    {
      "acc": 0.65047426,
      "epoch": 1.0139523084728563,
      "grad_norm": 6.71875,
      "learning_rate": 5.297842423599877e-06,
      "loss": 1.59526472,
      "memory(GiB)": 117.38,
      "step": 39970,
      "train_speed(iter/s)": 1.63912
    },
    {
      "acc": 0.66500549,
      "epoch": 1.0140791476407915,
      "grad_norm": 6.65625,
      "learning_rate": 5.296795655501714e-06,
      "loss": 1.62907639,
      "memory(GiB)": 117.38,
      "step": 39975,
      "train_speed(iter/s)": 1.639142
    },
    {
      "acc": 0.65319786,
      "epoch": 1.0142059868087265,
      "grad_norm": 5.6875,
      "learning_rate": 5.295748874349155e-06,
      "loss": 1.63263054,
      "memory(GiB)": 117.38,
      "step": 39980,
      "train_speed(iter/s)": 1.639162
    },
    {
      "acc": 0.65103827,
      "epoch": 1.0143328259766615,
      "grad_norm": 5.84375,
      "learning_rate": 5.294702080188236e-06,
      "loss": 1.56789045,
      "memory(GiB)": 117.38,
      "step": 39985,
      "train_speed(iter/s)": 1.639182
    },
    {
      "acc": 0.64973259,
      "epoch": 1.0144596651445967,
      "grad_norm": 5.4375,
      "learning_rate": 5.293655273065008e-06,
      "loss": 1.61104507,
      "memory(GiB)": 117.38,
      "step": 39990,
      "train_speed(iter/s)": 1.639203
    },
    {
      "acc": 0.66724234,
      "epoch": 1.0145865043125317,
      "grad_norm": 5.28125,
      "learning_rate": 5.2926084530255076e-06,
      "loss": 1.59788103,
      "memory(GiB)": 117.38,
      "step": 39995,
      "train_speed(iter/s)": 1.639225
    },
    {
      "acc": 0.6543807,
      "epoch": 1.0147133434804667,
      "grad_norm": 6.15625,
      "learning_rate": 5.291561620115781e-06,
      "loss": 1.59785461,
      "memory(GiB)": 117.38,
      "step": 40000,
      "train_speed(iter/s)": 1.639248
    },
    {
      "epoch": 1.0147133434804667,
      "eval_acc": 0.6462142142643837,
      "eval_loss": 1.5740785598754883,
      "eval_runtime": 57.9965,
      "eval_samples_per_second": 109.834,
      "eval_steps_per_second": 27.467,
      "step": 40000
    },
    {
      "acc": 0.65649223,
      "epoch": 1.014840182648402,
      "grad_norm": 6.40625,
      "learning_rate": 5.290514774381874e-06,
      "loss": 1.6269558,
      "memory(GiB)": 117.38,
      "step": 40005,
      "train_speed(iter/s)": 1.635093
    },
    {
      "acc": 0.65570307,
      "epoch": 1.014967021816337,
      "grad_norm": 6.15625,
      "learning_rate": 5.289467915869829e-06,
      "loss": 1.60519314,
      "memory(GiB)": 117.38,
      "step": 40010,
      "train_speed(iter/s)": 1.635113
    },
    {
      "acc": 0.66156893,
      "epoch": 1.015093860984272,
      "grad_norm": 6.5,
      "learning_rate": 5.288421044625694e-06,
      "loss": 1.59348755,
      "memory(GiB)": 117.38,
      "step": 40015,
      "train_speed(iter/s)": 1.635135
    },
    {
      "acc": 0.65349388,
      "epoch": 1.0152207001522071,
      "grad_norm": 5.3125,
      "learning_rate": 5.287374160695513e-06,
      "loss": 1.61469593,
      "memory(GiB)": 117.38,
      "step": 40020,
      "train_speed(iter/s)": 1.635157
    },
    {
      "acc": 0.64751792,
      "epoch": 1.015347539320142,
      "grad_norm": 5.46875,
      "learning_rate": 5.286327264125332e-06,
      "loss": 1.67454109,
      "memory(GiB)": 117.38,
      "step": 40025,
      "train_speed(iter/s)": 1.63518
    },
    {
      "acc": 0.64518757,
      "epoch": 1.015474378488077,
      "grad_norm": 6.28125,
      "learning_rate": 5.285280354961202e-06,
      "loss": 1.67606239,
      "memory(GiB)": 117.38,
      "step": 40030,
      "train_speed(iter/s)": 1.635203
    },
    {
      "acc": 0.65832996,
      "epoch": 1.015601217656012,
      "grad_norm": 5.59375,
      "learning_rate": 5.284233433249167e-06,
      "loss": 1.52517776,
      "memory(GiB)": 117.38,
      "step": 40035,
      "train_speed(iter/s)": 1.635225
    },
    {
      "acc": 0.65377994,
      "epoch": 1.0157280568239473,
      "grad_norm": 6.71875,
      "learning_rate": 5.283186499035276e-06,
      "loss": 1.5628643,
      "memory(GiB)": 117.38,
      "step": 40040,
      "train_speed(iter/s)": 1.635247
    },
    {
      "acc": 0.66414504,
      "epoch": 1.0158548959918823,
      "grad_norm": 5.5625,
      "learning_rate": 5.2821395523655795e-06,
      "loss": 1.53937492,
      "memory(GiB)": 117.38,
      "step": 40045,
      "train_speed(iter/s)": 1.635268
    },
    {
      "acc": 0.66452026,
      "epoch": 1.0159817351598173,
      "grad_norm": 5.53125,
      "learning_rate": 5.281092593286127e-06,
      "loss": 1.55305614,
      "memory(GiB)": 117.38,
      "step": 40050,
      "train_speed(iter/s)": 1.635287
    },
    {
      "acc": 0.6489089,
      "epoch": 1.0161085743277525,
      "grad_norm": 5.625,
      "learning_rate": 5.280045621842964e-06,
      "loss": 1.58909225,
      "memory(GiB)": 117.38,
      "step": 40055,
      "train_speed(iter/s)": 1.63531
    },
    {
      "acc": 0.63675885,
      "epoch": 1.0162354134956875,
      "grad_norm": 5.03125,
      "learning_rate": 5.278998638082148e-06,
      "loss": 1.62103424,
      "memory(GiB)": 117.38,
      "step": 40060,
      "train_speed(iter/s)": 1.635331
    },
    {
      "acc": 0.65525999,
      "epoch": 1.0163622526636225,
      "grad_norm": 5.03125,
      "learning_rate": 5.277951642049722e-06,
      "loss": 1.60936203,
      "memory(GiB)": 117.38,
      "step": 40065,
      "train_speed(iter/s)": 1.635352
    },
    {
      "acc": 0.66408048,
      "epoch": 1.0164890918315577,
      "grad_norm": 4.71875,
      "learning_rate": 5.276904633791745e-06,
      "loss": 1.56101933,
      "memory(GiB)": 117.38,
      "step": 40070,
      "train_speed(iter/s)": 1.635375
    },
    {
      "acc": 0.65598583,
      "epoch": 1.0166159309994927,
      "grad_norm": 5.875,
      "learning_rate": 5.275857613354265e-06,
      "loss": 1.55445404,
      "memory(GiB)": 117.38,
      "step": 40075,
      "train_speed(iter/s)": 1.635396
    },
    {
      "acc": 0.66829014,
      "epoch": 1.0167427701674276,
      "grad_norm": 4.90625,
      "learning_rate": 5.274810580783335e-06,
      "loss": 1.58074207,
      "memory(GiB)": 117.38,
      "step": 40080,
      "train_speed(iter/s)": 1.635419
    },
    {
      "acc": 0.67111392,
      "epoch": 1.0168696093353629,
      "grad_norm": 5.71875,
      "learning_rate": 5.2737635361250094e-06,
      "loss": 1.56327419,
      "memory(GiB)": 117.38,
      "step": 40085,
      "train_speed(iter/s)": 1.63544
    },
    {
      "acc": 0.65572786,
      "epoch": 1.0169964485032978,
      "grad_norm": 5.9375,
      "learning_rate": 5.2727164794253415e-06,
      "loss": 1.60113869,
      "memory(GiB)": 117.38,
      "step": 40090,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.64674392,
      "epoch": 1.0171232876712328,
      "grad_norm": 5.84375,
      "learning_rate": 5.271669410730384e-06,
      "loss": 1.60128365,
      "memory(GiB)": 117.38,
      "step": 40095,
      "train_speed(iter/s)": 1.635484
    },
    {
      "acc": 0.67051244,
      "epoch": 1.0172501268391678,
      "grad_norm": 5.78125,
      "learning_rate": 5.270622330086194e-06,
      "loss": 1.59918051,
      "memory(GiB)": 117.38,
      "step": 40100,
      "train_speed(iter/s)": 1.635507
    },
    {
      "acc": 0.63950601,
      "epoch": 1.017376966007103,
      "grad_norm": 5.15625,
      "learning_rate": 5.269575237538827e-06,
      "loss": 1.59302998,
      "memory(GiB)": 117.38,
      "step": 40105,
      "train_speed(iter/s)": 1.635529
    },
    {
      "acc": 0.64420271,
      "epoch": 1.017503805175038,
      "grad_norm": 6.0,
      "learning_rate": 5.268528133134335e-06,
      "loss": 1.65467072,
      "memory(GiB)": 117.38,
      "step": 40110,
      "train_speed(iter/s)": 1.635549
    },
    {
      "acc": 0.6618732,
      "epoch": 1.017630644342973,
      "grad_norm": 5.625,
      "learning_rate": 5.267481016918776e-06,
      "loss": 1.55507221,
      "memory(GiB)": 117.38,
      "step": 40115,
      "train_speed(iter/s)": 1.63557
    },
    {
      "acc": 0.65956802,
      "epoch": 1.0177574835109082,
      "grad_norm": 6.28125,
      "learning_rate": 5.266433888938212e-06,
      "loss": 1.60486946,
      "memory(GiB)": 117.38,
      "step": 40120,
      "train_speed(iter/s)": 1.635592
    },
    {
      "acc": 0.66995301,
      "epoch": 1.0178843226788432,
      "grad_norm": 5.28125,
      "learning_rate": 5.265386749238691e-06,
      "loss": 1.54338589,
      "memory(GiB)": 117.38,
      "step": 40125,
      "train_speed(iter/s)": 1.635614
    },
    {
      "acc": 0.64883652,
      "epoch": 1.0180111618467782,
      "grad_norm": 6.375,
      "learning_rate": 5.26433959786628e-06,
      "loss": 1.62892799,
      "memory(GiB)": 117.38,
      "step": 40130,
      "train_speed(iter/s)": 1.635635
    },
    {
      "acc": 0.64977026,
      "epoch": 1.0181380010147134,
      "grad_norm": 5.9375,
      "learning_rate": 5.263292434867031e-06,
      "loss": 1.59526377,
      "memory(GiB)": 117.38,
      "step": 40135,
      "train_speed(iter/s)": 1.635657
    },
    {
      "acc": 0.67420692,
      "epoch": 1.0182648401826484,
      "grad_norm": 5.125,
      "learning_rate": 5.262245260287006e-06,
      "loss": 1.48768463,
      "memory(GiB)": 117.38,
      "step": 40140,
      "train_speed(iter/s)": 1.635678
    },
    {
      "acc": 0.64456005,
      "epoch": 1.0183916793505834,
      "grad_norm": 8.25,
      "learning_rate": 5.261198074172262e-06,
      "loss": 1.69737701,
      "memory(GiB)": 117.38,
      "step": 40145,
      "train_speed(iter/s)": 1.635702
    },
    {
      "acc": 0.65967212,
      "epoch": 1.0185185185185186,
      "grad_norm": 5.625,
      "learning_rate": 5.260150876568862e-06,
      "loss": 1.61888218,
      "memory(GiB)": 117.38,
      "step": 40150,
      "train_speed(iter/s)": 1.635726
    },
    {
      "acc": 0.65892339,
      "epoch": 1.0186453576864536,
      "grad_norm": 4.71875,
      "learning_rate": 5.259103667522866e-06,
      "loss": 1.5869688,
      "memory(GiB)": 117.38,
      "step": 40155,
      "train_speed(iter/s)": 1.635748
    },
    {
      "acc": 0.65756025,
      "epoch": 1.0187721968543886,
      "grad_norm": 5.09375,
      "learning_rate": 5.258056447080333e-06,
      "loss": 1.64498558,
      "memory(GiB)": 117.38,
      "step": 40160,
      "train_speed(iter/s)": 1.635771
    },
    {
      "acc": 0.65321121,
      "epoch": 1.0188990360223238,
      "grad_norm": 5.53125,
      "learning_rate": 5.257009215287325e-06,
      "loss": 1.65939293,
      "memory(GiB)": 117.38,
      "step": 40165,
      "train_speed(iter/s)": 1.635793
    },
    {
      "acc": 0.65982928,
      "epoch": 1.0190258751902588,
      "grad_norm": 5.375,
      "learning_rate": 5.255961972189905e-06,
      "loss": 1.6098177,
      "memory(GiB)": 117.38,
      "step": 40170,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.64726415,
      "epoch": 1.0191527143581938,
      "grad_norm": 5.75,
      "learning_rate": 5.254914717834133e-06,
      "loss": 1.62165527,
      "memory(GiB)": 117.38,
      "step": 40175,
      "train_speed(iter/s)": 1.635838
    },
    {
      "acc": 0.65048866,
      "epoch": 1.019279553526129,
      "grad_norm": 5.03125,
      "learning_rate": 5.253867452266075e-06,
      "loss": 1.58091764,
      "memory(GiB)": 117.38,
      "step": 40180,
      "train_speed(iter/s)": 1.635861
    },
    {
      "acc": 0.67215166,
      "epoch": 1.019406392694064,
      "grad_norm": 9.0625,
      "learning_rate": 5.252820175531792e-06,
      "loss": 1.57077713,
      "memory(GiB)": 117.38,
      "step": 40185,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.64890275,
      "epoch": 1.019533231861999,
      "grad_norm": 7.46875,
      "learning_rate": 5.25177288767735e-06,
      "loss": 1.62759724,
      "memory(GiB)": 117.38,
      "step": 40190,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.66316366,
      "epoch": 1.019660071029934,
      "grad_norm": 5.96875,
      "learning_rate": 5.250725588748811e-06,
      "loss": 1.58214417,
      "memory(GiB)": 117.38,
      "step": 40195,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.6483737,
      "epoch": 1.0197869101978692,
      "grad_norm": 5.53125,
      "learning_rate": 5.249678278792243e-06,
      "loss": 1.68145676,
      "memory(GiB)": 117.38,
      "step": 40200,
      "train_speed(iter/s)": 1.635951
    },
    {
      "acc": 0.67354927,
      "epoch": 1.0199137493658041,
      "grad_norm": 5.625,
      "learning_rate": 5.248630957853708e-06,
      "loss": 1.54078636,
      "memory(GiB)": 117.38,
      "step": 40205,
      "train_speed(iter/s)": 1.635975
    },
    {
      "acc": 0.65906215,
      "epoch": 1.0200405885337391,
      "grad_norm": 5.46875,
      "learning_rate": 5.247583625979276e-06,
      "loss": 1.54860744,
      "memory(GiB)": 117.38,
      "step": 40210,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.64174495,
      "epoch": 1.0201674277016743,
      "grad_norm": 6.71875,
      "learning_rate": 5.246536283215007e-06,
      "loss": 1.58928385,
      "memory(GiB)": 117.38,
      "step": 40215,
      "train_speed(iter/s)": 1.636018
    },
    {
      "acc": 0.65218077,
      "epoch": 1.0202942668696093,
      "grad_norm": 5.5,
      "learning_rate": 5.245488929606974e-06,
      "loss": 1.62645531,
      "memory(GiB)": 117.38,
      "step": 40220,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.65695028,
      "epoch": 1.0204211060375443,
      "grad_norm": 4.75,
      "learning_rate": 5.244441565201241e-06,
      "loss": 1.61856422,
      "memory(GiB)": 117.38,
      "step": 40225,
      "train_speed(iter/s)": 1.63606
    },
    {
      "acc": 0.65484724,
      "epoch": 1.0205479452054795,
      "grad_norm": 5.8125,
      "learning_rate": 5.243394190043877e-06,
      "loss": 1.51860943,
      "memory(GiB)": 117.38,
      "step": 40230,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.66069336,
      "epoch": 1.0206747843734145,
      "grad_norm": 5.59375,
      "learning_rate": 5.242346804180949e-06,
      "loss": 1.58330793,
      "memory(GiB)": 117.38,
      "step": 40235,
      "train_speed(iter/s)": 1.636104
    },
    {
      "acc": 0.65965705,
      "epoch": 1.0208016235413495,
      "grad_norm": 5.375,
      "learning_rate": 5.241299407658528e-06,
      "loss": 1.61549454,
      "memory(GiB)": 117.38,
      "step": 40240,
      "train_speed(iter/s)": 1.636126
    },
    {
      "acc": 0.65619869,
      "epoch": 1.0209284627092847,
      "grad_norm": 5.28125,
      "learning_rate": 5.240252000522681e-06,
      "loss": 1.56270351,
      "memory(GiB)": 117.38,
      "step": 40245,
      "train_speed(iter/s)": 1.636148
    },
    {
      "acc": 0.64507804,
      "epoch": 1.0210553018772197,
      "grad_norm": 6.375,
      "learning_rate": 5.239204582819479e-06,
      "loss": 1.66113567,
      "memory(GiB)": 117.38,
      "step": 40250,
      "train_speed(iter/s)": 1.63617
    },
    {
      "acc": 0.65118189,
      "epoch": 1.0211821410451547,
      "grad_norm": 7.71875,
      "learning_rate": 5.238157154594989e-06,
      "loss": 1.60753536,
      "memory(GiB)": 117.38,
      "step": 40255,
      "train_speed(iter/s)": 1.636193
    },
    {
      "acc": 0.67268829,
      "epoch": 1.0213089802130897,
      "grad_norm": 6.34375,
      "learning_rate": 5.237109715895287e-06,
      "loss": 1.47926054,
      "memory(GiB)": 117.38,
      "step": 40260,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.66148472,
      "epoch": 1.021435819381025,
      "grad_norm": 5.875,
      "learning_rate": 5.2360622667664385e-06,
      "loss": 1.57624626,
      "memory(GiB)": 117.38,
      "step": 40265,
      "train_speed(iter/s)": 1.636233
    },
    {
      "acc": 0.66665912,
      "epoch": 1.02156265854896,
      "grad_norm": 6.8125,
      "learning_rate": 5.235014807254521e-06,
      "loss": 1.56256876,
      "memory(GiB)": 117.38,
      "step": 40270,
      "train_speed(iter/s)": 1.636254
    },
    {
      "acc": 0.64318552,
      "epoch": 1.0216894977168949,
      "grad_norm": 7.40625,
      "learning_rate": 5.233967337405599e-06,
      "loss": 1.61275978,
      "memory(GiB)": 117.38,
      "step": 40275,
      "train_speed(iter/s)": 1.636277
    },
    {
      "acc": 0.66389303,
      "epoch": 1.02181633688483,
      "grad_norm": 5.0625,
      "learning_rate": 5.232919857265752e-06,
      "loss": 1.61152496,
      "memory(GiB)": 117.38,
      "step": 40280,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.65124974,
      "epoch": 1.021943176052765,
      "grad_norm": 5.3125,
      "learning_rate": 5.231872366881048e-06,
      "loss": 1.61572285,
      "memory(GiB)": 117.38,
      "step": 40285,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.66670675,
      "epoch": 1.0220700152207,
      "grad_norm": 5.03125,
      "learning_rate": 5.230824866297563e-06,
      "loss": 1.5959487,
      "memory(GiB)": 117.38,
      "step": 40290,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65785418,
      "epoch": 1.0221968543886353,
      "grad_norm": 5.78125,
      "learning_rate": 5.229777355561368e-06,
      "loss": 1.58440151,
      "memory(GiB)": 117.38,
      "step": 40295,
      "train_speed(iter/s)": 1.636363
    },
    {
      "acc": 0.65784349,
      "epoch": 1.0223236935565703,
      "grad_norm": 7.0,
      "learning_rate": 5.2287298347185415e-06,
      "loss": 1.60356426,
      "memory(GiB)": 117.38,
      "step": 40300,
      "train_speed(iter/s)": 1.636384
    },
    {
      "acc": 0.63558426,
      "epoch": 1.0224505327245053,
      "grad_norm": 5.8125,
      "learning_rate": 5.227682303815155e-06,
      "loss": 1.63212929,
      "memory(GiB)": 117.38,
      "step": 40305,
      "train_speed(iter/s)": 1.636405
    },
    {
      "acc": 0.64989662,
      "epoch": 1.0225773718924405,
      "grad_norm": 5.3125,
      "learning_rate": 5.226634762897284e-06,
      "loss": 1.63894234,
      "memory(GiB)": 117.38,
      "step": 40310,
      "train_speed(iter/s)": 1.636428
    },
    {
      "acc": 0.65535831,
      "epoch": 1.0227042110603755,
      "grad_norm": 5.9375,
      "learning_rate": 5.225587212011004e-06,
      "loss": 1.54392014,
      "memory(GiB)": 117.38,
      "step": 40315,
      "train_speed(iter/s)": 1.63645
    },
    {
      "acc": 0.65255675,
      "epoch": 1.0228310502283104,
      "grad_norm": 5.59375,
      "learning_rate": 5.224539651202391e-06,
      "loss": 1.56442575,
      "memory(GiB)": 117.38,
      "step": 40320,
      "train_speed(iter/s)": 1.636473
    },
    {
      "acc": 0.65225158,
      "epoch": 1.0229578893962457,
      "grad_norm": 5.9375,
      "learning_rate": 5.223492080517523e-06,
      "loss": 1.61982002,
      "memory(GiB)": 117.38,
      "step": 40325,
      "train_speed(iter/s)": 1.636494
    },
    {
      "acc": 0.65033979,
      "epoch": 1.0230847285641806,
      "grad_norm": 6.90625,
      "learning_rate": 5.2224445000024744e-06,
      "loss": 1.65942116,
      "memory(GiB)": 117.38,
      "step": 40330,
      "train_speed(iter/s)": 1.636516
    },
    {
      "acc": 0.65091619,
      "epoch": 1.0232115677321156,
      "grad_norm": 5.78125,
      "learning_rate": 5.221396909703322e-06,
      "loss": 1.58307533,
      "memory(GiB)": 117.38,
      "step": 40335,
      "train_speed(iter/s)": 1.636539
    },
    {
      "acc": 0.66293764,
      "epoch": 1.0233384069000508,
      "grad_norm": 5.90625,
      "learning_rate": 5.220349309666148e-06,
      "loss": 1.53305664,
      "memory(GiB)": 117.38,
      "step": 40340,
      "train_speed(iter/s)": 1.636562
    },
    {
      "acc": 0.65347309,
      "epoch": 1.0234652460679858,
      "grad_norm": 6.78125,
      "learning_rate": 5.2193016999370265e-06,
      "loss": 1.61133499,
      "memory(GiB)": 117.38,
      "step": 40345,
      "train_speed(iter/s)": 1.636584
    },
    {
      "acc": 0.65706906,
      "epoch": 1.0235920852359208,
      "grad_norm": 5.28125,
      "learning_rate": 5.218254080562038e-06,
      "loss": 1.55974197,
      "memory(GiB)": 117.38,
      "step": 40350,
      "train_speed(iter/s)": 1.636607
    },
    {
      "acc": 0.67610836,
      "epoch": 1.0237189244038558,
      "grad_norm": 7.5,
      "learning_rate": 5.2172064515872585e-06,
      "loss": 1.49498997,
      "memory(GiB)": 117.38,
      "step": 40355,
      "train_speed(iter/s)": 1.63663
    },
    {
      "acc": 0.66202326,
      "epoch": 1.023845763571791,
      "grad_norm": 6.65625,
      "learning_rate": 5.21615881305877e-06,
      "loss": 1.62622356,
      "memory(GiB)": 117.38,
      "step": 40360,
      "train_speed(iter/s)": 1.636652
    },
    {
      "acc": 0.65148506,
      "epoch": 1.023972602739726,
      "grad_norm": 5.71875,
      "learning_rate": 5.215111165022653e-06,
      "loss": 1.57873535,
      "memory(GiB)": 117.38,
      "step": 40365,
      "train_speed(iter/s)": 1.636674
    },
    {
      "acc": 0.67259207,
      "epoch": 1.024099441907661,
      "grad_norm": 5.375,
      "learning_rate": 5.2140635075249856e-06,
      "loss": 1.5061265,
      "memory(GiB)": 117.38,
      "step": 40370,
      "train_speed(iter/s)": 1.636695
    },
    {
      "acc": 0.6600646,
      "epoch": 1.0242262810755962,
      "grad_norm": 5.6875,
      "learning_rate": 5.213015840611851e-06,
      "loss": 1.5737422,
      "memory(GiB)": 117.38,
      "step": 40375,
      "train_speed(iter/s)": 1.636717
    },
    {
      "acc": 0.64788723,
      "epoch": 1.0243531202435312,
      "grad_norm": 6.75,
      "learning_rate": 5.211968164329328e-06,
      "loss": 1.64896126,
      "memory(GiB)": 117.38,
      "step": 40380,
      "train_speed(iter/s)": 1.636738
    },
    {
      "acc": 0.66225376,
      "epoch": 1.0244799594114662,
      "grad_norm": 5.59375,
      "learning_rate": 5.210920478723497e-06,
      "loss": 1.61479454,
      "memory(GiB)": 117.38,
      "step": 40385,
      "train_speed(iter/s)": 1.63676
    },
    {
      "acc": 0.64224291,
      "epoch": 1.0246067985794014,
      "grad_norm": 5.59375,
      "learning_rate": 5.209872783840443e-06,
      "loss": 1.6535284,
      "memory(GiB)": 117.38,
      "step": 40390,
      "train_speed(iter/s)": 1.636781
    },
    {
      "acc": 0.65905256,
      "epoch": 1.0247336377473364,
      "grad_norm": 5.03125,
      "learning_rate": 5.208825079726248e-06,
      "loss": 1.60687561,
      "memory(GiB)": 117.38,
      "step": 40395,
      "train_speed(iter/s)": 1.636802
    },
    {
      "acc": 0.66541691,
      "epoch": 1.0248604769152714,
      "grad_norm": 5.71875,
      "learning_rate": 5.207777366426992e-06,
      "loss": 1.588976,
      "memory(GiB)": 117.38,
      "step": 40400,
      "train_speed(iter/s)": 1.636822
    },
    {
      "acc": 0.65608292,
      "epoch": 1.0249873160832066,
      "grad_norm": 5.5,
      "learning_rate": 5.206729643988759e-06,
      "loss": 1.58177643,
      "memory(GiB)": 117.38,
      "step": 40405,
      "train_speed(iter/s)": 1.636845
    },
    {
      "acc": 0.66607685,
      "epoch": 1.0251141552511416,
      "grad_norm": 5.875,
      "learning_rate": 5.205681912457635e-06,
      "loss": 1.57152004,
      "memory(GiB)": 117.38,
      "step": 40410,
      "train_speed(iter/s)": 1.636866
    },
    {
      "acc": 0.63905668,
      "epoch": 1.0252409944190766,
      "grad_norm": 7.34375,
      "learning_rate": 5.204634171879701e-06,
      "loss": 1.69107399,
      "memory(GiB)": 117.38,
      "step": 40415,
      "train_speed(iter/s)": 1.636886
    },
    {
      "acc": 0.65537505,
      "epoch": 1.0253678335870116,
      "grad_norm": 4.71875,
      "learning_rate": 5.2035864223010445e-06,
      "loss": 1.51729584,
      "memory(GiB)": 117.38,
      "step": 40420,
      "train_speed(iter/s)": 1.636907
    },
    {
      "acc": 0.65592699,
      "epoch": 1.0254946727549468,
      "grad_norm": 6.90625,
      "learning_rate": 5.202538663767746e-06,
      "loss": 1.6054409,
      "memory(GiB)": 117.38,
      "step": 40425,
      "train_speed(iter/s)": 1.636927
    },
    {
      "acc": 0.65538116,
      "epoch": 1.0256215119228818,
      "grad_norm": 5.96875,
      "learning_rate": 5.201490896325895e-06,
      "loss": 1.63512344,
      "memory(GiB)": 117.38,
      "step": 40430,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.66591787,
      "epoch": 1.0257483510908167,
      "grad_norm": 5.78125,
      "learning_rate": 5.200443120021572e-06,
      "loss": 1.56082277,
      "memory(GiB)": 117.38,
      "step": 40435,
      "train_speed(iter/s)": 1.636971
    },
    {
      "acc": 0.66506791,
      "epoch": 1.025875190258752,
      "grad_norm": 5.8125,
      "learning_rate": 5.199395334900868e-06,
      "loss": 1.53434143,
      "memory(GiB)": 117.38,
      "step": 40440,
      "train_speed(iter/s)": 1.636993
    },
    {
      "acc": 0.65561771,
      "epoch": 1.026002029426687,
      "grad_norm": 6.53125,
      "learning_rate": 5.198347541009866e-06,
      "loss": 1.60469818,
      "memory(GiB)": 117.38,
      "step": 40445,
      "train_speed(iter/s)": 1.637015
    },
    {
      "acc": 0.65198641,
      "epoch": 1.026128868594622,
      "grad_norm": 6.125,
      "learning_rate": 5.197299738394654e-06,
      "loss": 1.64730339,
      "memory(GiB)": 117.38,
      "step": 40450,
      "train_speed(iter/s)": 1.637036
    },
    {
      "acc": 0.65640302,
      "epoch": 1.0262557077625571,
      "grad_norm": 5.25,
      "learning_rate": 5.196251927101318e-06,
      "loss": 1.61294098,
      "memory(GiB)": 117.38,
      "step": 40455,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.66402311,
      "epoch": 1.0263825469304921,
      "grad_norm": 5.96875,
      "learning_rate": 5.195204107175946e-06,
      "loss": 1.50397654,
      "memory(GiB)": 117.38,
      "step": 40460,
      "train_speed(iter/s)": 1.637079
    },
    {
      "acc": 0.66228719,
      "epoch": 1.0265093860984271,
      "grad_norm": 6.53125,
      "learning_rate": 5.194156278664627e-06,
      "loss": 1.62087059,
      "memory(GiB)": 117.38,
      "step": 40465,
      "train_speed(iter/s)": 1.637101
    },
    {
      "acc": 0.65959306,
      "epoch": 1.0266362252663623,
      "grad_norm": 4.96875,
      "learning_rate": 5.1931084416134466e-06,
      "loss": 1.57213478,
      "memory(GiB)": 117.38,
      "step": 40470,
      "train_speed(iter/s)": 1.637122
    },
    {
      "acc": 0.65139093,
      "epoch": 1.0267630644342973,
      "grad_norm": 4.8125,
      "learning_rate": 5.192060596068496e-06,
      "loss": 1.62109947,
      "memory(GiB)": 117.38,
      "step": 40475,
      "train_speed(iter/s)": 1.637144
    },
    {
      "acc": 0.65920916,
      "epoch": 1.0268899036022323,
      "grad_norm": 5.375,
      "learning_rate": 5.191012742075863e-06,
      "loss": 1.5731863,
      "memory(GiB)": 117.38,
      "step": 40480,
      "train_speed(iter/s)": 1.637165
    },
    {
      "acc": 0.66571569,
      "epoch": 1.0270167427701675,
      "grad_norm": 7.53125,
      "learning_rate": 5.189964879681635e-06,
      "loss": 1.59552307,
      "memory(GiB)": 117.38,
      "step": 40485,
      "train_speed(iter/s)": 1.637186
    },
    {
      "acc": 0.64958324,
      "epoch": 1.0271435819381025,
      "grad_norm": 6.0,
      "learning_rate": 5.188917008931905e-06,
      "loss": 1.62662163,
      "memory(GiB)": 117.38,
      "step": 40490,
      "train_speed(iter/s)": 1.637207
    },
    {
      "acc": 0.6562439,
      "epoch": 1.0272704211060375,
      "grad_norm": 7.21875,
      "learning_rate": 5.18786912987276e-06,
      "loss": 1.58037453,
      "memory(GiB)": 117.38,
      "step": 40495,
      "train_speed(iter/s)": 1.63723
    },
    {
      "acc": 0.66030588,
      "epoch": 1.0273972602739727,
      "grad_norm": 5.5,
      "learning_rate": 5.186821242550294e-06,
      "loss": 1.56659536,
      "memory(GiB)": 117.38,
      "step": 40500,
      "train_speed(iter/s)": 1.637251
    },
    {
      "acc": 0.64224744,
      "epoch": 1.0275240994419077,
      "grad_norm": 6.0625,
      "learning_rate": 5.185773347010594e-06,
      "loss": 1.64190331,
      "memory(GiB)": 117.38,
      "step": 40505,
      "train_speed(iter/s)": 1.637273
    },
    {
      "acc": 0.65694728,
      "epoch": 1.0276509386098427,
      "grad_norm": 5.53125,
      "learning_rate": 5.184725443299753e-06,
      "loss": 1.59824219,
      "memory(GiB)": 117.38,
      "step": 40510,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.66392894,
      "epoch": 1.0277777777777777,
      "grad_norm": 5.8125,
      "learning_rate": 5.183677531463863e-06,
      "loss": 1.56978006,
      "memory(GiB)": 117.38,
      "step": 40515,
      "train_speed(iter/s)": 1.637317
    },
    {
      "acc": 0.64516563,
      "epoch": 1.027904616945713,
      "grad_norm": 6.15625,
      "learning_rate": 5.182629611549015e-06,
      "loss": 1.65863686,
      "memory(GiB)": 117.38,
      "step": 40520,
      "train_speed(iter/s)": 1.637339
    },
    {
      "acc": 0.65516596,
      "epoch": 1.0280314561136479,
      "grad_norm": 6.0625,
      "learning_rate": 5.181581683601301e-06,
      "loss": 1.5598835,
      "memory(GiB)": 117.38,
      "step": 40525,
      "train_speed(iter/s)": 1.63736
    },
    {
      "acc": 0.65747061,
      "epoch": 1.0281582952815829,
      "grad_norm": 6.34375,
      "learning_rate": 5.1805337476668135e-06,
      "loss": 1.69331379,
      "memory(GiB)": 117.38,
      "step": 40530,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.66326451,
      "epoch": 1.028285134449518,
      "grad_norm": 4.375,
      "learning_rate": 5.179485803791646e-06,
      "loss": 1.62275791,
      "memory(GiB)": 117.38,
      "step": 40535,
      "train_speed(iter/s)": 1.637403
    },
    {
      "acc": 0.65492725,
      "epoch": 1.028411973617453,
      "grad_norm": 5.78125,
      "learning_rate": 5.178437852021892e-06,
      "loss": 1.67811279,
      "memory(GiB)": 117.38,
      "step": 40540,
      "train_speed(iter/s)": 1.637425
    },
    {
      "acc": 0.66354737,
      "epoch": 1.028538812785388,
      "grad_norm": 6.75,
      "learning_rate": 5.177389892403645e-06,
      "loss": 1.5604188,
      "memory(GiB)": 117.38,
      "step": 40545,
      "train_speed(iter/s)": 1.637447
    },
    {
      "acc": 0.65004511,
      "epoch": 1.0286656519533233,
      "grad_norm": 5.53125,
      "learning_rate": 5.176341924982997e-06,
      "loss": 1.58811197,
      "memory(GiB)": 117.38,
      "step": 40550,
      "train_speed(iter/s)": 1.637468
    },
    {
      "acc": 0.6679503,
      "epoch": 1.0287924911212583,
      "grad_norm": 6.59375,
      "learning_rate": 5.1752939498060435e-06,
      "loss": 1.56891251,
      "memory(GiB)": 117.38,
      "step": 40555,
      "train_speed(iter/s)": 1.637491
    },
    {
      "acc": 0.64970713,
      "epoch": 1.0289193302891932,
      "grad_norm": 5.59375,
      "learning_rate": 5.174245966918883e-06,
      "loss": 1.69450188,
      "memory(GiB)": 117.38,
      "step": 40560,
      "train_speed(iter/s)": 1.637513
    },
    {
      "acc": 0.63921604,
      "epoch": 1.0290461694571285,
      "grad_norm": 5.6875,
      "learning_rate": 5.173197976367603e-06,
      "loss": 1.6202755,
      "memory(GiB)": 117.38,
      "step": 40565,
      "train_speed(iter/s)": 1.637536
    },
    {
      "acc": 0.6622056,
      "epoch": 1.0291730086250634,
      "grad_norm": 5.53125,
      "learning_rate": 5.1721499781983055e-06,
      "loss": 1.54897089,
      "memory(GiB)": 117.38,
      "step": 40570,
      "train_speed(iter/s)": 1.637557
    },
    {
      "acc": 0.64063449,
      "epoch": 1.0292998477929984,
      "grad_norm": 4.96875,
      "learning_rate": 5.171101972457081e-06,
      "loss": 1.59721107,
      "memory(GiB)": 117.38,
      "step": 40575,
      "train_speed(iter/s)": 1.637579
    },
    {
      "acc": 0.65204802,
      "epoch": 1.0294266869609334,
      "grad_norm": 5.625,
      "learning_rate": 5.170053959190029e-06,
      "loss": 1.59807692,
      "memory(GiB)": 117.38,
      "step": 40580,
      "train_speed(iter/s)": 1.6376
    },
    {
      "acc": 0.65249763,
      "epoch": 1.0295535261288686,
      "grad_norm": 6.65625,
      "learning_rate": 5.169005938443245e-06,
      "loss": 1.61302605,
      "memory(GiB)": 117.38,
      "step": 40585,
      "train_speed(iter/s)": 1.637622
    },
    {
      "acc": 0.66014142,
      "epoch": 1.0296803652968036,
      "grad_norm": 5.46875,
      "learning_rate": 5.1679579102628245e-06,
      "loss": 1.59491444,
      "memory(GiB)": 117.38,
      "step": 40590,
      "train_speed(iter/s)": 1.637644
    },
    {
      "acc": 0.6484828,
      "epoch": 1.0298072044647386,
      "grad_norm": 5.28125,
      "learning_rate": 5.166909874694866e-06,
      "loss": 1.59603996,
      "memory(GiB)": 117.38,
      "step": 40595,
      "train_speed(iter/s)": 1.637666
    },
    {
      "acc": 0.64561977,
      "epoch": 1.0299340436326738,
      "grad_norm": 7.4375,
      "learning_rate": 5.165861831785465e-06,
      "loss": 1.69799347,
      "memory(GiB)": 117.38,
      "step": 40600,
      "train_speed(iter/s)": 1.637688
    },
    {
      "acc": 0.64031649,
      "epoch": 1.0300608828006088,
      "grad_norm": 5.75,
      "learning_rate": 5.164813781580721e-06,
      "loss": 1.62498169,
      "memory(GiB)": 117.38,
      "step": 40605,
      "train_speed(iter/s)": 1.637709
    },
    {
      "acc": 0.65559216,
      "epoch": 1.0301877219685438,
      "grad_norm": 5.125,
      "learning_rate": 5.16376572412673e-06,
      "loss": 1.57135181,
      "memory(GiB)": 117.38,
      "step": 40610,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.66612349,
      "epoch": 1.030314561136479,
      "grad_norm": 5.6875,
      "learning_rate": 5.162717659469593e-06,
      "loss": 1.6295536,
      "memory(GiB)": 117.38,
      "step": 40615,
      "train_speed(iter/s)": 1.637752
    },
    {
      "acc": 0.67055264,
      "epoch": 1.030441400304414,
      "grad_norm": 5.46875,
      "learning_rate": 5.161669587655406e-06,
      "loss": 1.52363548,
      "memory(GiB)": 117.38,
      "step": 40620,
      "train_speed(iter/s)": 1.637772
    },
    {
      "acc": 0.6556602,
      "epoch": 1.030568239472349,
      "grad_norm": 7.34375,
      "learning_rate": 5.160621508730267e-06,
      "loss": 1.56594753,
      "memory(GiB)": 117.38,
      "step": 40625,
      "train_speed(iter/s)": 1.637794
    },
    {
      "acc": 0.65218773,
      "epoch": 1.0306950786402842,
      "grad_norm": 5.65625,
      "learning_rate": 5.15957342274028e-06,
      "loss": 1.6275959,
      "memory(GiB)": 117.38,
      "step": 40630,
      "train_speed(iter/s)": 1.637816
    },
    {
      "acc": 0.64686565,
      "epoch": 1.0308219178082192,
      "grad_norm": 8.125,
      "learning_rate": 5.158525329731539e-06,
      "loss": 1.62003994,
      "memory(GiB)": 117.38,
      "step": 40635,
      "train_speed(iter/s)": 1.637837
    },
    {
      "acc": 0.65470314,
      "epoch": 1.0309487569761542,
      "grad_norm": 6.5625,
      "learning_rate": 5.157477229750149e-06,
      "loss": 1.67855587,
      "memory(GiB)": 117.38,
      "step": 40640,
      "train_speed(iter/s)": 1.63786
    },
    {
      "acc": 0.66197357,
      "epoch": 1.0310755961440894,
      "grad_norm": 5.46875,
      "learning_rate": 5.156429122842204e-06,
      "loss": 1.60478783,
      "memory(GiB)": 117.38,
      "step": 40645,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.66182189,
      "epoch": 1.0312024353120244,
      "grad_norm": 6.28125,
      "learning_rate": 5.15538100905381e-06,
      "loss": 1.57081213,
      "memory(GiB)": 117.38,
      "step": 40650,
      "train_speed(iter/s)": 1.637901
    },
    {
      "acc": 0.656461,
      "epoch": 1.0313292744799594,
      "grad_norm": 4.8125,
      "learning_rate": 5.154332888431064e-06,
      "loss": 1.58842239,
      "memory(GiB)": 117.38,
      "step": 40655,
      "train_speed(iter/s)": 1.637923
    },
    {
      "acc": 0.65025449,
      "epoch": 1.0314561136478946,
      "grad_norm": 6.15625,
      "learning_rate": 5.15328476102007e-06,
      "loss": 1.61803246,
      "memory(GiB)": 117.38,
      "step": 40660,
      "train_speed(iter/s)": 1.637945
    },
    {
      "acc": 0.6445549,
      "epoch": 1.0315829528158296,
      "grad_norm": 5.34375,
      "learning_rate": 5.1522366268669264e-06,
      "loss": 1.62090549,
      "memory(GiB)": 117.38,
      "step": 40665,
      "train_speed(iter/s)": 1.637966
    },
    {
      "acc": 0.65344067,
      "epoch": 1.0317097919837646,
      "grad_norm": 6.25,
      "learning_rate": 5.1511884860177376e-06,
      "loss": 1.59310169,
      "memory(GiB)": 117.38,
      "step": 40670,
      "train_speed(iter/s)": 1.637987
    },
    {
      "acc": 0.65493965,
      "epoch": 1.0318366311516995,
      "grad_norm": 5.96875,
      "learning_rate": 5.150140338518603e-06,
      "loss": 1.6139164,
      "memory(GiB)": 117.38,
      "step": 40675,
      "train_speed(iter/s)": 1.638008
    },
    {
      "acc": 0.66272364,
      "epoch": 1.0319634703196348,
      "grad_norm": 5.09375,
      "learning_rate": 5.149092184415627e-06,
      "loss": 1.52428246,
      "memory(GiB)": 117.38,
      "step": 40680,
      "train_speed(iter/s)": 1.63803
    },
    {
      "acc": 0.67664566,
      "epoch": 1.0320903094875697,
      "grad_norm": 5.15625,
      "learning_rate": 5.148044023754911e-06,
      "loss": 1.56870518,
      "memory(GiB)": 117.38,
      "step": 40685,
      "train_speed(iter/s)": 1.638052
    },
    {
      "acc": 0.64395576,
      "epoch": 1.0322171486555047,
      "grad_norm": 4.78125,
      "learning_rate": 5.146995856582557e-06,
      "loss": 1.62173195,
      "memory(GiB)": 117.38,
      "step": 40690,
      "train_speed(iter/s)": 1.638074
    },
    {
      "acc": 0.65144897,
      "epoch": 1.03234398782344,
      "grad_norm": 5.46875,
      "learning_rate": 5.14594768294467e-06,
      "loss": 1.65414658,
      "memory(GiB)": 117.38,
      "step": 40695,
      "train_speed(iter/s)": 1.638097
    },
    {
      "acc": 0.6609879,
      "epoch": 1.032470826991375,
      "grad_norm": 6.09375,
      "learning_rate": 5.1448995028873515e-06,
      "loss": 1.60159454,
      "memory(GiB)": 117.38,
      "step": 40700,
      "train_speed(iter/s)": 1.638117
    },
    {
      "acc": 0.65242009,
      "epoch": 1.03259766615931,
      "grad_norm": 5.5625,
      "learning_rate": 5.143851316456706e-06,
      "loss": 1.57185392,
      "memory(GiB)": 117.38,
      "step": 40705,
      "train_speed(iter/s)": 1.638139
    },
    {
      "acc": 0.64776373,
      "epoch": 1.0327245053272451,
      "grad_norm": 6.84375,
      "learning_rate": 5.142803123698838e-06,
      "loss": 1.64793663,
      "memory(GiB)": 117.38,
      "step": 40710,
      "train_speed(iter/s)": 1.63816
    },
    {
      "acc": 0.66486349,
      "epoch": 1.0328513444951801,
      "grad_norm": 5.625,
      "learning_rate": 5.14175492465985e-06,
      "loss": 1.52684345,
      "memory(GiB)": 117.38,
      "step": 40715,
      "train_speed(iter/s)": 1.638182
    },
    {
      "acc": 0.65848589,
      "epoch": 1.0329781836631151,
      "grad_norm": 5.8125,
      "learning_rate": 5.14070671938585e-06,
      "loss": 1.62021332,
      "memory(GiB)": 117.38,
      "step": 40720,
      "train_speed(iter/s)": 1.638203
    },
    {
      "acc": 0.65813723,
      "epoch": 1.0331050228310503,
      "grad_norm": 5.59375,
      "learning_rate": 5.139658507922937e-06,
      "loss": 1.58810902,
      "memory(GiB)": 117.38,
      "step": 40725,
      "train_speed(iter/s)": 1.638225
    },
    {
      "acc": 0.6496624,
      "epoch": 1.0332318619989853,
      "grad_norm": 6.84375,
      "learning_rate": 5.138610290317221e-06,
      "loss": 1.6423027,
      "memory(GiB)": 117.38,
      "step": 40730,
      "train_speed(iter/s)": 1.638247
    },
    {
      "acc": 0.6481904,
      "epoch": 1.0333587011669203,
      "grad_norm": 5.8125,
      "learning_rate": 5.137562066614805e-06,
      "loss": 1.61454353,
      "memory(GiB)": 117.38,
      "step": 40735,
      "train_speed(iter/s)": 1.638269
    },
    {
      "acc": 0.66562257,
      "epoch": 1.0334855403348553,
      "grad_norm": 8.0,
      "learning_rate": 5.136513836861795e-06,
      "loss": 1.53041515,
      "memory(GiB)": 117.38,
      "step": 40740,
      "train_speed(iter/s)": 1.638293
    },
    {
      "acc": 0.66315899,
      "epoch": 1.0336123795027905,
      "grad_norm": 5.09375,
      "learning_rate": 5.135465601104298e-06,
      "loss": 1.65367966,
      "memory(GiB)": 117.38,
      "step": 40745,
      "train_speed(iter/s)": 1.638314
    },
    {
      "acc": 0.68181005,
      "epoch": 1.0337392186707255,
      "grad_norm": 5.1875,
      "learning_rate": 5.134417359388418e-06,
      "loss": 1.53159122,
      "memory(GiB)": 117.38,
      "step": 40750,
      "train_speed(iter/s)": 1.638336
    },
    {
      "acc": 0.65063896,
      "epoch": 1.0338660578386605,
      "grad_norm": 6.0625,
      "learning_rate": 5.133369111760264e-06,
      "loss": 1.60261383,
      "memory(GiB)": 117.38,
      "step": 40755,
      "train_speed(iter/s)": 1.638358
    },
    {
      "acc": 0.66811352,
      "epoch": 1.0339928970065957,
      "grad_norm": 7.40625,
      "learning_rate": 5.132320858265939e-06,
      "loss": 1.56639633,
      "memory(GiB)": 117.38,
      "step": 40760,
      "train_speed(iter/s)": 1.638381
    },
    {
      "acc": 0.66822629,
      "epoch": 1.0341197361745307,
      "grad_norm": 7.40625,
      "learning_rate": 5.131272598951554e-06,
      "loss": 1.58580513,
      "memory(GiB)": 117.38,
      "step": 40765,
      "train_speed(iter/s)": 1.638401
    },
    {
      "acc": 0.66615558,
      "epoch": 1.0342465753424657,
      "grad_norm": 4.625,
      "learning_rate": 5.130224333863212e-06,
      "loss": 1.49721498,
      "memory(GiB)": 117.38,
      "step": 40770,
      "train_speed(iter/s)": 1.638422
    },
    {
      "acc": 0.65318537,
      "epoch": 1.0343734145104009,
      "grad_norm": 6.125,
      "learning_rate": 5.129176063047022e-06,
      "loss": 1.60644112,
      "memory(GiB)": 117.38,
      "step": 40775,
      "train_speed(iter/s)": 1.638445
    },
    {
      "acc": 0.64278841,
      "epoch": 1.0345002536783359,
      "grad_norm": 6.09375,
      "learning_rate": 5.128127786549094e-06,
      "loss": 1.67408009,
      "memory(GiB)": 117.38,
      "step": 40780,
      "train_speed(iter/s)": 1.638467
    },
    {
      "acc": 0.65311394,
      "epoch": 1.0346270928462709,
      "grad_norm": 5.40625,
      "learning_rate": 5.127079504415532e-06,
      "loss": 1.6130373,
      "memory(GiB)": 117.38,
      "step": 40785,
      "train_speed(iter/s)": 1.638487
    },
    {
      "acc": 0.67491827,
      "epoch": 1.034753932014206,
      "grad_norm": 6.53125,
      "learning_rate": 5.126031216692449e-06,
      "loss": 1.52196407,
      "memory(GiB)": 117.38,
      "step": 40790,
      "train_speed(iter/s)": 1.638508
    },
    {
      "acc": 0.65585279,
      "epoch": 1.034880771182141,
      "grad_norm": 4.71875,
      "learning_rate": 5.124982923425947e-06,
      "loss": 1.58482971,
      "memory(GiB)": 117.38,
      "step": 40795,
      "train_speed(iter/s)": 1.638527
    },
    {
      "acc": 0.65393801,
      "epoch": 1.035007610350076,
      "grad_norm": 4.875,
      "learning_rate": 5.123934624662139e-06,
      "loss": 1.61756001,
      "memory(GiB)": 117.38,
      "step": 40800,
      "train_speed(iter/s)": 1.638548
    },
    {
      "acc": 0.638975,
      "epoch": 1.0351344495180113,
      "grad_norm": 7.75,
      "learning_rate": 5.1228863204471335e-06,
      "loss": 1.65982056,
      "memory(GiB)": 117.38,
      "step": 40805,
      "train_speed(iter/s)": 1.63857
    },
    {
      "acc": 0.66074066,
      "epoch": 1.0352612886859462,
      "grad_norm": 4.5,
      "learning_rate": 5.121838010827039e-06,
      "loss": 1.56035404,
      "memory(GiB)": 117.38,
      "step": 40810,
      "train_speed(iter/s)": 1.638591
    },
    {
      "acc": 0.6509923,
      "epoch": 1.0353881278538812,
      "grad_norm": 5.4375,
      "learning_rate": 5.120789695847965e-06,
      "loss": 1.61882,
      "memory(GiB)": 117.38,
      "step": 40815,
      "train_speed(iter/s)": 1.638613
    },
    {
      "acc": 0.65156822,
      "epoch": 1.0355149670218164,
      "grad_norm": 5.875,
      "learning_rate": 5.119741375556021e-06,
      "loss": 1.62566681,
      "memory(GiB)": 117.38,
      "step": 40820,
      "train_speed(iter/s)": 1.638635
    },
    {
      "acc": 0.68064942,
      "epoch": 1.0356418061897514,
      "grad_norm": 6.25,
      "learning_rate": 5.118693049997316e-06,
      "loss": 1.51892948,
      "memory(GiB)": 117.38,
      "step": 40825,
      "train_speed(iter/s)": 1.638657
    },
    {
      "acc": 0.64971361,
      "epoch": 1.0357686453576864,
      "grad_norm": 6.4375,
      "learning_rate": 5.117644719217961e-06,
      "loss": 1.613204,
      "memory(GiB)": 117.38,
      "step": 40830,
      "train_speed(iter/s)": 1.638679
    },
    {
      "acc": 0.65265303,
      "epoch": 1.0358954845256214,
      "grad_norm": 6.09375,
      "learning_rate": 5.116596383264066e-06,
      "loss": 1.59452076,
      "memory(GiB)": 117.38,
      "step": 40835,
      "train_speed(iter/s)": 1.6387
    },
    {
      "acc": 0.64392495,
      "epoch": 1.0360223236935566,
      "grad_norm": 9.125,
      "learning_rate": 5.115548042181742e-06,
      "loss": 1.65839653,
      "memory(GiB)": 117.38,
      "step": 40840,
      "train_speed(iter/s)": 1.638721
    },
    {
      "acc": 0.66803961,
      "epoch": 1.0361491628614916,
      "grad_norm": 4.875,
      "learning_rate": 5.114499696017098e-06,
      "loss": 1.50407085,
      "memory(GiB)": 117.38,
      "step": 40845,
      "train_speed(iter/s)": 1.638742
    },
    {
      "acc": 0.65188293,
      "epoch": 1.0362760020294266,
      "grad_norm": 5.21875,
      "learning_rate": 5.1134513448162475e-06,
      "loss": 1.66241341,
      "memory(GiB)": 117.38,
      "step": 40850,
      "train_speed(iter/s)": 1.638764
    },
    {
      "acc": 0.6425087,
      "epoch": 1.0364028411973618,
      "grad_norm": 5.3125,
      "learning_rate": 5.112402988625299e-06,
      "loss": 1.62109718,
      "memory(GiB)": 117.38,
      "step": 40855,
      "train_speed(iter/s)": 1.638786
    },
    {
      "acc": 0.66601357,
      "epoch": 1.0365296803652968,
      "grad_norm": 6.21875,
      "learning_rate": 5.111354627490367e-06,
      "loss": 1.54996109,
      "memory(GiB)": 117.38,
      "step": 40860,
      "train_speed(iter/s)": 1.638806
    },
    {
      "acc": 0.64003944,
      "epoch": 1.0366565195332318,
      "grad_norm": 5.46875,
      "learning_rate": 5.110306261457559e-06,
      "loss": 1.66370449,
      "memory(GiB)": 117.38,
      "step": 40865,
      "train_speed(iter/s)": 1.638829
    },
    {
      "acc": 0.65367227,
      "epoch": 1.036783358701167,
      "grad_norm": 5.03125,
      "learning_rate": 5.109257890572991e-06,
      "loss": 1.66479588,
      "memory(GiB)": 117.38,
      "step": 40870,
      "train_speed(iter/s)": 1.638851
    },
    {
      "acc": 0.65728073,
      "epoch": 1.036910197869102,
      "grad_norm": 6.71875,
      "learning_rate": 5.108209514882772e-06,
      "loss": 1.58111229,
      "memory(GiB)": 117.38,
      "step": 40875,
      "train_speed(iter/s)": 1.638873
    },
    {
      "acc": 0.64618697,
      "epoch": 1.037037037037037,
      "grad_norm": 6.59375,
      "learning_rate": 5.107161134433017e-06,
      "loss": 1.64139786,
      "memory(GiB)": 117.38,
      "step": 40880,
      "train_speed(iter/s)": 1.638893
    },
    {
      "acc": 0.65989714,
      "epoch": 1.0371638762049722,
      "grad_norm": 6.84375,
      "learning_rate": 5.106112749269835e-06,
      "loss": 1.56313238,
      "memory(GiB)": 117.38,
      "step": 40885,
      "train_speed(iter/s)": 1.638915
    },
    {
      "acc": 0.65228424,
      "epoch": 1.0372907153729072,
      "grad_norm": 10.0,
      "learning_rate": 5.105064359439341e-06,
      "loss": 1.64385548,
      "memory(GiB)": 117.38,
      "step": 40890,
      "train_speed(iter/s)": 1.638936
    },
    {
      "acc": 0.6694046,
      "epoch": 1.0374175545408422,
      "grad_norm": 5.9375,
      "learning_rate": 5.1040159649876485e-06,
      "loss": 1.50057182,
      "memory(GiB)": 117.38,
      "step": 40895,
      "train_speed(iter/s)": 1.638957
    },
    {
      "acc": 0.64544067,
      "epoch": 1.0375443937087772,
      "grad_norm": 6.5625,
      "learning_rate": 5.102967565960868e-06,
      "loss": 1.65781517,
      "memory(GiB)": 117.38,
      "step": 40900,
      "train_speed(iter/s)": 1.638977
    },
    {
      "acc": 0.65514412,
      "epoch": 1.0376712328767124,
      "grad_norm": 5.59375,
      "learning_rate": 5.101919162405116e-06,
      "loss": 1.61020088,
      "memory(GiB)": 117.38,
      "step": 40905,
      "train_speed(iter/s)": 1.638997
    },
    {
      "acc": 0.66149073,
      "epoch": 1.0377980720446474,
      "grad_norm": 5.6875,
      "learning_rate": 5.100870754366503e-06,
      "loss": 1.56553469,
      "memory(GiB)": 117.38,
      "step": 40910,
      "train_speed(iter/s)": 1.639019
    },
    {
      "acc": 0.67709308,
      "epoch": 1.0379249112125823,
      "grad_norm": 6.6875,
      "learning_rate": 5.099822341891144e-06,
      "loss": 1.50459242,
      "memory(GiB)": 117.38,
      "step": 40915,
      "train_speed(iter/s)": 1.639041
    },
    {
      "acc": 0.64800062,
      "epoch": 1.0380517503805176,
      "grad_norm": 5.875,
      "learning_rate": 5.098773925025152e-06,
      "loss": 1.61434307,
      "memory(GiB)": 117.38,
      "step": 40920,
      "train_speed(iter/s)": 1.639063
    },
    {
      "acc": 0.63802557,
      "epoch": 1.0381785895484525,
      "grad_norm": 5.84375,
      "learning_rate": 5.097725503814643e-06,
      "loss": 1.68307896,
      "memory(GiB)": 117.38,
      "step": 40925,
      "train_speed(iter/s)": 1.639084
    },
    {
      "acc": 0.65785446,
      "epoch": 1.0383054287163875,
      "grad_norm": 4.90625,
      "learning_rate": 5.09667707830573e-06,
      "loss": 1.56549129,
      "memory(GiB)": 117.38,
      "step": 40930,
      "train_speed(iter/s)": 1.639107
    },
    {
      "acc": 0.67706718,
      "epoch": 1.0384322678843227,
      "grad_norm": 6.3125,
      "learning_rate": 5.095628648544526e-06,
      "loss": 1.47671614,
      "memory(GiB)": 117.38,
      "step": 40935,
      "train_speed(iter/s)": 1.639128
    },
    {
      "acc": 0.65696793,
      "epoch": 1.0385591070522577,
      "grad_norm": 5.65625,
      "learning_rate": 5.0945802145771495e-06,
      "loss": 1.56331949,
      "memory(GiB)": 117.38,
      "step": 40940,
      "train_speed(iter/s)": 1.63915
    },
    {
      "acc": 0.66067467,
      "epoch": 1.0386859462201927,
      "grad_norm": 5.5,
      "learning_rate": 5.093531776449711e-06,
      "loss": 1.54960203,
      "memory(GiB)": 117.38,
      "step": 40945,
      "train_speed(iter/s)": 1.63917
    },
    {
      "acc": 0.65931311,
      "epoch": 1.038812785388128,
      "grad_norm": 5.875,
      "learning_rate": 5.092483334208327e-06,
      "loss": 1.60602036,
      "memory(GiB)": 117.38,
      "step": 40950,
      "train_speed(iter/s)": 1.639191
    },
    {
      "acc": 0.64333758,
      "epoch": 1.038939624556063,
      "grad_norm": 6.34375,
      "learning_rate": 5.091434887899114e-06,
      "loss": 1.63052254,
      "memory(GiB)": 117.38,
      "step": 40955,
      "train_speed(iter/s)": 1.639212
    },
    {
      "acc": 0.65728874,
      "epoch": 1.039066463723998,
      "grad_norm": 6.03125,
      "learning_rate": 5.0903864375681866e-06,
      "loss": 1.6105011,
      "memory(GiB)": 117.38,
      "step": 40960,
      "train_speed(iter/s)": 1.639233
    },
    {
      "acc": 0.66116171,
      "epoch": 1.0391933028919331,
      "grad_norm": 5.46875,
      "learning_rate": 5.0893379832616594e-06,
      "loss": 1.5755868,
      "memory(GiB)": 117.38,
      "step": 40965,
      "train_speed(iter/s)": 1.639252
    },
    {
      "acc": 0.66144433,
      "epoch": 1.0393201420598681,
      "grad_norm": 5.875,
      "learning_rate": 5.08828952502565e-06,
      "loss": 1.57303867,
      "memory(GiB)": 117.38,
      "step": 40970,
      "train_speed(iter/s)": 1.639274
    },
    {
      "acc": 0.65965929,
      "epoch": 1.039446981227803,
      "grad_norm": 5.03125,
      "learning_rate": 5.087241062906272e-06,
      "loss": 1.60068188,
      "memory(GiB)": 117.38,
      "step": 40975,
      "train_speed(iter/s)": 1.639295
    },
    {
      "acc": 0.63852367,
      "epoch": 1.0395738203957383,
      "grad_norm": 4.96875,
      "learning_rate": 5.086192596949643e-06,
      "loss": 1.64146194,
      "memory(GiB)": 117.38,
      "step": 40980,
      "train_speed(iter/s)": 1.639316
    },
    {
      "acc": 0.66799808,
      "epoch": 1.0397006595636733,
      "grad_norm": 5.53125,
      "learning_rate": 5.085144127201879e-06,
      "loss": 1.55656929,
      "memory(GiB)": 117.38,
      "step": 40985,
      "train_speed(iter/s)": 1.639338
    },
    {
      "acc": 0.66920452,
      "epoch": 1.0398274987316083,
      "grad_norm": 6.84375,
      "learning_rate": 5.084095653709096e-06,
      "loss": 1.48647404,
      "memory(GiB)": 117.38,
      "step": 40990,
      "train_speed(iter/s)": 1.639358
    },
    {
      "acc": 0.676647,
      "epoch": 1.0399543378995433,
      "grad_norm": 5.8125,
      "learning_rate": 5.0830471765174096e-06,
      "loss": 1.5691246,
      "memory(GiB)": 117.38,
      "step": 40995,
      "train_speed(iter/s)": 1.639379
    },
    {
      "acc": 0.65418453,
      "epoch": 1.0400811770674785,
      "grad_norm": 6.59375,
      "learning_rate": 5.0819986956729395e-06,
      "loss": 1.62014542,
      "memory(GiB)": 117.38,
      "step": 41000,
      "train_speed(iter/s)": 1.639401
    },
    {
      "epoch": 1.0400811770674785,
      "eval_acc": 0.646207949346047,
      "eval_loss": 1.5738683938980103,
      "eval_runtime": 58.4919,
      "eval_samples_per_second": 108.904,
      "eval_steps_per_second": 27.235,
      "step": 41000
    },
    {
      "acc": 0.67275405,
      "epoch": 1.0402080162354135,
      "grad_norm": 7.375,
      "learning_rate": 5.080950211221799e-06,
      "loss": 1.5470417,
      "memory(GiB)": 117.38,
      "step": 41005,
      "train_speed(iter/s)": 1.635319
    },
    {
      "acc": 0.65025458,
      "epoch": 1.0403348554033485,
      "grad_norm": 5.90625,
      "learning_rate": 5.079901723210109e-06,
      "loss": 1.574189,
      "memory(GiB)": 117.38,
      "step": 41010,
      "train_speed(iter/s)": 1.635339
    },
    {
      "acc": 0.65324278,
      "epoch": 1.0404616945712837,
      "grad_norm": 5.875,
      "learning_rate": 5.078853231683981e-06,
      "loss": 1.58726282,
      "memory(GiB)": 117.38,
      "step": 41015,
      "train_speed(iter/s)": 1.635358
    },
    {
      "acc": 0.65567102,
      "epoch": 1.0405885337392187,
      "grad_norm": 5.59375,
      "learning_rate": 5.077804736689539e-06,
      "loss": 1.64108505,
      "memory(GiB)": 117.38,
      "step": 41020,
      "train_speed(iter/s)": 1.635379
    },
    {
      "acc": 0.66853514,
      "epoch": 1.0407153729071537,
      "grad_norm": 6.15625,
      "learning_rate": 5.0767562382728955e-06,
      "loss": 1.54305401,
      "memory(GiB)": 117.38,
      "step": 41025,
      "train_speed(iter/s)": 1.635399
    },
    {
      "acc": 0.64062362,
      "epoch": 1.0408422120750889,
      "grad_norm": 6.84375,
      "learning_rate": 5.075707736480171e-06,
      "loss": 1.70396004,
      "memory(GiB)": 117.38,
      "step": 41030,
      "train_speed(iter/s)": 1.635419
    },
    {
      "acc": 0.65617466,
      "epoch": 1.0409690512430239,
      "grad_norm": 5.75,
      "learning_rate": 5.074659231357482e-06,
      "loss": 1.58250599,
      "memory(GiB)": 117.38,
      "step": 41035,
      "train_speed(iter/s)": 1.635438
    },
    {
      "acc": 0.65760169,
      "epoch": 1.0410958904109588,
      "grad_norm": 4.5625,
      "learning_rate": 5.073610722950947e-06,
      "loss": 1.60189304,
      "memory(GiB)": 117.38,
      "step": 41040,
      "train_speed(iter/s)": 1.635458
    },
    {
      "acc": 0.64942598,
      "epoch": 1.041222729578894,
      "grad_norm": 7.3125,
      "learning_rate": 5.072562211306683e-06,
      "loss": 1.58427477,
      "memory(GiB)": 117.38,
      "step": 41045,
      "train_speed(iter/s)": 1.635477
    },
    {
      "acc": 0.64224863,
      "epoch": 1.041349568746829,
      "grad_norm": 6.4375,
      "learning_rate": 5.071513696470809e-06,
      "loss": 1.65665932,
      "memory(GiB)": 117.38,
      "step": 41050,
      "train_speed(iter/s)": 1.635495
    },
    {
      "acc": 0.66928444,
      "epoch": 1.041476407914764,
      "grad_norm": 7.15625,
      "learning_rate": 5.070465178489443e-06,
      "loss": 1.58906279,
      "memory(GiB)": 117.38,
      "step": 41055,
      "train_speed(iter/s)": 1.635515
    },
    {
      "acc": 0.66286411,
      "epoch": 1.041603247082699,
      "grad_norm": 5.375,
      "learning_rate": 5.069416657408704e-06,
      "loss": 1.57125578,
      "memory(GiB)": 117.38,
      "step": 41060,
      "train_speed(iter/s)": 1.635533
    },
    {
      "acc": 0.65803609,
      "epoch": 1.0417300862506342,
      "grad_norm": 4.78125,
      "learning_rate": 5.0683681332747105e-06,
      "loss": 1.64628754,
      "memory(GiB)": 117.38,
      "step": 41065,
      "train_speed(iter/s)": 1.635552
    },
    {
      "acc": 0.658601,
      "epoch": 1.0418569254185692,
      "grad_norm": 5.0625,
      "learning_rate": 5.067319606133583e-06,
      "loss": 1.63151436,
      "memory(GiB)": 117.38,
      "step": 41070,
      "train_speed(iter/s)": 1.635572
    },
    {
      "acc": 0.67268991,
      "epoch": 1.0419837645865042,
      "grad_norm": 6.625,
      "learning_rate": 5.066271076031436e-06,
      "loss": 1.55479927,
      "memory(GiB)": 117.38,
      "step": 41075,
      "train_speed(iter/s)": 1.635592
    },
    {
      "acc": 0.65339289,
      "epoch": 1.0421106037544394,
      "grad_norm": 6.125,
      "learning_rate": 5.065222543014394e-06,
      "loss": 1.62807522,
      "memory(GiB)": 117.38,
      "step": 41080,
      "train_speed(iter/s)": 1.63561
    },
    {
      "acc": 0.652844,
      "epoch": 1.0422374429223744,
      "grad_norm": 7.78125,
      "learning_rate": 5.06417400712857e-06,
      "loss": 1.62595863,
      "memory(GiB)": 117.38,
      "step": 41085,
      "train_speed(iter/s)": 1.63563
    },
    {
      "acc": 0.66335192,
      "epoch": 1.0423642820903094,
      "grad_norm": 5.40625,
      "learning_rate": 5.0631254684200906e-06,
      "loss": 1.53417234,
      "memory(GiB)": 117.38,
      "step": 41090,
      "train_speed(iter/s)": 1.63565
    },
    {
      "acc": 0.66178837,
      "epoch": 1.0424911212582446,
      "grad_norm": 5.75,
      "learning_rate": 5.062076926935068e-06,
      "loss": 1.59790726,
      "memory(GiB)": 117.38,
      "step": 41095,
      "train_speed(iter/s)": 1.635672
    },
    {
      "acc": 0.66018434,
      "epoch": 1.0426179604261796,
      "grad_norm": 6.15625,
      "learning_rate": 5.061028382719626e-06,
      "loss": 1.54263077,
      "memory(GiB)": 117.38,
      "step": 41100,
      "train_speed(iter/s)": 1.635693
    },
    {
      "acc": 0.66236649,
      "epoch": 1.0427447995941146,
      "grad_norm": 5.875,
      "learning_rate": 5.0599798358198835e-06,
      "loss": 1.57164869,
      "memory(GiB)": 117.38,
      "step": 41105,
      "train_speed(iter/s)": 1.635714
    },
    {
      "acc": 0.65376139,
      "epoch": 1.0428716387620498,
      "grad_norm": 5.625,
      "learning_rate": 5.0589312862819605e-06,
      "loss": 1.60471878,
      "memory(GiB)": 117.38,
      "step": 41110,
      "train_speed(iter/s)": 1.635734
    },
    {
      "acc": 0.66896567,
      "epoch": 1.0429984779299848,
      "grad_norm": 5.28125,
      "learning_rate": 5.057882734151977e-06,
      "loss": 1.52460117,
      "memory(GiB)": 117.38,
      "step": 41115,
      "train_speed(iter/s)": 1.635754
    },
    {
      "acc": 0.65420294,
      "epoch": 1.0431253170979198,
      "grad_norm": 5.625,
      "learning_rate": 5.05683417947605e-06,
      "loss": 1.60587082,
      "memory(GiB)": 117.38,
      "step": 41120,
      "train_speed(iter/s)": 1.635775
    },
    {
      "acc": 0.65285158,
      "epoch": 1.043252156265855,
      "grad_norm": 6.28125,
      "learning_rate": 5.055785622300303e-06,
      "loss": 1.61026115,
      "memory(GiB)": 117.38,
      "step": 41125,
      "train_speed(iter/s)": 1.635796
    },
    {
      "acc": 0.65437546,
      "epoch": 1.04337899543379,
      "grad_norm": 4.90625,
      "learning_rate": 5.054737062670857e-06,
      "loss": 1.62572021,
      "memory(GiB)": 117.38,
      "step": 41130,
      "train_speed(iter/s)": 1.635816
    },
    {
      "acc": 0.65306063,
      "epoch": 1.043505834601725,
      "grad_norm": 5.53125,
      "learning_rate": 5.053688500633828e-06,
      "loss": 1.57150707,
      "memory(GiB)": 117.38,
      "step": 41135,
      "train_speed(iter/s)": 1.635836
    },
    {
      "acc": 0.66121545,
      "epoch": 1.0436326737696602,
      "grad_norm": 6.28125,
      "learning_rate": 5.052639936235341e-06,
      "loss": 1.55277061,
      "memory(GiB)": 117.38,
      "step": 41140,
      "train_speed(iter/s)": 1.63585
    },
    {
      "acc": 0.66214256,
      "epoch": 1.0437595129375952,
      "grad_norm": 7.0625,
      "learning_rate": 5.051591369521513e-06,
      "loss": 1.6219347,
      "memory(GiB)": 117.38,
      "step": 41145,
      "train_speed(iter/s)": 1.635871
    },
    {
      "acc": 0.64794827,
      "epoch": 1.0438863521055302,
      "grad_norm": 5.96875,
      "learning_rate": 5.050542800538469e-06,
      "loss": 1.58073626,
      "memory(GiB)": 117.38,
      "step": 41150,
      "train_speed(iter/s)": 1.635893
    },
    {
      "acc": 0.66172285,
      "epoch": 1.0440131912734651,
      "grad_norm": 5.71875,
      "learning_rate": 5.049494229332324e-06,
      "loss": 1.63676605,
      "memory(GiB)": 117.38,
      "step": 41155,
      "train_speed(iter/s)": 1.635915
    },
    {
      "acc": 0.66545863,
      "epoch": 1.0441400304414004,
      "grad_norm": 5.21875,
      "learning_rate": 5.048445655949204e-06,
      "loss": 1.60956802,
      "memory(GiB)": 117.38,
      "step": 41160,
      "train_speed(iter/s)": 1.635935
    },
    {
      "acc": 0.64485331,
      "epoch": 1.0442668696093353,
      "grad_norm": 5.78125,
      "learning_rate": 5.047397080435225e-06,
      "loss": 1.69102707,
      "memory(GiB)": 117.38,
      "step": 41165,
      "train_speed(iter/s)": 1.635956
    },
    {
      "acc": 0.65639114,
      "epoch": 1.0443937087772703,
      "grad_norm": 5.46875,
      "learning_rate": 5.046348502836512e-06,
      "loss": 1.65284081,
      "memory(GiB)": 117.38,
      "step": 41170,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.64845624,
      "epoch": 1.0445205479452055,
      "grad_norm": 5.1875,
      "learning_rate": 5.045299923199186e-06,
      "loss": 1.61099606,
      "memory(GiB)": 117.38,
      "step": 41175,
      "train_speed(iter/s)": 1.635998
    },
    {
      "acc": 0.65766449,
      "epoch": 1.0446473871131405,
      "grad_norm": 5.0,
      "learning_rate": 5.044251341569366e-06,
      "loss": 1.57406473,
      "memory(GiB)": 117.38,
      "step": 41180,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.63691139,
      "epoch": 1.0447742262810755,
      "grad_norm": 7.0625,
      "learning_rate": 5.043202757993175e-06,
      "loss": 1.70735168,
      "memory(GiB)": 117.38,
      "step": 41185,
      "train_speed(iter/s)": 1.636038
    },
    {
      "acc": 0.64051337,
      "epoch": 1.0449010654490107,
      "grad_norm": 7.75,
      "learning_rate": 5.042154172516734e-06,
      "loss": 1.65134926,
      "memory(GiB)": 117.38,
      "step": 41190,
      "train_speed(iter/s)": 1.636059
    },
    {
      "acc": 0.66136446,
      "epoch": 1.0450279046169457,
      "grad_norm": 5.5,
      "learning_rate": 5.041105585186164e-06,
      "loss": 1.59882412,
      "memory(GiB)": 117.38,
      "step": 41195,
      "train_speed(iter/s)": 1.636079
    },
    {
      "acc": 0.67160959,
      "epoch": 1.0451547437848807,
      "grad_norm": 4.9375,
      "learning_rate": 5.040056996047587e-06,
      "loss": 1.6220026,
      "memory(GiB)": 117.38,
      "step": 41200,
      "train_speed(iter/s)": 1.6361
    },
    {
      "acc": 0.66530619,
      "epoch": 1.045281582952816,
      "grad_norm": 5.59375,
      "learning_rate": 5.039008405147125e-06,
      "loss": 1.54878445,
      "memory(GiB)": 117.38,
      "step": 41205,
      "train_speed(iter/s)": 1.63612
    },
    {
      "acc": 0.64306927,
      "epoch": 1.045408422120751,
      "grad_norm": 6.53125,
      "learning_rate": 5.0379598125308984e-06,
      "loss": 1.64910774,
      "memory(GiB)": 117.38,
      "step": 41210,
      "train_speed(iter/s)": 1.636141
    },
    {
      "acc": 0.64517002,
      "epoch": 1.045535261288686,
      "grad_norm": 6.3125,
      "learning_rate": 5.036911218245029e-06,
      "loss": 1.68365669,
      "memory(GiB)": 117.38,
      "step": 41215,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.65690842,
      "epoch": 1.045662100456621,
      "grad_norm": 6.03125,
      "learning_rate": 5.035862622335641e-06,
      "loss": 1.59201756,
      "memory(GiB)": 117.38,
      "step": 41220,
      "train_speed(iter/s)": 1.636181
    },
    {
      "acc": 0.66846271,
      "epoch": 1.045788939624556,
      "grad_norm": 5.34375,
      "learning_rate": 5.034814024848853e-06,
      "loss": 1.55792236,
      "memory(GiB)": 117.38,
      "step": 41225,
      "train_speed(iter/s)": 1.636202
    },
    {
      "acc": 0.66630754,
      "epoch": 1.045915778792491,
      "grad_norm": 5.1875,
      "learning_rate": 5.033765425830791e-06,
      "loss": 1.5664361,
      "memory(GiB)": 117.38,
      "step": 41230,
      "train_speed(iter/s)": 1.636223
    },
    {
      "acc": 0.66141944,
      "epoch": 1.046042617960426,
      "grad_norm": 7.3125,
      "learning_rate": 5.032716825327573e-06,
      "loss": 1.5978425,
      "memory(GiB)": 117.38,
      "step": 41235,
      "train_speed(iter/s)": 1.636243
    },
    {
      "acc": 0.65883985,
      "epoch": 1.0461694571283613,
      "grad_norm": 6.375,
      "learning_rate": 5.031668223385323e-06,
      "loss": 1.58060741,
      "memory(GiB)": 117.38,
      "step": 41240,
      "train_speed(iter/s)": 1.636265
    },
    {
      "acc": 0.64730096,
      "epoch": 1.0462962962962963,
      "grad_norm": 6.03125,
      "learning_rate": 5.030619620050163e-06,
      "loss": 1.61114693,
      "memory(GiB)": 117.38,
      "step": 41245,
      "train_speed(iter/s)": 1.636287
    },
    {
      "acc": 0.6748704,
      "epoch": 1.0464231354642313,
      "grad_norm": 5.96875,
      "learning_rate": 5.029571015368217e-06,
      "loss": 1.47259388,
      "memory(GiB)": 117.38,
      "step": 41250,
      "train_speed(iter/s)": 1.636307
    },
    {
      "acc": 0.64501467,
      "epoch": 1.0465499746321665,
      "grad_norm": 5.40625,
      "learning_rate": 5.028522409385605e-06,
      "loss": 1.59153805,
      "memory(GiB)": 117.38,
      "step": 41255,
      "train_speed(iter/s)": 1.636329
    },
    {
      "acc": 0.66175165,
      "epoch": 1.0466768138001015,
      "grad_norm": 5.03125,
      "learning_rate": 5.0274738021484495e-06,
      "loss": 1.52612638,
      "memory(GiB)": 117.38,
      "step": 41260,
      "train_speed(iter/s)": 1.636351
    },
    {
      "acc": 0.64561868,
      "epoch": 1.0468036529680365,
      "grad_norm": 5.96875,
      "learning_rate": 5.026425193702874e-06,
      "loss": 1.63842659,
      "memory(GiB)": 117.38,
      "step": 41265,
      "train_speed(iter/s)": 1.636372
    },
    {
      "acc": 0.68379211,
      "epoch": 1.0469304921359717,
      "grad_norm": 5.53125,
      "learning_rate": 5.025376584095001e-06,
      "loss": 1.4420042,
      "memory(GiB)": 117.38,
      "step": 41270,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.66751194,
      "epoch": 1.0470573313039067,
      "grad_norm": 6.375,
      "learning_rate": 5.024327973370951e-06,
      "loss": 1.48990412,
      "memory(GiB)": 117.38,
      "step": 41275,
      "train_speed(iter/s)": 1.636417
    },
    {
      "acc": 0.64170256,
      "epoch": 1.0471841704718416,
      "grad_norm": 6.78125,
      "learning_rate": 5.02327936157685e-06,
      "loss": 1.65868225,
      "memory(GiB)": 117.38,
      "step": 41280,
      "train_speed(iter/s)": 1.636439
    },
    {
      "acc": 0.65187325,
      "epoch": 1.0473110096397769,
      "grad_norm": 5.78125,
      "learning_rate": 5.022230748758816e-06,
      "loss": 1.55477715,
      "memory(GiB)": 117.38,
      "step": 41285,
      "train_speed(iter/s)": 1.636458
    },
    {
      "acc": 0.66740675,
      "epoch": 1.0474378488077118,
      "grad_norm": 6.53125,
      "learning_rate": 5.021182134962978e-06,
      "loss": 1.59187527,
      "memory(GiB)": 117.38,
      "step": 41290,
      "train_speed(iter/s)": 1.636478
    },
    {
      "acc": 0.66480932,
      "epoch": 1.0475646879756468,
      "grad_norm": 4.65625,
      "learning_rate": 5.020133520235453e-06,
      "loss": 1.59848013,
      "memory(GiB)": 117.38,
      "step": 41295,
      "train_speed(iter/s)": 1.636498
    },
    {
      "acc": 0.64622912,
      "epoch": 1.047691527143582,
      "grad_norm": 5.28125,
      "learning_rate": 5.019084904622367e-06,
      "loss": 1.6313488,
      "memory(GiB)": 117.38,
      "step": 41300,
      "train_speed(iter/s)": 1.636518
    },
    {
      "acc": 0.65072818,
      "epoch": 1.047818366311517,
      "grad_norm": 4.53125,
      "learning_rate": 5.01803628816984e-06,
      "loss": 1.59510641,
      "memory(GiB)": 117.38,
      "step": 41305,
      "train_speed(iter/s)": 1.636539
    },
    {
      "acc": 0.65696106,
      "epoch": 1.047945205479452,
      "grad_norm": 6.625,
      "learning_rate": 5.016987670923998e-06,
      "loss": 1.53928385,
      "memory(GiB)": 117.38,
      "step": 41310,
      "train_speed(iter/s)": 1.63656
    },
    {
      "acc": 0.66856894,
      "epoch": 1.048072044647387,
      "grad_norm": 5.90625,
      "learning_rate": 5.0159390529309615e-06,
      "loss": 1.48899651,
      "memory(GiB)": 117.38,
      "step": 41315,
      "train_speed(iter/s)": 1.636582
    },
    {
      "acc": 0.65610499,
      "epoch": 1.0481988838153222,
      "grad_norm": 6.6875,
      "learning_rate": 5.014890434236854e-06,
      "loss": 1.56075859,
      "memory(GiB)": 117.38,
      "step": 41320,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.63999534,
      "epoch": 1.0483257229832572,
      "grad_norm": 7.21875,
      "learning_rate": 5.0138418148878e-06,
      "loss": 1.68905296,
      "memory(GiB)": 117.38,
      "step": 41325,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.66330452,
      "epoch": 1.0484525621511922,
      "grad_norm": 6.96875,
      "learning_rate": 5.01279319492992e-06,
      "loss": 1.60673504,
      "memory(GiB)": 117.38,
      "step": 41330,
      "train_speed(iter/s)": 1.636644
    },
    {
      "acc": 0.64284172,
      "epoch": 1.0485794013191274,
      "grad_norm": 5.40625,
      "learning_rate": 5.01174457440934e-06,
      "loss": 1.60006599,
      "memory(GiB)": 117.38,
      "step": 41335,
      "train_speed(iter/s)": 1.636665
    },
    {
      "acc": 0.67079792,
      "epoch": 1.0487062404870624,
      "grad_norm": 5.09375,
      "learning_rate": 5.010695953372179e-06,
      "loss": 1.55845718,
      "memory(GiB)": 117.38,
      "step": 41340,
      "train_speed(iter/s)": 1.636686
    },
    {
      "acc": 0.66199512,
      "epoch": 1.0488330796549974,
      "grad_norm": 5.8125,
      "learning_rate": 5.009647331864563e-06,
      "loss": 1.55456257,
      "memory(GiB)": 117.38,
      "step": 41345,
      "train_speed(iter/s)": 1.636707
    },
    {
      "acc": 0.66360178,
      "epoch": 1.0489599188229326,
      "grad_norm": 5.875,
      "learning_rate": 5.008598709932615e-06,
      "loss": 1.60426788,
      "memory(GiB)": 117.38,
      "step": 41350,
      "train_speed(iter/s)": 1.636728
    },
    {
      "acc": 0.65038514,
      "epoch": 1.0490867579908676,
      "grad_norm": 5.53125,
      "learning_rate": 5.007550087622456e-06,
      "loss": 1.64997749,
      "memory(GiB)": 117.38,
      "step": 41355,
      "train_speed(iter/s)": 1.636748
    },
    {
      "acc": 0.66603189,
      "epoch": 1.0492135971588026,
      "grad_norm": 6.90625,
      "learning_rate": 5.0065014649802124e-06,
      "loss": 1.55911579,
      "memory(GiB)": 117.38,
      "step": 41360,
      "train_speed(iter/s)": 1.636771
    },
    {
      "acc": 0.65281935,
      "epoch": 1.0493404363267378,
      "grad_norm": 5.1875,
      "learning_rate": 5.005452842052003e-06,
      "loss": 1.62591591,
      "memory(GiB)": 117.38,
      "step": 41365,
      "train_speed(iter/s)": 1.636791
    },
    {
      "acc": 0.65610533,
      "epoch": 1.0494672754946728,
      "grad_norm": 5.59375,
      "learning_rate": 5.004404218883955e-06,
      "loss": 1.61340828,
      "memory(GiB)": 117.38,
      "step": 41370,
      "train_speed(iter/s)": 1.636812
    },
    {
      "acc": 0.67170458,
      "epoch": 1.0495941146626078,
      "grad_norm": 6.375,
      "learning_rate": 5.0033555955221875e-06,
      "loss": 1.58470221,
      "memory(GiB)": 117.38,
      "step": 41375,
      "train_speed(iter/s)": 1.636833
    },
    {
      "acc": 0.66129222,
      "epoch": 1.0497209538305428,
      "grad_norm": 6.0,
      "learning_rate": 5.002306972012829e-06,
      "loss": 1.58265219,
      "memory(GiB)": 117.38,
      "step": 41380,
      "train_speed(iter/s)": 1.636853
    },
    {
      "acc": 0.65790405,
      "epoch": 1.049847792998478,
      "grad_norm": 4.875,
      "learning_rate": 5.001258348401998e-06,
      "loss": 1.59193287,
      "memory(GiB)": 117.38,
      "step": 41385,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.66113191,
      "epoch": 1.049974632166413,
      "grad_norm": 5.78125,
      "learning_rate": 5.000209724735819e-06,
      "loss": 1.61167622,
      "memory(GiB)": 117.38,
      "step": 41390,
      "train_speed(iter/s)": 1.636893
    },
    {
      "acc": 0.65508175,
      "epoch": 1.050101471334348,
      "grad_norm": 6.09375,
      "learning_rate": 4.999161101060416e-06,
      "loss": 1.62819176,
      "memory(GiB)": 117.38,
      "step": 41395,
      "train_speed(iter/s)": 1.636915
    },
    {
      "acc": 0.64078255,
      "epoch": 1.0502283105022832,
      "grad_norm": 6.78125,
      "learning_rate": 4.99811247742191e-06,
      "loss": 1.67368488,
      "memory(GiB)": 117.38,
      "step": 41400,
      "train_speed(iter/s)": 1.636936
    },
    {
      "acc": 0.65263281,
      "epoch": 1.0503551496702181,
      "grad_norm": 6.25,
      "learning_rate": 4.9970638538664275e-06,
      "loss": 1.61442719,
      "memory(GiB)": 117.38,
      "step": 41405,
      "train_speed(iter/s)": 1.636955
    },
    {
      "acc": 0.65208826,
      "epoch": 1.0504819888381531,
      "grad_norm": 4.84375,
      "learning_rate": 4.996015230440091e-06,
      "loss": 1.56453342,
      "memory(GiB)": 117.38,
      "step": 41410,
      "train_speed(iter/s)": 1.636975
    },
    {
      "acc": 0.66709089,
      "epoch": 1.0506088280060883,
      "grad_norm": 5.75,
      "learning_rate": 4.99496660718902e-06,
      "loss": 1.5060358,
      "memory(GiB)": 117.38,
      "step": 41415,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.65774541,
      "epoch": 1.0507356671740233,
      "grad_norm": 5.0625,
      "learning_rate": 4.99391798415934e-06,
      "loss": 1.5928319,
      "memory(GiB)": 117.38,
      "step": 41420,
      "train_speed(iter/s)": 1.637016
    },
    {
      "acc": 0.65854659,
      "epoch": 1.0508625063419583,
      "grad_norm": 6.0625,
      "learning_rate": 4.992869361397175e-06,
      "loss": 1.5839098,
      "memory(GiB)": 117.38,
      "step": 41425,
      "train_speed(iter/s)": 1.637038
    },
    {
      "acc": 0.65281224,
      "epoch": 1.0509893455098935,
      "grad_norm": 5.4375,
      "learning_rate": 4.991820738948649e-06,
      "loss": 1.5649519,
      "memory(GiB)": 117.38,
      "step": 41430,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.64837828,
      "epoch": 1.0511161846778285,
      "grad_norm": 6.65625,
      "learning_rate": 4.9907721168598805e-06,
      "loss": 1.6430418,
      "memory(GiB)": 117.38,
      "step": 41435,
      "train_speed(iter/s)": 1.637078
    },
    {
      "acc": 0.67129793,
      "epoch": 1.0512430238457635,
      "grad_norm": 5.0625,
      "learning_rate": 4.989723495176997e-06,
      "loss": 1.54342794,
      "memory(GiB)": 117.38,
      "step": 41440,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.67158484,
      "epoch": 1.0513698630136987,
      "grad_norm": 5.34375,
      "learning_rate": 4.988674873946118e-06,
      "loss": 1.53408966,
      "memory(GiB)": 117.38,
      "step": 41445,
      "train_speed(iter/s)": 1.637119
    },
    {
      "acc": 0.66758113,
      "epoch": 1.0514967021816337,
      "grad_norm": 6.875,
      "learning_rate": 4.987626253213373e-06,
      "loss": 1.57656155,
      "memory(GiB)": 117.38,
      "step": 41450,
      "train_speed(iter/s)": 1.637138
    },
    {
      "acc": 0.66543889,
      "epoch": 1.0516235413495687,
      "grad_norm": 5.90625,
      "learning_rate": 4.986577633024877e-06,
      "loss": 1.59522905,
      "memory(GiB)": 117.38,
      "step": 41455,
      "train_speed(iter/s)": 1.637158
    },
    {
      "acc": 0.64872437,
      "epoch": 1.051750380517504,
      "grad_norm": 5.96875,
      "learning_rate": 4.985529013426758e-06,
      "loss": 1.61923275,
      "memory(GiB)": 117.38,
      "step": 41460,
      "train_speed(iter/s)": 1.637179
    },
    {
      "acc": 0.65970845,
      "epoch": 1.051877219685439,
      "grad_norm": 5.5,
      "learning_rate": 4.984480394465136e-06,
      "loss": 1.54783516,
      "memory(GiB)": 117.38,
      "step": 41465,
      "train_speed(iter/s)": 1.637197
    },
    {
      "acc": 0.65071931,
      "epoch": 1.052004058853374,
      "grad_norm": 4.5,
      "learning_rate": 4.9834317761861385e-06,
      "loss": 1.59526424,
      "memory(GiB)": 117.38,
      "step": 41470,
      "train_speed(iter/s)": 1.637218
    },
    {
      "acc": 0.64619188,
      "epoch": 1.0521308980213089,
      "grad_norm": 6.125,
      "learning_rate": 4.982383158635884e-06,
      "loss": 1.61843891,
      "memory(GiB)": 117.38,
      "step": 41475,
      "train_speed(iter/s)": 1.637238
    },
    {
      "acc": 0.65058489,
      "epoch": 1.052257737189244,
      "grad_norm": 6.0,
      "learning_rate": 4.981334541860496e-06,
      "loss": 1.62189579,
      "memory(GiB)": 117.38,
      "step": 41480,
      "train_speed(iter/s)": 1.637258
    },
    {
      "acc": 0.65288968,
      "epoch": 1.052384576357179,
      "grad_norm": 6.09375,
      "learning_rate": 4.980285925906098e-06,
      "loss": 1.64552898,
      "memory(GiB)": 117.38,
      "step": 41485,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.64913392,
      "epoch": 1.052511415525114,
      "grad_norm": 6.09375,
      "learning_rate": 4.9792373108188155e-06,
      "loss": 1.63377304,
      "memory(GiB)": 117.38,
      "step": 41490,
      "train_speed(iter/s)": 1.6373
    },
    {
      "acc": 0.66187911,
      "epoch": 1.0526382546930493,
      "grad_norm": 7.625,
      "learning_rate": 4.978188696644767e-06,
      "loss": 1.55747643,
      "memory(GiB)": 117.38,
      "step": 41495,
      "train_speed(iter/s)": 1.637319
    },
    {
      "acc": 0.6660954,
      "epoch": 1.0527650938609843,
      "grad_norm": 5.15625,
      "learning_rate": 4.977140083430075e-06,
      "loss": 1.54084549,
      "memory(GiB)": 117.38,
      "step": 41500,
      "train_speed(iter/s)": 1.637339
    },
    {
      "acc": 0.66575518,
      "epoch": 1.0528919330289193,
      "grad_norm": 4.46875,
      "learning_rate": 4.976091471220867e-06,
      "loss": 1.5681531,
      "memory(GiB)": 117.38,
      "step": 41505,
      "train_speed(iter/s)": 1.637359
    },
    {
      "acc": 0.65210896,
      "epoch": 1.0530187721968545,
      "grad_norm": 4.84375,
      "learning_rate": 4.975042860063263e-06,
      "loss": 1.61182365,
      "memory(GiB)": 117.38,
      "step": 41510,
      "train_speed(iter/s)": 1.63738
    },
    {
      "acc": 0.6524231,
      "epoch": 1.0531456113647895,
      "grad_norm": 5.4375,
      "learning_rate": 4.973994250003384e-06,
      "loss": 1.63421822,
      "memory(GiB)": 117.38,
      "step": 41515,
      "train_speed(iter/s)": 1.6374
    },
    {
      "acc": 0.64087076,
      "epoch": 1.0532724505327244,
      "grad_norm": 5.59375,
      "learning_rate": 4.972945641087355e-06,
      "loss": 1.69444199,
      "memory(GiB)": 117.38,
      "step": 41520,
      "train_speed(iter/s)": 1.63742
    },
    {
      "acc": 0.65947309,
      "epoch": 1.0533992897006597,
      "grad_norm": 5.21875,
      "learning_rate": 4.9718970333612955e-06,
      "loss": 1.5772234,
      "memory(GiB)": 117.38,
      "step": 41525,
      "train_speed(iter/s)": 1.63744
    },
    {
      "acc": 0.65320401,
      "epoch": 1.0535261288685946,
      "grad_norm": 6.0625,
      "learning_rate": 4.970848426871333e-06,
      "loss": 1.58386173,
      "memory(GiB)": 117.38,
      "step": 41530,
      "train_speed(iter/s)": 1.637461
    },
    {
      "acc": 0.64789886,
      "epoch": 1.0536529680365296,
      "grad_norm": 5.375,
      "learning_rate": 4.9697998216635854e-06,
      "loss": 1.64066391,
      "memory(GiB)": 117.38,
      "step": 41535,
      "train_speed(iter/s)": 1.63748
    },
    {
      "acc": 0.65942602,
      "epoch": 1.0537798072044646,
      "grad_norm": 5.0,
      "learning_rate": 4.9687512177841765e-06,
      "loss": 1.55954399,
      "memory(GiB)": 117.38,
      "step": 41540,
      "train_speed(iter/s)": 1.637501
    },
    {
      "acc": 0.64906888,
      "epoch": 1.0539066463723998,
      "grad_norm": 5.28125,
      "learning_rate": 4.967702615279227e-06,
      "loss": 1.62640915,
      "memory(GiB)": 117.38,
      "step": 41545,
      "train_speed(iter/s)": 1.637519
    },
    {
      "acc": 0.64940128,
      "epoch": 1.0540334855403348,
      "grad_norm": 5.5,
      "learning_rate": 4.966654014194863e-06,
      "loss": 1.65984116,
      "memory(GiB)": 117.38,
      "step": 41550,
      "train_speed(iter/s)": 1.637539
    },
    {
      "acc": 0.66228552,
      "epoch": 1.0541603247082698,
      "grad_norm": 5.0,
      "learning_rate": 4.965605414577204e-06,
      "loss": 1.50570974,
      "memory(GiB)": 117.38,
      "step": 41555,
      "train_speed(iter/s)": 1.637557
    },
    {
      "acc": 0.67069626,
      "epoch": 1.054287163876205,
      "grad_norm": 5.78125,
      "learning_rate": 4.964556816472371e-06,
      "loss": 1.5725421,
      "memory(GiB)": 117.38,
      "step": 41560,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.6603281,
      "epoch": 1.05441400304414,
      "grad_norm": 4.6875,
      "learning_rate": 4.9635082199264874e-06,
      "loss": 1.62031517,
      "memory(GiB)": 117.38,
      "step": 41565,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.66995239,
      "epoch": 1.054540842212075,
      "grad_norm": 5.3125,
      "learning_rate": 4.962459624985677e-06,
      "loss": 1.58806696,
      "memory(GiB)": 117.38,
      "step": 41570,
      "train_speed(iter/s)": 1.637616
    },
    {
      "acc": 0.65389962,
      "epoch": 1.0546676813800102,
      "grad_norm": 6.0625,
      "learning_rate": 4.961411031696059e-06,
      "loss": 1.63474693,
      "memory(GiB)": 117.38,
      "step": 41575,
      "train_speed(iter/s)": 1.637635
    },
    {
      "acc": 0.66805277,
      "epoch": 1.0547945205479452,
      "grad_norm": 6.53125,
      "learning_rate": 4.960362440103756e-06,
      "loss": 1.60070915,
      "memory(GiB)": 117.38,
      "step": 41580,
      "train_speed(iter/s)": 1.637656
    },
    {
      "acc": 0.67776632,
      "epoch": 1.0549213597158802,
      "grad_norm": 6.84375,
      "learning_rate": 4.95931385025489e-06,
      "loss": 1.51767006,
      "memory(GiB)": 117.38,
      "step": 41585,
      "train_speed(iter/s)": 1.637677
    },
    {
      "acc": 0.66229258,
      "epoch": 1.0550481988838154,
      "grad_norm": 5.375,
      "learning_rate": 4.958265262195584e-06,
      "loss": 1.58374205,
      "memory(GiB)": 117.38,
      "step": 41590,
      "train_speed(iter/s)": 1.637698
    },
    {
      "acc": 0.67434092,
      "epoch": 1.0551750380517504,
      "grad_norm": 5.84375,
      "learning_rate": 4.957216675971955e-06,
      "loss": 1.57769642,
      "memory(GiB)": 117.38,
      "step": 41595,
      "train_speed(iter/s)": 1.637718
    },
    {
      "acc": 0.65956807,
      "epoch": 1.0553018772196854,
      "grad_norm": 6.25,
      "learning_rate": 4.9561680916301295e-06,
      "loss": 1.65626869,
      "memory(GiB)": 117.38,
      "step": 41600,
      "train_speed(iter/s)": 1.637737
    },
    {
      "acc": 0.65970173,
      "epoch": 1.0554287163876206,
      "grad_norm": 5.90625,
      "learning_rate": 4.955119509216226e-06,
      "loss": 1.56891727,
      "memory(GiB)": 117.38,
      "step": 41605,
      "train_speed(iter/s)": 1.637755
    },
    {
      "acc": 0.66244111,
      "epoch": 1.0555555555555556,
      "grad_norm": 5.5625,
      "learning_rate": 4.9540709287763685e-06,
      "loss": 1.6115097,
      "memory(GiB)": 117.38,
      "step": 41610,
      "train_speed(iter/s)": 1.637776
    },
    {
      "acc": 0.65042858,
      "epoch": 1.0556823947234906,
      "grad_norm": 7.46875,
      "learning_rate": 4.953022350356676e-06,
      "loss": 1.6223484,
      "memory(GiB)": 117.38,
      "step": 41615,
      "train_speed(iter/s)": 1.637797
    },
    {
      "acc": 0.65889292,
      "epoch": 1.0558092338914258,
      "grad_norm": 7.25,
      "learning_rate": 4.951973774003269e-06,
      "loss": 1.61314983,
      "memory(GiB)": 117.38,
      "step": 41620,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.65645108,
      "epoch": 1.0559360730593608,
      "grad_norm": 6.4375,
      "learning_rate": 4.950925199762271e-06,
      "loss": 1.64923573,
      "memory(GiB)": 117.38,
      "step": 41625,
      "train_speed(iter/s)": 1.637838
    },
    {
      "acc": 0.65752707,
      "epoch": 1.0560629122272958,
      "grad_norm": 6.1875,
      "learning_rate": 4.949876627679803e-06,
      "loss": 1.61506424,
      "memory(GiB)": 117.38,
      "step": 41630,
      "train_speed(iter/s)": 1.637857
    },
    {
      "acc": 0.66689143,
      "epoch": 1.0561897513952307,
      "grad_norm": 5.75,
      "learning_rate": 4.948828057801983e-06,
      "loss": 1.57217751,
      "memory(GiB)": 117.38,
      "step": 41635,
      "train_speed(iter/s)": 1.637877
    },
    {
      "acc": 0.64063044,
      "epoch": 1.056316590563166,
      "grad_norm": 6.34375,
      "learning_rate": 4.947779490174933e-06,
      "loss": 1.64298973,
      "memory(GiB)": 117.38,
      "step": 41640,
      "train_speed(iter/s)": 1.637898
    },
    {
      "acc": 0.65305147,
      "epoch": 1.056443429731101,
      "grad_norm": 6.25,
      "learning_rate": 4.946730924844775e-06,
      "loss": 1.6481638,
      "memory(GiB)": 117.38,
      "step": 41645,
      "train_speed(iter/s)": 1.637917
    },
    {
      "acc": 0.64655704,
      "epoch": 1.056570268899036,
      "grad_norm": 5.71875,
      "learning_rate": 4.945682361857631e-06,
      "loss": 1.61778393,
      "memory(GiB)": 117.38,
      "step": 41650,
      "train_speed(iter/s)": 1.637937
    },
    {
      "acc": 0.66422472,
      "epoch": 1.0566971080669711,
      "grad_norm": 5.71875,
      "learning_rate": 4.944633801259615e-06,
      "loss": 1.58619061,
      "memory(GiB)": 117.38,
      "step": 41655,
      "train_speed(iter/s)": 1.637954
    },
    {
      "acc": 0.63881536,
      "epoch": 1.0568239472349061,
      "grad_norm": 7.65625,
      "learning_rate": 4.943585243096854e-06,
      "loss": 1.67109737,
      "memory(GiB)": 117.38,
      "step": 41660,
      "train_speed(iter/s)": 1.637975
    },
    {
      "acc": 0.66965723,
      "epoch": 1.0569507864028411,
      "grad_norm": 5.34375,
      "learning_rate": 4.942536687415465e-06,
      "loss": 1.53907242,
      "memory(GiB)": 117.38,
      "step": 41665,
      "train_speed(iter/s)": 1.637996
    },
    {
      "acc": 0.63948097,
      "epoch": 1.0570776255707763,
      "grad_norm": 6.0625,
      "learning_rate": 4.941488134261571e-06,
      "loss": 1.69211578,
      "memory(GiB)": 117.38,
      "step": 41670,
      "train_speed(iter/s)": 1.638016
    },
    {
      "acc": 0.66704359,
      "epoch": 1.0572044647387113,
      "grad_norm": 6.21875,
      "learning_rate": 4.940439583681288e-06,
      "loss": 1.47416096,
      "memory(GiB)": 117.38,
      "step": 41675,
      "train_speed(iter/s)": 1.638034
    },
    {
      "acc": 0.65645337,
      "epoch": 1.0573313039066463,
      "grad_norm": 5.71875,
      "learning_rate": 4.939391035720739e-06,
      "loss": 1.54929352,
      "memory(GiB)": 117.38,
      "step": 41680,
      "train_speed(iter/s)": 1.638053
    },
    {
      "acc": 0.67585878,
      "epoch": 1.0574581430745815,
      "grad_norm": 5.78125,
      "learning_rate": 4.938342490426041e-06,
      "loss": 1.5336607,
      "memory(GiB)": 117.38,
      "step": 41685,
      "train_speed(iter/s)": 1.638071
    },
    {
      "acc": 0.65449138,
      "epoch": 1.0575849822425165,
      "grad_norm": 6.46875,
      "learning_rate": 4.937293947843318e-06,
      "loss": 1.60138302,
      "memory(GiB)": 117.38,
      "step": 41690,
      "train_speed(iter/s)": 1.638091
    },
    {
      "acc": 0.663904,
      "epoch": 1.0577118214104515,
      "grad_norm": 4.71875,
      "learning_rate": 4.936245408018687e-06,
      "loss": 1.5568924,
      "memory(GiB)": 117.38,
      "step": 41695,
      "train_speed(iter/s)": 1.63811
    },
    {
      "acc": 0.65379052,
      "epoch": 1.0578386605783865,
      "grad_norm": 5.84375,
      "learning_rate": 4.935196870998265e-06,
      "loss": 1.64195786,
      "memory(GiB)": 117.38,
      "step": 41700,
      "train_speed(iter/s)": 1.638129
    },
    {
      "acc": 0.654774,
      "epoch": 1.0579654997463217,
      "grad_norm": 15.6875,
      "learning_rate": 4.934148336828176e-06,
      "loss": 1.6594841,
      "memory(GiB)": 117.38,
      "step": 41705,
      "train_speed(iter/s)": 1.638149
    },
    {
      "acc": 0.63969841,
      "epoch": 1.0580923389142567,
      "grad_norm": 6.8125,
      "learning_rate": 4.933099805554538e-06,
      "loss": 1.7236372,
      "memory(GiB)": 117.38,
      "step": 41710,
      "train_speed(iter/s)": 1.63817
    },
    {
      "acc": 0.65071011,
      "epoch": 1.0582191780821917,
      "grad_norm": 5.4375,
      "learning_rate": 4.932051277223468e-06,
      "loss": 1.65656586,
      "memory(GiB)": 117.38,
      "step": 41715,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.66019106,
      "epoch": 1.058346017250127,
      "grad_norm": 6.3125,
      "learning_rate": 4.931002751881086e-06,
      "loss": 1.60512905,
      "memory(GiB)": 117.38,
      "step": 41720,
      "train_speed(iter/s)": 1.638206
    },
    {
      "acc": 0.65448594,
      "epoch": 1.0584728564180619,
      "grad_norm": 9.125,
      "learning_rate": 4.929954229573512e-06,
      "loss": 1.60945301,
      "memory(GiB)": 117.38,
      "step": 41725,
      "train_speed(iter/s)": 1.638227
    },
    {
      "acc": 0.65172033,
      "epoch": 1.0585996955859969,
      "grad_norm": 5.28125,
      "learning_rate": 4.9289057103468635e-06,
      "loss": 1.57258883,
      "memory(GiB)": 117.38,
      "step": 41730,
      "train_speed(iter/s)": 1.638246
    },
    {
      "acc": 0.65867891,
      "epoch": 1.058726534753932,
      "grad_norm": 5.125,
      "learning_rate": 4.927857194247258e-06,
      "loss": 1.56267366,
      "memory(GiB)": 117.38,
      "step": 41735,
      "train_speed(iter/s)": 1.638265
    },
    {
      "acc": 0.63669672,
      "epoch": 1.058853373921867,
      "grad_norm": 5.34375,
      "learning_rate": 4.926808681320816e-06,
      "loss": 1.64934692,
      "memory(GiB)": 117.38,
      "step": 41740,
      "train_speed(iter/s)": 1.638286
    },
    {
      "acc": 0.63982091,
      "epoch": 1.058980213089802,
      "grad_norm": 7.4375,
      "learning_rate": 4.925760171613654e-06,
      "loss": 1.65690918,
      "memory(GiB)": 117.38,
      "step": 41745,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.64528608,
      "epoch": 1.0591070522577373,
      "grad_norm": 5.90625,
      "learning_rate": 4.9247116651718925e-06,
      "loss": 1.64378071,
      "memory(GiB)": 117.38,
      "step": 41750,
      "train_speed(iter/s)": 1.638325
    },
    {
      "acc": 0.65477071,
      "epoch": 1.0592338914256723,
      "grad_norm": 7.9375,
      "learning_rate": 4.9236631620416486e-06,
      "loss": 1.66486778,
      "memory(GiB)": 117.38,
      "step": 41755,
      "train_speed(iter/s)": 1.638344
    },
    {
      "acc": 0.65830355,
      "epoch": 1.0593607305936072,
      "grad_norm": 5.09375,
      "learning_rate": 4.922614662269038e-06,
      "loss": 1.61162148,
      "memory(GiB)": 117.38,
      "step": 41760,
      "train_speed(iter/s)": 1.638363
    },
    {
      "acc": 0.66571188,
      "epoch": 1.0594875697615425,
      "grad_norm": 5.0625,
      "learning_rate": 4.9215661659001805e-06,
      "loss": 1.54778767,
      "memory(GiB)": 117.38,
      "step": 41765,
      "train_speed(iter/s)": 1.638382
    },
    {
      "acc": 0.68263235,
      "epoch": 1.0596144089294774,
      "grad_norm": 6.28125,
      "learning_rate": 4.920517672981195e-06,
      "loss": 1.49134645,
      "memory(GiB)": 117.38,
      "step": 41770,
      "train_speed(iter/s)": 1.638401
    },
    {
      "acc": 0.65683346,
      "epoch": 1.0597412480974124,
      "grad_norm": 5.65625,
      "learning_rate": 4.919469183558195e-06,
      "loss": 1.60307827,
      "memory(GiB)": 117.38,
      "step": 41775,
      "train_speed(iter/s)": 1.63842
    },
    {
      "acc": 0.66316805,
      "epoch": 1.0598680872653476,
      "grad_norm": 5.28125,
      "learning_rate": 4.9184206976773e-06,
      "loss": 1.57670088,
      "memory(GiB)": 117.38,
      "step": 41780,
      "train_speed(iter/s)": 1.638439
    },
    {
      "acc": 0.65350456,
      "epoch": 1.0599949264332826,
      "grad_norm": 6.5,
      "learning_rate": 4.917372215384627e-06,
      "loss": 1.66546021,
      "memory(GiB)": 117.38,
      "step": 41785,
      "train_speed(iter/s)": 1.638458
    },
    {
      "acc": 0.65337143,
      "epoch": 1.0601217656012176,
      "grad_norm": 6.28125,
      "learning_rate": 4.916323736726295e-06,
      "loss": 1.57240086,
      "memory(GiB)": 117.38,
      "step": 41790,
      "train_speed(iter/s)": 1.638478
    },
    {
      "acc": 0.65314188,
      "epoch": 1.0602486047691526,
      "grad_norm": 5.40625,
      "learning_rate": 4.9152752617484156e-06,
      "loss": 1.60025139,
      "memory(GiB)": 117.38,
      "step": 41795,
      "train_speed(iter/s)": 1.638498
    },
    {
      "acc": 0.64789534,
      "epoch": 1.0603754439370878,
      "grad_norm": 5.6875,
      "learning_rate": 4.91422679049711e-06,
      "loss": 1.62481232,
      "memory(GiB)": 117.38,
      "step": 41800,
      "train_speed(iter/s)": 1.638517
    },
    {
      "acc": 0.66380148,
      "epoch": 1.0605022831050228,
      "grad_norm": 5.59375,
      "learning_rate": 4.913178323018493e-06,
      "loss": 1.54954538,
      "memory(GiB)": 117.38,
      "step": 41805,
      "train_speed(iter/s)": 1.638536
    },
    {
      "acc": 0.65828438,
      "epoch": 1.0606291222729578,
      "grad_norm": 7.71875,
      "learning_rate": 4.912129859358682e-06,
      "loss": 1.6011261,
      "memory(GiB)": 117.38,
      "step": 41810,
      "train_speed(iter/s)": 1.638555
    },
    {
      "acc": 0.66674891,
      "epoch": 1.060755961440893,
      "grad_norm": 6.25,
      "learning_rate": 4.9110813995637905e-06,
      "loss": 1.55645895,
      "memory(GiB)": 117.38,
      "step": 41815,
      "train_speed(iter/s)": 1.638574
    },
    {
      "acc": 0.65907483,
      "epoch": 1.060882800608828,
      "grad_norm": 5.09375,
      "learning_rate": 4.910032943679936e-06,
      "loss": 1.56672668,
      "memory(GiB)": 117.38,
      "step": 41820,
      "train_speed(iter/s)": 1.638593
    },
    {
      "acc": 0.66251602,
      "epoch": 1.061009639776763,
      "grad_norm": 6.125,
      "learning_rate": 4.908984491753234e-06,
      "loss": 1.56733284,
      "memory(GiB)": 117.38,
      "step": 41825,
      "train_speed(iter/s)": 1.638613
    },
    {
      "acc": 0.6519701,
      "epoch": 1.0611364789446982,
      "grad_norm": 6.125,
      "learning_rate": 4.907936043829802e-06,
      "loss": 1.61102676,
      "memory(GiB)": 117.38,
      "step": 41830,
      "train_speed(iter/s)": 1.638632
    },
    {
      "acc": 0.65537128,
      "epoch": 1.0612633181126332,
      "grad_norm": 5.40625,
      "learning_rate": 4.906887599955754e-06,
      "loss": 1.59698248,
      "memory(GiB)": 117.38,
      "step": 41835,
      "train_speed(iter/s)": 1.638649
    },
    {
      "acc": 0.66442413,
      "epoch": 1.0613901572805682,
      "grad_norm": 6.0625,
      "learning_rate": 4.905839160177203e-06,
      "loss": 1.52137699,
      "memory(GiB)": 117.38,
      "step": 41840,
      "train_speed(iter/s)": 1.638669
    },
    {
      "acc": 0.65538521,
      "epoch": 1.0615169964485034,
      "grad_norm": 6.34375,
      "learning_rate": 4.904790724540267e-06,
      "loss": 1.57740974,
      "memory(GiB)": 117.38,
      "step": 41845,
      "train_speed(iter/s)": 1.638687
    },
    {
      "acc": 0.66363306,
      "epoch": 1.0616438356164384,
      "grad_norm": 5.9375,
      "learning_rate": 4.903742293091061e-06,
      "loss": 1.56996784,
      "memory(GiB)": 117.38,
      "step": 41850,
      "train_speed(iter/s)": 1.638707
    },
    {
      "acc": 0.65466814,
      "epoch": 1.0617706747843734,
      "grad_norm": 5.4375,
      "learning_rate": 4.902693865875698e-06,
      "loss": 1.63040752,
      "memory(GiB)": 117.38,
      "step": 41855,
      "train_speed(iter/s)": 1.638727
    },
    {
      "acc": 0.65475426,
      "epoch": 1.0618975139523084,
      "grad_norm": 6.46875,
      "learning_rate": 4.901645442940293e-06,
      "loss": 1.60180779,
      "memory(GiB)": 117.38,
      "step": 41860,
      "train_speed(iter/s)": 1.638746
    },
    {
      "acc": 0.66409035,
      "epoch": 1.0620243531202436,
      "grad_norm": 5.59375,
      "learning_rate": 4.900597024330961e-06,
      "loss": 1.58837566,
      "memory(GiB)": 117.38,
      "step": 41865,
      "train_speed(iter/s)": 1.638764
    },
    {
      "acc": 0.65501347,
      "epoch": 1.0621511922881786,
      "grad_norm": 6.5625,
      "learning_rate": 4.899548610093816e-06,
      "loss": 1.63757057,
      "memory(GiB)": 117.38,
      "step": 41870,
      "train_speed(iter/s)": 1.638784
    },
    {
      "acc": 0.66580958,
      "epoch": 1.0622780314561135,
      "grad_norm": 8.5,
      "learning_rate": 4.89850020027497e-06,
      "loss": 1.56432114,
      "memory(GiB)": 117.38,
      "step": 41875,
      "train_speed(iter/s)": 1.638803
    },
    {
      "acc": 0.66607304,
      "epoch": 1.0624048706240488,
      "grad_norm": 5.8125,
      "learning_rate": 4.89745179492054e-06,
      "loss": 1.56290388,
      "memory(GiB)": 117.38,
      "step": 41880,
      "train_speed(iter/s)": 1.63882
    },
    {
      "acc": 0.65710521,
      "epoch": 1.0625317097919837,
      "grad_norm": 6.625,
      "learning_rate": 4.896403394076636e-06,
      "loss": 1.61378155,
      "memory(GiB)": 117.38,
      "step": 41885,
      "train_speed(iter/s)": 1.638839
    },
    {
      "acc": 0.65485053,
      "epoch": 1.0626585489599187,
      "grad_norm": 9.0625,
      "learning_rate": 4.895354997789377e-06,
      "loss": 1.65956841,
      "memory(GiB)": 117.38,
      "step": 41890,
      "train_speed(iter/s)": 1.638858
    },
    {
      "acc": 0.66044173,
      "epoch": 1.062785388127854,
      "grad_norm": 5.5,
      "learning_rate": 4.894306606104869e-06,
      "loss": 1.59045801,
      "memory(GiB)": 117.38,
      "step": 41895,
      "train_speed(iter/s)": 1.638877
    },
    {
      "acc": 0.66803007,
      "epoch": 1.062912227295789,
      "grad_norm": 5.0,
      "learning_rate": 4.893258219069229e-06,
      "loss": 1.57747107,
      "memory(GiB)": 117.38,
      "step": 41900,
      "train_speed(iter/s)": 1.638895
    },
    {
      "acc": 0.66822958,
      "epoch": 1.063039066463724,
      "grad_norm": 5.6875,
      "learning_rate": 4.892209836728569e-06,
      "loss": 1.5444109,
      "memory(GiB)": 117.38,
      "step": 41905,
      "train_speed(iter/s)": 1.638913
    },
    {
      "acc": 0.65718756,
      "epoch": 1.0631659056316591,
      "grad_norm": 5.34375,
      "learning_rate": 4.891161459129003e-06,
      "loss": 1.61944962,
      "memory(GiB)": 117.38,
      "step": 41910,
      "train_speed(iter/s)": 1.638932
    },
    {
      "acc": 0.66516099,
      "epoch": 1.0632927447995941,
      "grad_norm": 6.78125,
      "learning_rate": 4.890113086316641e-06,
      "loss": 1.58834476,
      "memory(GiB)": 117.38,
      "step": 41915,
      "train_speed(iter/s)": 1.63895
    },
    {
      "acc": 0.65930672,
      "epoch": 1.0634195839675291,
      "grad_norm": 5.03125,
      "learning_rate": 4.889064718337595e-06,
      "loss": 1.58916779,
      "memory(GiB)": 117.38,
      "step": 41920,
      "train_speed(iter/s)": 1.638971
    },
    {
      "acc": 0.66039019,
      "epoch": 1.0635464231354643,
      "grad_norm": 4.78125,
      "learning_rate": 4.888016355237979e-06,
      "loss": 1.58848038,
      "memory(GiB)": 117.38,
      "step": 41925,
      "train_speed(iter/s)": 1.63899
    },
    {
      "acc": 0.64981003,
      "epoch": 1.0636732623033993,
      "grad_norm": 12.3125,
      "learning_rate": 4.886967997063905e-06,
      "loss": 1.58340187,
      "memory(GiB)": 117.38,
      "step": 41930,
      "train_speed(iter/s)": 1.63901
    },
    {
      "acc": 0.65391531,
      "epoch": 1.0638001014713343,
      "grad_norm": 6.34375,
      "learning_rate": 4.885919643861482e-06,
      "loss": 1.52405014,
      "memory(GiB)": 117.38,
      "step": 41935,
      "train_speed(iter/s)": 1.639031
    },
    {
      "acc": 0.67146859,
      "epoch": 1.0639269406392695,
      "grad_norm": 7.78125,
      "learning_rate": 4.884871295676821e-06,
      "loss": 1.48230019,
      "memory(GiB)": 117.38,
      "step": 41940,
      "train_speed(iter/s)": 1.639049
    },
    {
      "acc": 0.66181154,
      "epoch": 1.0640537798072045,
      "grad_norm": 6.1875,
      "learning_rate": 4.883822952556036e-06,
      "loss": 1.57460527,
      "memory(GiB)": 117.38,
      "step": 41945,
      "train_speed(iter/s)": 1.639069
    },
    {
      "acc": 0.65440249,
      "epoch": 1.0641806189751395,
      "grad_norm": 6.0625,
      "learning_rate": 4.882774614545237e-06,
      "loss": 1.54522963,
      "memory(GiB)": 117.38,
      "step": 41950,
      "train_speed(iter/s)": 1.639088
    },
    {
      "acc": 0.65802555,
      "epoch": 1.0643074581430745,
      "grad_norm": 6.28125,
      "learning_rate": 4.881726281690531e-06,
      "loss": 1.620014,
      "memory(GiB)": 117.38,
      "step": 41955,
      "train_speed(iter/s)": 1.639107
    },
    {
      "acc": 0.64371748,
      "epoch": 1.0644342973110097,
      "grad_norm": 6.84375,
      "learning_rate": 4.8806779540380335e-06,
      "loss": 1.65082741,
      "memory(GiB)": 117.38,
      "step": 41960,
      "train_speed(iter/s)": 1.639128
    },
    {
      "acc": 0.67126255,
      "epoch": 1.0645611364789447,
      "grad_norm": 4.875,
      "learning_rate": 4.879629631633851e-06,
      "loss": 1.58161755,
      "memory(GiB)": 117.38,
      "step": 41965,
      "train_speed(iter/s)": 1.639148
    },
    {
      "acc": 0.64680166,
      "epoch": 1.0646879756468797,
      "grad_norm": 7.4375,
      "learning_rate": 4.8785813145240965e-06,
      "loss": 1.61356754,
      "memory(GiB)": 117.38,
      "step": 41970,
      "train_speed(iter/s)": 1.639168
    },
    {
      "acc": 0.65687032,
      "epoch": 1.0648148148148149,
      "grad_norm": 5.625,
      "learning_rate": 4.877533002754877e-06,
      "loss": 1.58165016,
      "memory(GiB)": 117.38,
      "step": 41975,
      "train_speed(iter/s)": 1.639189
    },
    {
      "acc": 0.63934741,
      "epoch": 1.0649416539827499,
      "grad_norm": 5.03125,
      "learning_rate": 4.8764846963723025e-06,
      "loss": 1.66624241,
      "memory(GiB)": 117.38,
      "step": 41980,
      "train_speed(iter/s)": 1.63921
    },
    {
      "acc": 0.65329185,
      "epoch": 1.0650684931506849,
      "grad_norm": 6.8125,
      "learning_rate": 4.875436395422481e-06,
      "loss": 1.60341434,
      "memory(GiB)": 117.38,
      "step": 41985,
      "train_speed(iter/s)": 1.639229
    },
    {
      "acc": 0.65343442,
      "epoch": 1.06519533231862,
      "grad_norm": 5.375,
      "learning_rate": 4.874388099951527e-06,
      "loss": 1.57683716,
      "memory(GiB)": 117.38,
      "step": 41990,
      "train_speed(iter/s)": 1.639249
    },
    {
      "acc": 0.68604231,
      "epoch": 1.065322171486555,
      "grad_norm": 5.09375,
      "learning_rate": 4.873339810005543e-06,
      "loss": 1.5390049,
      "memory(GiB)": 117.38,
      "step": 41995,
      "train_speed(iter/s)": 1.639268
    },
    {
      "acc": 0.64881458,
      "epoch": 1.06544901065449,
      "grad_norm": 6.25,
      "learning_rate": 4.872291525630638e-06,
      "loss": 1.67170715,
      "memory(GiB)": 117.38,
      "step": 42000,
      "train_speed(iter/s)": 1.639288
    },
    {
      "epoch": 1.06544901065449,
      "eval_acc": 0.6461908252359264,
      "eval_loss": 1.5735667943954468,
      "eval_runtime": 58.6735,
      "eval_samples_per_second": 108.567,
      "eval_steps_per_second": 27.15,
      "step": 42000
    },
    {
      "acc": 0.66267004,
      "epoch": 1.0655758498224253,
      "grad_norm": 6.15625,
      "learning_rate": 4.871243246872923e-06,
      "loss": 1.5742548,
      "memory(GiB)": 117.38,
      "step": 42005,
      "train_speed(iter/s)": 1.635292
    },
    {
      "acc": 0.65942197,
      "epoch": 1.0657026889903602,
      "grad_norm": 5.75,
      "learning_rate": 4.870194973778506e-06,
      "loss": 1.58884411,
      "memory(GiB)": 117.38,
      "step": 42010,
      "train_speed(iter/s)": 1.635312
    },
    {
      "acc": 0.64643865,
      "epoch": 1.0658295281582952,
      "grad_norm": 5.59375,
      "learning_rate": 4.869146706393493e-06,
      "loss": 1.65027313,
      "memory(GiB)": 117.38,
      "step": 42015,
      "train_speed(iter/s)": 1.635334
    },
    {
      "acc": 0.65906415,
      "epoch": 1.0659563673262302,
      "grad_norm": 6.25,
      "learning_rate": 4.868098444763991e-06,
      "loss": 1.54234753,
      "memory(GiB)": 117.38,
      "step": 42020,
      "train_speed(iter/s)": 1.635351
    },
    {
      "acc": 0.65209513,
      "epoch": 1.0660832064941654,
      "grad_norm": 8.375,
      "learning_rate": 4.86705018893611e-06,
      "loss": 1.59075565,
      "memory(GiB)": 117.38,
      "step": 42025,
      "train_speed(iter/s)": 1.635374
    },
    {
      "acc": 0.65900273,
      "epoch": 1.0662100456621004,
      "grad_norm": 5.09375,
      "learning_rate": 4.866001938955955e-06,
      "loss": 1.57277384,
      "memory(GiB)": 117.38,
      "step": 42030,
      "train_speed(iter/s)": 1.635395
    },
    {
      "acc": 0.6565896,
      "epoch": 1.0663368848300354,
      "grad_norm": 4.8125,
      "learning_rate": 4.864953694869632e-06,
      "loss": 1.57699404,
      "memory(GiB)": 117.38,
      "step": 42035,
      "train_speed(iter/s)": 1.635417
    },
    {
      "acc": 0.66661692,
      "epoch": 1.0664637239979706,
      "grad_norm": 8.5625,
      "learning_rate": 4.863905456723249e-06,
      "loss": 1.61874542,
      "memory(GiB)": 117.38,
      "step": 42040,
      "train_speed(iter/s)": 1.635439
    },
    {
      "acc": 0.6373539,
      "epoch": 1.0665905631659056,
      "grad_norm": 5.84375,
      "learning_rate": 4.8628572245629105e-06,
      "loss": 1.68152905,
      "memory(GiB)": 117.38,
      "step": 42045,
      "train_speed(iter/s)": 1.63546
    },
    {
      "acc": 0.64584494,
      "epoch": 1.0667174023338406,
      "grad_norm": 6.46875,
      "learning_rate": 4.861808998434726e-06,
      "loss": 1.60825424,
      "memory(GiB)": 117.38,
      "step": 42050,
      "train_speed(iter/s)": 1.635481
    },
    {
      "acc": 0.66562757,
      "epoch": 1.0668442415017758,
      "grad_norm": 6.8125,
      "learning_rate": 4.860760778384797e-06,
      "loss": 1.51508074,
      "memory(GiB)": 117.38,
      "step": 42055,
      "train_speed(iter/s)": 1.635504
    },
    {
      "acc": 0.65530353,
      "epoch": 1.0669710806697108,
      "grad_norm": 5.4375,
      "learning_rate": 4.85971256445923e-06,
      "loss": 1.54585171,
      "memory(GiB)": 117.38,
      "step": 42060,
      "train_speed(iter/s)": 1.635524
    },
    {
      "acc": 0.65522218,
      "epoch": 1.0670979198376458,
      "grad_norm": 6.65625,
      "learning_rate": 4.858664356704131e-06,
      "loss": 1.62526665,
      "memory(GiB)": 117.38,
      "step": 42065,
      "train_speed(iter/s)": 1.635545
    },
    {
      "acc": 0.68565059,
      "epoch": 1.067224759005581,
      "grad_norm": 6.71875,
      "learning_rate": 4.857616155165606e-06,
      "loss": 1.43904161,
      "memory(GiB)": 117.38,
      "step": 42070,
      "train_speed(iter/s)": 1.635567
    },
    {
      "acc": 0.631318,
      "epoch": 1.067351598173516,
      "grad_norm": 5.03125,
      "learning_rate": 4.856567959889758e-06,
      "loss": 1.66722412,
      "memory(GiB)": 117.38,
      "step": 42075,
      "train_speed(iter/s)": 1.635589
    },
    {
      "acc": 0.65941119,
      "epoch": 1.067478437341451,
      "grad_norm": 5.90625,
      "learning_rate": 4.855519770922691e-06,
      "loss": 1.62293243,
      "memory(GiB)": 117.38,
      "step": 42080,
      "train_speed(iter/s)": 1.63561
    },
    {
      "acc": 0.64963002,
      "epoch": 1.0676052765093862,
      "grad_norm": 6.625,
      "learning_rate": 4.8544715883105084e-06,
      "loss": 1.59394016,
      "memory(GiB)": 117.38,
      "step": 42085,
      "train_speed(iter/s)": 1.635632
    },
    {
      "acc": 0.65316563,
      "epoch": 1.0677321156773212,
      "grad_norm": 6.3125,
      "learning_rate": 4.853423412099318e-06,
      "loss": 1.61296577,
      "memory(GiB)": 117.38,
      "step": 42090,
      "train_speed(iter/s)": 1.635654
    },
    {
      "acc": 0.64957895,
      "epoch": 1.0678589548452562,
      "grad_norm": 5.71875,
      "learning_rate": 4.852375242335217e-06,
      "loss": 1.64058018,
      "memory(GiB)": 117.38,
      "step": 42095,
      "train_speed(iter/s)": 1.635676
    },
    {
      "acc": 0.65379176,
      "epoch": 1.0679857940131914,
      "grad_norm": 5.28125,
      "learning_rate": 4.851327079064314e-06,
      "loss": 1.59891739,
      "memory(GiB)": 117.38,
      "step": 42100,
      "train_speed(iter/s)": 1.635696
    },
    {
      "acc": 0.6463707,
      "epoch": 1.0681126331811264,
      "grad_norm": 6.03125,
      "learning_rate": 4.850278922332708e-06,
      "loss": 1.70048218,
      "memory(GiB)": 117.38,
      "step": 42105,
      "train_speed(iter/s)": 1.635718
    },
    {
      "acc": 0.65905781,
      "epoch": 1.0682394723490614,
      "grad_norm": 6.46875,
      "learning_rate": 4.849230772186508e-06,
      "loss": 1.61065731,
      "memory(GiB)": 117.38,
      "step": 42110,
      "train_speed(iter/s)": 1.635738
    },
    {
      "acc": 0.66508641,
      "epoch": 1.0683663115169963,
      "grad_norm": 5.25,
      "learning_rate": 4.848182628671806e-06,
      "loss": 1.54283333,
      "memory(GiB)": 117.38,
      "step": 42115,
      "train_speed(iter/s)": 1.635759
    },
    {
      "acc": 0.67158318,
      "epoch": 1.0684931506849316,
      "grad_norm": 5.03125,
      "learning_rate": 4.847134491834713e-06,
      "loss": 1.54667645,
      "memory(GiB)": 117.38,
      "step": 42120,
      "train_speed(iter/s)": 1.63578
    },
    {
      "acc": 0.66108823,
      "epoch": 1.0686199898528665,
      "grad_norm": 5.1875,
      "learning_rate": 4.846086361721326e-06,
      "loss": 1.54157867,
      "memory(GiB)": 117.38,
      "step": 42125,
      "train_speed(iter/s)": 1.635803
    },
    {
      "acc": 0.64167342,
      "epoch": 1.0687468290208015,
      "grad_norm": 6.46875,
      "learning_rate": 4.84503823837775e-06,
      "loss": 1.61747932,
      "memory(GiB)": 117.38,
      "step": 42130,
      "train_speed(iter/s)": 1.635826
    },
    {
      "acc": 0.6583787,
      "epoch": 1.0688736681887367,
      "grad_norm": 5.15625,
      "learning_rate": 4.843990121850083e-06,
      "loss": 1.55810575,
      "memory(GiB)": 117.38,
      "step": 42135,
      "train_speed(iter/s)": 1.635848
    },
    {
      "acc": 0.64197702,
      "epoch": 1.0690005073566717,
      "grad_norm": 5.875,
      "learning_rate": 4.842942012184426e-06,
      "loss": 1.62249298,
      "memory(GiB)": 117.38,
      "step": 42140,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.64114494,
      "epoch": 1.0691273465246067,
      "grad_norm": 5.0,
      "learning_rate": 4.841893909426881e-06,
      "loss": 1.64206657,
      "memory(GiB)": 117.38,
      "step": 42145,
      "train_speed(iter/s)": 1.635891
    },
    {
      "acc": 0.65284929,
      "epoch": 1.069254185692542,
      "grad_norm": 7.625,
      "learning_rate": 4.84084581362355e-06,
      "loss": 1.65577507,
      "memory(GiB)": 117.38,
      "step": 42150,
      "train_speed(iter/s)": 1.635912
    },
    {
      "acc": 0.65538263,
      "epoch": 1.069381024860477,
      "grad_norm": 5.375,
      "learning_rate": 4.839797724820529e-06,
      "loss": 1.61465034,
      "memory(GiB)": 117.38,
      "step": 42155,
      "train_speed(iter/s)": 1.635932
    },
    {
      "acc": 0.6463922,
      "epoch": 1.069507864028412,
      "grad_norm": 6.84375,
      "learning_rate": 4.838749643063918e-06,
      "loss": 1.62764587,
      "memory(GiB)": 117.38,
      "step": 42160,
      "train_speed(iter/s)": 1.635954
    },
    {
      "acc": 0.65110922,
      "epoch": 1.0696347031963471,
      "grad_norm": 5.46875,
      "learning_rate": 4.837701568399819e-06,
      "loss": 1.62532349,
      "memory(GiB)": 117.38,
      "step": 42165,
      "train_speed(iter/s)": 1.635975
    },
    {
      "acc": 0.64151073,
      "epoch": 1.0697615423642821,
      "grad_norm": 4.6875,
      "learning_rate": 4.836653500874331e-06,
      "loss": 1.6430233,
      "memory(GiB)": 117.38,
      "step": 42170,
      "train_speed(iter/s)": 1.635997
    },
    {
      "acc": 0.6517695,
      "epoch": 1.069888381532217,
      "grad_norm": 6.25,
      "learning_rate": 4.835605440533549e-06,
      "loss": 1.64731331,
      "memory(GiB)": 117.38,
      "step": 42175,
      "train_speed(iter/s)": 1.636018
    },
    {
      "acc": 0.66772804,
      "epoch": 1.070015220700152,
      "grad_norm": 6.625,
      "learning_rate": 4.834557387423575e-06,
      "loss": 1.50288582,
      "memory(GiB)": 117.38,
      "step": 42180,
      "train_speed(iter/s)": 1.63604
    },
    {
      "acc": 0.66155567,
      "epoch": 1.0701420598680873,
      "grad_norm": 5.1875,
      "learning_rate": 4.833509341590503e-06,
      "loss": 1.61410942,
      "memory(GiB)": 117.38,
      "step": 42185,
      "train_speed(iter/s)": 1.636061
    },
    {
      "acc": 0.66356831,
      "epoch": 1.0702688990360223,
      "grad_norm": 5.59375,
      "learning_rate": 4.8324613030804374e-06,
      "loss": 1.53381405,
      "memory(GiB)": 117.38,
      "step": 42190,
      "train_speed(iter/s)": 1.636081
    },
    {
      "acc": 0.64900713,
      "epoch": 1.0703957382039573,
      "grad_norm": 6.03125,
      "learning_rate": 4.83141327193947e-06,
      "loss": 1.64698257,
      "memory(GiB)": 117.38,
      "step": 42195,
      "train_speed(iter/s)": 1.636102
    },
    {
      "acc": 0.66689458,
      "epoch": 1.0705225773718925,
      "grad_norm": 5.71875,
      "learning_rate": 4.8303652482137e-06,
      "loss": 1.55340824,
      "memory(GiB)": 117.38,
      "step": 42200,
      "train_speed(iter/s)": 1.636123
    },
    {
      "acc": 0.63156323,
      "epoch": 1.0706494165398275,
      "grad_norm": 6.90625,
      "learning_rate": 4.829317231949222e-06,
      "loss": 1.68912773,
      "memory(GiB)": 117.38,
      "step": 42205,
      "train_speed(iter/s)": 1.636145
    },
    {
      "acc": 0.66935859,
      "epoch": 1.0707762557077625,
      "grad_norm": 5.875,
      "learning_rate": 4.828269223192137e-06,
      "loss": 1.58948174,
      "memory(GiB)": 117.38,
      "step": 42210,
      "train_speed(iter/s)": 1.636165
    },
    {
      "acc": 0.65661216,
      "epoch": 1.0709030948756977,
      "grad_norm": 6.0,
      "learning_rate": 4.827221221988537e-06,
      "loss": 1.56759233,
      "memory(GiB)": 117.38,
      "step": 42215,
      "train_speed(iter/s)": 1.636185
    },
    {
      "acc": 0.6571384,
      "epoch": 1.0710299340436327,
      "grad_norm": 5.28125,
      "learning_rate": 4.826173228384518e-06,
      "loss": 1.56036091,
      "memory(GiB)": 117.38,
      "step": 42220,
      "train_speed(iter/s)": 1.636207
    },
    {
      "acc": 0.66064982,
      "epoch": 1.0711567732115677,
      "grad_norm": 5.3125,
      "learning_rate": 4.8251252424261775e-06,
      "loss": 1.61549644,
      "memory(GiB)": 117.38,
      "step": 42225,
      "train_speed(iter/s)": 1.636228
    },
    {
      "acc": 0.67536812,
      "epoch": 1.0712836123795029,
      "grad_norm": 4.9375,
      "learning_rate": 4.8240772641596105e-06,
      "loss": 1.49482794,
      "memory(GiB)": 117.38,
      "step": 42230,
      "train_speed(iter/s)": 1.636248
    },
    {
      "acc": 0.66572533,
      "epoch": 1.0714104515474379,
      "grad_norm": 5.625,
      "learning_rate": 4.82302929363091e-06,
      "loss": 1.60782547,
      "memory(GiB)": 117.38,
      "step": 42235,
      "train_speed(iter/s)": 1.63627
    },
    {
      "acc": 0.65329785,
      "epoch": 1.0715372907153728,
      "grad_norm": 6.0,
      "learning_rate": 4.8219813308861705e-06,
      "loss": 1.6175312,
      "memory(GiB)": 117.38,
      "step": 42240,
      "train_speed(iter/s)": 1.63629
    },
    {
      "acc": 0.6694169,
      "epoch": 1.071664129883308,
      "grad_norm": 5.6875,
      "learning_rate": 4.820933375971487e-06,
      "loss": 1.53876801,
      "memory(GiB)": 117.38,
      "step": 42245,
      "train_speed(iter/s)": 1.636311
    },
    {
      "acc": 0.64812908,
      "epoch": 1.071790969051243,
      "grad_norm": 6.90625,
      "learning_rate": 4.819885428932955e-06,
      "loss": 1.61719284,
      "memory(GiB)": 117.38,
      "step": 42250,
      "train_speed(iter/s)": 1.636333
    },
    {
      "acc": 0.6650456,
      "epoch": 1.071917808219178,
      "grad_norm": 5.46875,
      "learning_rate": 4.818837489816664e-06,
      "loss": 1.55047035,
      "memory(GiB)": 117.38,
      "step": 42255,
      "train_speed(iter/s)": 1.636354
    },
    {
      "acc": 0.64859395,
      "epoch": 1.0720446473871132,
      "grad_norm": 5.25,
      "learning_rate": 4.81778955866871e-06,
      "loss": 1.64661102,
      "memory(GiB)": 117.38,
      "step": 42260,
      "train_speed(iter/s)": 1.636375
    },
    {
      "acc": 0.67052026,
      "epoch": 1.0721714865550482,
      "grad_norm": 5.40625,
      "learning_rate": 4.816741635535183e-06,
      "loss": 1.56664429,
      "memory(GiB)": 117.38,
      "step": 42265,
      "train_speed(iter/s)": 1.636396
    },
    {
      "acc": 0.6433629,
      "epoch": 1.0722983257229832,
      "grad_norm": 5.6875,
      "learning_rate": 4.81569372046218e-06,
      "loss": 1.68916283,
      "memory(GiB)": 117.38,
      "step": 42270,
      "train_speed(iter/s)": 1.636416
    },
    {
      "acc": 0.64638591,
      "epoch": 1.0724251648909182,
      "grad_norm": 6.375,
      "learning_rate": 4.814645813495788e-06,
      "loss": 1.65827789,
      "memory(GiB)": 117.38,
      "step": 42275,
      "train_speed(iter/s)": 1.636436
    },
    {
      "acc": 0.65866117,
      "epoch": 1.0725520040588534,
      "grad_norm": 4.5625,
      "learning_rate": 4.8135979146821e-06,
      "loss": 1.62065964,
      "memory(GiB)": 117.38,
      "step": 42280,
      "train_speed(iter/s)": 1.636457
    },
    {
      "acc": 0.64541955,
      "epoch": 1.0726788432267884,
      "grad_norm": 5.96875,
      "learning_rate": 4.81255002406721e-06,
      "loss": 1.61616898,
      "memory(GiB)": 117.38,
      "step": 42285,
      "train_speed(iter/s)": 1.636478
    },
    {
      "acc": 0.67276602,
      "epoch": 1.0728056823947234,
      "grad_norm": 6.84375,
      "learning_rate": 4.811502141697206e-06,
      "loss": 1.51470623,
      "memory(GiB)": 117.38,
      "step": 42290,
      "train_speed(iter/s)": 1.636499
    },
    {
      "acc": 0.65773182,
      "epoch": 1.0729325215626586,
      "grad_norm": 4.8125,
      "learning_rate": 4.81045426761818e-06,
      "loss": 1.58835955,
      "memory(GiB)": 117.38,
      "step": 42295,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.6545176,
      "epoch": 1.0730593607305936,
      "grad_norm": 4.8125,
      "learning_rate": 4.80940640187622e-06,
      "loss": 1.60705605,
      "memory(GiB)": 117.38,
      "step": 42300,
      "train_speed(iter/s)": 1.63654
    },
    {
      "acc": 0.66162596,
      "epoch": 1.0731861998985286,
      "grad_norm": 6.0,
      "learning_rate": 4.808358544517418e-06,
      "loss": 1.52777958,
      "memory(GiB)": 117.38,
      "step": 42305,
      "train_speed(iter/s)": 1.636561
    },
    {
      "acc": 0.66089416,
      "epoch": 1.0733130390664638,
      "grad_norm": 6.21875,
      "learning_rate": 4.807310695587865e-06,
      "loss": 1.52624264,
      "memory(GiB)": 117.38,
      "step": 42310,
      "train_speed(iter/s)": 1.636582
    },
    {
      "acc": 0.64437509,
      "epoch": 1.0734398782343988,
      "grad_norm": 5.46875,
      "learning_rate": 4.8062628551336445e-06,
      "loss": 1.6733593,
      "memory(GiB)": 117.38,
      "step": 42315,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.65218344,
      "epoch": 1.0735667174023338,
      "grad_norm": 5.5625,
      "learning_rate": 4.80521502320085e-06,
      "loss": 1.6339716,
      "memory(GiB)": 117.38,
      "step": 42320,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.65528421,
      "epoch": 1.073693556570269,
      "grad_norm": 6.5625,
      "learning_rate": 4.804167199835567e-06,
      "loss": 1.59885778,
      "memory(GiB)": 117.38,
      "step": 42325,
      "train_speed(iter/s)": 1.636642
    },
    {
      "acc": 0.66165695,
      "epoch": 1.073820395738204,
      "grad_norm": 5.5625,
      "learning_rate": 4.8031193850838894e-06,
      "loss": 1.54507418,
      "memory(GiB)": 117.38,
      "step": 42330,
      "train_speed(iter/s)": 1.636662
    },
    {
      "acc": 0.6511714,
      "epoch": 1.073947234906139,
      "grad_norm": 5.34375,
      "learning_rate": 4.802071578991896e-06,
      "loss": 1.6204174,
      "memory(GiB)": 117.38,
      "step": 42335,
      "train_speed(iter/s)": 1.636684
    },
    {
      "acc": 0.65364795,
      "epoch": 1.074074074074074,
      "grad_norm": 5.09375,
      "learning_rate": 4.801023781605679e-06,
      "loss": 1.62786865,
      "memory(GiB)": 117.38,
      "step": 42340,
      "train_speed(iter/s)": 1.636705
    },
    {
      "acc": 0.65816154,
      "epoch": 1.0742009132420092,
      "grad_norm": 4.8125,
      "learning_rate": 4.799975992971325e-06,
      "loss": 1.59796553,
      "memory(GiB)": 117.38,
      "step": 42345,
      "train_speed(iter/s)": 1.636725
    },
    {
      "acc": 0.6649786,
      "epoch": 1.0743277524099442,
      "grad_norm": 5.46875,
      "learning_rate": 4.798928213134921e-06,
      "loss": 1.54111347,
      "memory(GiB)": 117.38,
      "step": 42350,
      "train_speed(iter/s)": 1.636747
    },
    {
      "acc": 0.66917233,
      "epoch": 1.0744545915778791,
      "grad_norm": 6.0,
      "learning_rate": 4.797880442142551e-06,
      "loss": 1.61523819,
      "memory(GiB)": 117.38,
      "step": 42355,
      "train_speed(iter/s)": 1.636767
    },
    {
      "acc": 0.65927925,
      "epoch": 1.0745814307458144,
      "grad_norm": 5.71875,
      "learning_rate": 4.7968326800403e-06,
      "loss": 1.57280722,
      "memory(GiB)": 117.38,
      "step": 42360,
      "train_speed(iter/s)": 1.636788
    },
    {
      "acc": 0.64824514,
      "epoch": 1.0747082699137493,
      "grad_norm": 7.15625,
      "learning_rate": 4.795784926874255e-06,
      "loss": 1.64902573,
      "memory(GiB)": 117.38,
      "step": 42365,
      "train_speed(iter/s)": 1.63681
    },
    {
      "acc": 0.66020918,
      "epoch": 1.0748351090816843,
      "grad_norm": 7.25,
      "learning_rate": 4.794737182690503e-06,
      "loss": 1.59338675,
      "memory(GiB)": 117.38,
      "step": 42370,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.63986688,
      "epoch": 1.0749619482496195,
      "grad_norm": 5.03125,
      "learning_rate": 4.793689447535126e-06,
      "loss": 1.58832169,
      "memory(GiB)": 117.38,
      "step": 42375,
      "train_speed(iter/s)": 1.636852
    },
    {
      "acc": 0.67833238,
      "epoch": 1.0750887874175545,
      "grad_norm": 6.34375,
      "learning_rate": 4.792641721454206e-06,
      "loss": 1.57167778,
      "memory(GiB)": 117.38,
      "step": 42380,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.66357117,
      "epoch": 1.0752156265854895,
      "grad_norm": 5.40625,
      "learning_rate": 4.79159400449383e-06,
      "loss": 1.56718082,
      "memory(GiB)": 117.38,
      "step": 42385,
      "train_speed(iter/s)": 1.636895
    },
    {
      "acc": 0.66061506,
      "epoch": 1.0753424657534247,
      "grad_norm": 5.59375,
      "learning_rate": 4.7905462967000816e-06,
      "loss": 1.62065353,
      "memory(GiB)": 117.38,
      "step": 42390,
      "train_speed(iter/s)": 1.636915
    },
    {
      "acc": 0.65036502,
      "epoch": 1.0754693049213597,
      "grad_norm": 6.21875,
      "learning_rate": 4.789498598119039e-06,
      "loss": 1.63716011,
      "memory(GiB)": 117.38,
      "step": 42395,
      "train_speed(iter/s)": 1.636937
    },
    {
      "acc": 0.65736675,
      "epoch": 1.0755961440892947,
      "grad_norm": 5.46875,
      "learning_rate": 4.78845090879679e-06,
      "loss": 1.58783159,
      "memory(GiB)": 117.38,
      "step": 42400,
      "train_speed(iter/s)": 1.636958
    },
    {
      "acc": 0.65636415,
      "epoch": 1.07572298325723,
      "grad_norm": 6.5,
      "learning_rate": 4.787403228779413e-06,
      "loss": 1.58273706,
      "memory(GiB)": 117.38,
      "step": 42405,
      "train_speed(iter/s)": 1.636979
    },
    {
      "acc": 0.65250158,
      "epoch": 1.075849822425165,
      "grad_norm": 7.5,
      "learning_rate": 4.786355558112994e-06,
      "loss": 1.65883331,
      "memory(GiB)": 117.38,
      "step": 42410,
      "train_speed(iter/s)": 1.637
    },
    {
      "acc": 0.67009602,
      "epoch": 1.0759766615931,
      "grad_norm": 5.3125,
      "learning_rate": 4.78530789684361e-06,
      "loss": 1.54517412,
      "memory(GiB)": 117.38,
      "step": 42415,
      "train_speed(iter/s)": 1.637022
    },
    {
      "acc": 0.64447303,
      "epoch": 1.0761035007610351,
      "grad_norm": 4.4375,
      "learning_rate": 4.784260245017343e-06,
      "loss": 1.65884037,
      "memory(GiB)": 117.38,
      "step": 42420,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.66058936,
      "epoch": 1.07623033992897,
      "grad_norm": 5.46875,
      "learning_rate": 4.7832126026802725e-06,
      "loss": 1.60292816,
      "memory(GiB)": 117.38,
      "step": 42425,
      "train_speed(iter/s)": 1.637064
    },
    {
      "acc": 0.66112666,
      "epoch": 1.076357179096905,
      "grad_norm": 6.28125,
      "learning_rate": 4.782164969878482e-06,
      "loss": 1.5541626,
      "memory(GiB)": 117.38,
      "step": 42430,
      "train_speed(iter/s)": 1.637086
    },
    {
      "acc": 0.63486061,
      "epoch": 1.07648401826484,
      "grad_norm": 4.9375,
      "learning_rate": 4.781117346658047e-06,
      "loss": 1.62688007,
      "memory(GiB)": 117.38,
      "step": 42435,
      "train_speed(iter/s)": 1.637106
    },
    {
      "acc": 0.64601574,
      "epoch": 1.0766108574327753,
      "grad_norm": 5.6875,
      "learning_rate": 4.780069733065048e-06,
      "loss": 1.60367508,
      "memory(GiB)": 117.38,
      "step": 42440,
      "train_speed(iter/s)": 1.637127
    },
    {
      "acc": 0.6611485,
      "epoch": 1.0767376966007103,
      "grad_norm": 8.9375,
      "learning_rate": 4.779022129145566e-06,
      "loss": 1.55333118,
      "memory(GiB)": 117.38,
      "step": 42445,
      "train_speed(iter/s)": 1.637149
    },
    {
      "acc": 0.65445738,
      "epoch": 1.0768645357686453,
      "grad_norm": 5.59375,
      "learning_rate": 4.777974534945677e-06,
      "loss": 1.66348419,
      "memory(GiB)": 117.38,
      "step": 42450,
      "train_speed(iter/s)": 1.63717
    },
    {
      "acc": 0.65219812,
      "epoch": 1.0769913749365805,
      "grad_norm": 4.65625,
      "learning_rate": 4.776926950511457e-06,
      "loss": 1.65615845,
      "memory(GiB)": 117.38,
      "step": 42455,
      "train_speed(iter/s)": 1.637191
    },
    {
      "acc": 0.63933501,
      "epoch": 1.0771182141045155,
      "grad_norm": 5.5,
      "learning_rate": 4.775879375888986e-06,
      "loss": 1.68618317,
      "memory(GiB)": 117.38,
      "step": 42460,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.66758986,
      "epoch": 1.0772450532724505,
      "grad_norm": 5.34375,
      "learning_rate": 4.774831811124343e-06,
      "loss": 1.58380451,
      "memory(GiB)": 117.38,
      "step": 42465,
      "train_speed(iter/s)": 1.63723
    },
    {
      "acc": 0.65434456,
      "epoch": 1.0773718924403857,
      "grad_norm": 5.375,
      "learning_rate": 4.773784256263601e-06,
      "loss": 1.59232121,
      "memory(GiB)": 117.38,
      "step": 42470,
      "train_speed(iter/s)": 1.637252
    },
    {
      "acc": 0.64999151,
      "epoch": 1.0774987316083207,
      "grad_norm": 5.34375,
      "learning_rate": 4.7727367113528374e-06,
      "loss": 1.61950951,
      "memory(GiB)": 117.38,
      "step": 42475,
      "train_speed(iter/s)": 1.637272
    },
    {
      "acc": 0.6488997,
      "epoch": 1.0776255707762556,
      "grad_norm": 5.375,
      "learning_rate": 4.771689176438128e-06,
      "loss": 1.64509106,
      "memory(GiB)": 117.38,
      "step": 42480,
      "train_speed(iter/s)": 1.637294
    },
    {
      "acc": 0.65637975,
      "epoch": 1.0777524099441909,
      "grad_norm": 6.96875,
      "learning_rate": 4.770641651565546e-06,
      "loss": 1.53319883,
      "memory(GiB)": 117.38,
      "step": 42485,
      "train_speed(iter/s)": 1.637315
    },
    {
      "acc": 0.65938635,
      "epoch": 1.0778792491121258,
      "grad_norm": 5.6875,
      "learning_rate": 4.769594136781172e-06,
      "loss": 1.56976175,
      "memory(GiB)": 117.38,
      "step": 42490,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.66910419,
      "epoch": 1.0780060882800608,
      "grad_norm": 6.65625,
      "learning_rate": 4.768546632131074e-06,
      "loss": 1.58377504,
      "memory(GiB)": 117.38,
      "step": 42495,
      "train_speed(iter/s)": 1.637359
    },
    {
      "acc": 0.65934429,
      "epoch": 1.0781329274479958,
      "grad_norm": 8.3125,
      "learning_rate": 4.767499137661328e-06,
      "loss": 1.65002556,
      "memory(GiB)": 117.38,
      "step": 42500,
      "train_speed(iter/s)": 1.637379
    },
    {
      "acc": 0.67151184,
      "epoch": 1.078259766615931,
      "grad_norm": 4.59375,
      "learning_rate": 4.76645165341801e-06,
      "loss": 1.50512838,
      "memory(GiB)": 117.38,
      "step": 42505,
      "train_speed(iter/s)": 1.637399
    },
    {
      "acc": 0.64283657,
      "epoch": 1.078386605783866,
      "grad_norm": 5.84375,
      "learning_rate": 4.76540417944719e-06,
      "loss": 1.59125729,
      "memory(GiB)": 117.38,
      "step": 42510,
      "train_speed(iter/s)": 1.637419
    },
    {
      "acc": 0.66571417,
      "epoch": 1.078513444951801,
      "grad_norm": 4.71875,
      "learning_rate": 4.764356715794942e-06,
      "loss": 1.60510559,
      "memory(GiB)": 117.38,
      "step": 42515,
      "train_speed(iter/s)": 1.63744
    },
    {
      "acc": 0.6593575,
      "epoch": 1.0786402841197362,
      "grad_norm": 7.96875,
      "learning_rate": 4.763309262507336e-06,
      "loss": 1.55107212,
      "memory(GiB)": 117.38,
      "step": 42520,
      "train_speed(iter/s)": 1.63746
    },
    {
      "acc": 0.65673194,
      "epoch": 1.0787671232876712,
      "grad_norm": 6.875,
      "learning_rate": 4.762261819630447e-06,
      "loss": 1.6110527,
      "memory(GiB)": 117.38,
      "step": 42525,
      "train_speed(iter/s)": 1.637481
    },
    {
      "acc": 0.66404743,
      "epoch": 1.0788939624556062,
      "grad_norm": 5.75,
      "learning_rate": 4.761214387210345e-06,
      "loss": 1.5447691,
      "memory(GiB)": 117.38,
      "step": 42530,
      "train_speed(iter/s)": 1.637503
    },
    {
      "acc": 0.67320657,
      "epoch": 1.0790208016235414,
      "grad_norm": 5.375,
      "learning_rate": 4.760166965293099e-06,
      "loss": 1.58602543,
      "memory(GiB)": 117.38,
      "step": 42535,
      "train_speed(iter/s)": 1.637524
    },
    {
      "acc": 0.65891619,
      "epoch": 1.0791476407914764,
      "grad_norm": 6.59375,
      "learning_rate": 4.759119553924781e-06,
      "loss": 1.59398317,
      "memory(GiB)": 117.38,
      "step": 42540,
      "train_speed(iter/s)": 1.637544
    },
    {
      "acc": 0.65041962,
      "epoch": 1.0792744799594114,
      "grad_norm": 5.53125,
      "learning_rate": 4.758072153151461e-06,
      "loss": 1.56685324,
      "memory(GiB)": 117.38,
      "step": 42545,
      "train_speed(iter/s)": 1.637565
    },
    {
      "acc": 0.65652957,
      "epoch": 1.0794013191273466,
      "grad_norm": 5.65625,
      "learning_rate": 4.757024763019209e-06,
      "loss": 1.54740772,
      "memory(GiB)": 117.38,
      "step": 42550,
      "train_speed(iter/s)": 1.637585
    },
    {
      "acc": 0.65493364,
      "epoch": 1.0795281582952816,
      "grad_norm": 5.09375,
      "learning_rate": 4.755977383574091e-06,
      "loss": 1.58426991,
      "memory(GiB)": 117.38,
      "step": 42555,
      "train_speed(iter/s)": 1.637607
    },
    {
      "acc": 0.66231947,
      "epoch": 1.0796549974632166,
      "grad_norm": 5.6875,
      "learning_rate": 4.754930014862177e-06,
      "loss": 1.60242119,
      "memory(GiB)": 117.38,
      "step": 42560,
      "train_speed(iter/s)": 1.637628
    },
    {
      "acc": 0.66401873,
      "epoch": 1.0797818366311518,
      "grad_norm": 6.3125,
      "learning_rate": 4.753882656929535e-06,
      "loss": 1.64540787,
      "memory(GiB)": 117.38,
      "step": 42565,
      "train_speed(iter/s)": 1.637649
    },
    {
      "acc": 0.6570085,
      "epoch": 1.0799086757990868,
      "grad_norm": 5.25,
      "learning_rate": 4.752835309822234e-06,
      "loss": 1.60429153,
      "memory(GiB)": 117.38,
      "step": 42570,
      "train_speed(iter/s)": 1.63767
    },
    {
      "acc": 0.66406903,
      "epoch": 1.0800355149670218,
      "grad_norm": 5.96875,
      "learning_rate": 4.7517879735863385e-06,
      "loss": 1.57594776,
      "memory(GiB)": 117.38,
      "step": 42575,
      "train_speed(iter/s)": 1.637691
    },
    {
      "acc": 0.64655027,
      "epoch": 1.080162354134957,
      "grad_norm": 4.84375,
      "learning_rate": 4.750740648267916e-06,
      "loss": 1.62976341,
      "memory(GiB)": 117.38,
      "step": 42580,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.6545217,
      "epoch": 1.080289193302892,
      "grad_norm": 4.875,
      "learning_rate": 4.749693333913033e-06,
      "loss": 1.60688286,
      "memory(GiB)": 117.38,
      "step": 42585,
      "train_speed(iter/s)": 1.637734
    },
    {
      "acc": 0.65543194,
      "epoch": 1.080416032470827,
      "grad_norm": 5.84375,
      "learning_rate": 4.748646030567755e-06,
      "loss": 1.61598892,
      "memory(GiB)": 117.38,
      "step": 42590,
      "train_speed(iter/s)": 1.637755
    },
    {
      "acc": 0.66029997,
      "epoch": 1.080542871638762,
      "grad_norm": 6.3125,
      "learning_rate": 4.747598738278147e-06,
      "loss": 1.61568012,
      "memory(GiB)": 117.38,
      "step": 42595,
      "train_speed(iter/s)": 1.637778
    },
    {
      "acc": 0.6595809,
      "epoch": 1.0806697108066972,
      "grad_norm": 4.9375,
      "learning_rate": 4.746551457090272e-06,
      "loss": 1.61032467,
      "memory(GiB)": 117.38,
      "step": 42600,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.65252342,
      "epoch": 1.0807965499746321,
      "grad_norm": 5.25,
      "learning_rate": 4.745504187050197e-06,
      "loss": 1.65675735,
      "memory(GiB)": 117.38,
      "step": 42605,
      "train_speed(iter/s)": 1.637819
    },
    {
      "acc": 0.64243746,
      "epoch": 1.0809233891425671,
      "grad_norm": 5.5625,
      "learning_rate": 4.744456928203985e-06,
      "loss": 1.63453903,
      "memory(GiB)": 117.38,
      "step": 42610,
      "train_speed(iter/s)": 1.637839
    },
    {
      "acc": 0.65615311,
      "epoch": 1.0810502283105023,
      "grad_norm": 6.75,
      "learning_rate": 4.743409680597695e-06,
      "loss": 1.58209953,
      "memory(GiB)": 117.38,
      "step": 42615,
      "train_speed(iter/s)": 1.637861
    },
    {
      "acc": 0.64392343,
      "epoch": 1.0811770674784373,
      "grad_norm": 5.8125,
      "learning_rate": 4.742362444277394e-06,
      "loss": 1.64784203,
      "memory(GiB)": 117.38,
      "step": 42620,
      "train_speed(iter/s)": 1.637884
    },
    {
      "acc": 0.65613842,
      "epoch": 1.0813039066463723,
      "grad_norm": 5.28125,
      "learning_rate": 4.741315219289142e-06,
      "loss": 1.63932877,
      "memory(GiB)": 117.38,
      "step": 42625,
      "train_speed(iter/s)": 1.637905
    },
    {
      "acc": 0.6664278,
      "epoch": 1.0814307458143075,
      "grad_norm": 6.15625,
      "learning_rate": 4.740268005679005e-06,
      "loss": 1.55967646,
      "memory(GiB)": 117.38,
      "step": 42630,
      "train_speed(iter/s)": 1.637926
    },
    {
      "acc": 0.66025348,
      "epoch": 1.0815575849822425,
      "grad_norm": 7.03125,
      "learning_rate": 4.739220803493039e-06,
      "loss": 1.55208797,
      "memory(GiB)": 117.38,
      "step": 42635,
      "train_speed(iter/s)": 1.637946
    },
    {
      "acc": 0.65910797,
      "epoch": 1.0816844241501775,
      "grad_norm": 6.65625,
      "learning_rate": 4.738173612777306e-06,
      "loss": 1.65165787,
      "memory(GiB)": 117.38,
      "step": 42640,
      "train_speed(iter/s)": 1.63797
    },
    {
      "acc": 0.64050398,
      "epoch": 1.0818112633181127,
      "grad_norm": 5.3125,
      "learning_rate": 4.737126433577866e-06,
      "loss": 1.69772148,
      "memory(GiB)": 117.38,
      "step": 42645,
      "train_speed(iter/s)": 1.637991
    },
    {
      "acc": 0.6448998,
      "epoch": 1.0819381024860477,
      "grad_norm": 5.9375,
      "learning_rate": 4.736079265940781e-06,
      "loss": 1.64091969,
      "memory(GiB)": 117.38,
      "step": 42650,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.64359446,
      "epoch": 1.0820649416539827,
      "grad_norm": 6.46875,
      "learning_rate": 4.735032109912107e-06,
      "loss": 1.6278429,
      "memory(GiB)": 117.38,
      "step": 42655,
      "train_speed(iter/s)": 1.638034
    },
    {
      "acc": 0.64896083,
      "epoch": 1.0821917808219177,
      "grad_norm": 5.53125,
      "learning_rate": 4.733984965537903e-06,
      "loss": 1.59420853,
      "memory(GiB)": 117.38,
      "step": 42660,
      "train_speed(iter/s)": 1.638055
    },
    {
      "acc": 0.67471514,
      "epoch": 1.082318619989853,
      "grad_norm": 6.5625,
      "learning_rate": 4.732937832864229e-06,
      "loss": 1.52894554,
      "memory(GiB)": 117.38,
      "step": 42665,
      "train_speed(iter/s)": 1.638076
    },
    {
      "acc": 0.64166245,
      "epoch": 1.082445459157788,
      "grad_norm": 6.34375,
      "learning_rate": 4.731890711937141e-06,
      "loss": 1.64043427,
      "memory(GiB)": 117.38,
      "step": 42670,
      "train_speed(iter/s)": 1.638098
    },
    {
      "acc": 0.65850821,
      "epoch": 1.0825722983257229,
      "grad_norm": 7.15625,
      "learning_rate": 4.730843602802696e-06,
      "loss": 1.57808895,
      "memory(GiB)": 117.38,
      "step": 42675,
      "train_speed(iter/s)": 1.638119
    },
    {
      "acc": 0.66012197,
      "epoch": 1.082699137493658,
      "grad_norm": 7.0625,
      "learning_rate": 4.729796505506951e-06,
      "loss": 1.57560768,
      "memory(GiB)": 117.38,
      "step": 42680,
      "train_speed(iter/s)": 1.638141
    },
    {
      "acc": 0.6368959,
      "epoch": 1.082825976661593,
      "grad_norm": 5.0,
      "learning_rate": 4.728749420095964e-06,
      "loss": 1.65686798,
      "memory(GiB)": 117.38,
      "step": 42685,
      "train_speed(iter/s)": 1.638162
    },
    {
      "acc": 0.66116824,
      "epoch": 1.082952815829528,
      "grad_norm": 5.71875,
      "learning_rate": 4.727702346615788e-06,
      "loss": 1.65334988,
      "memory(GiB)": 117.38,
      "step": 42690,
      "train_speed(iter/s)": 1.638184
    },
    {
      "acc": 0.66066723,
      "epoch": 1.0830796549974633,
      "grad_norm": 4.625,
      "learning_rate": 4.726655285112477e-06,
      "loss": 1.56014986,
      "memory(GiB)": 117.38,
      "step": 42695,
      "train_speed(iter/s)": 1.638204
    },
    {
      "acc": 0.64992914,
      "epoch": 1.0832064941653983,
      "grad_norm": 6.21875,
      "learning_rate": 4.725608235632088e-06,
      "loss": 1.62344704,
      "memory(GiB)": 117.38,
      "step": 42700,
      "train_speed(iter/s)": 1.638225
    },
    {
      "acc": 0.65785198,
      "epoch": 1.0833333333333333,
      "grad_norm": 6.25,
      "learning_rate": 4.724561198220672e-06,
      "loss": 1.60169411,
      "memory(GiB)": 117.38,
      "step": 42705,
      "train_speed(iter/s)": 1.638245
    },
    {
      "acc": 0.6505753,
      "epoch": 1.0834601725012685,
      "grad_norm": 5.21875,
      "learning_rate": 4.723514172924287e-06,
      "loss": 1.62297802,
      "memory(GiB)": 117.38,
      "step": 42710,
      "train_speed(iter/s)": 1.638266
    },
    {
      "acc": 0.64711261,
      "epoch": 1.0835870116692035,
      "grad_norm": 5.0625,
      "learning_rate": 4.7224671597889825e-06,
      "loss": 1.62159233,
      "memory(GiB)": 117.38,
      "step": 42715,
      "train_speed(iter/s)": 1.638286
    },
    {
      "acc": 0.64322829,
      "epoch": 1.0837138508371384,
      "grad_norm": 6.84375,
      "learning_rate": 4.72142015886081e-06,
      "loss": 1.66040306,
      "memory(GiB)": 117.38,
      "step": 42720,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.67230005,
      "epoch": 1.0838406900050737,
      "grad_norm": 5.46875,
      "learning_rate": 4.720373170185823e-06,
      "loss": 1.54251385,
      "memory(GiB)": 117.38,
      "step": 42725,
      "train_speed(iter/s)": 1.638327
    },
    {
      "acc": 0.66142106,
      "epoch": 1.0839675291730086,
      "grad_norm": 6.53125,
      "learning_rate": 4.719326193810075e-06,
      "loss": 1.64745331,
      "memory(GiB)": 117.38,
      "step": 42730,
      "train_speed(iter/s)": 1.638349
    },
    {
      "acc": 0.64883218,
      "epoch": 1.0840943683409436,
      "grad_norm": 7.40625,
      "learning_rate": 4.718279229779612e-06,
      "loss": 1.65219688,
      "memory(GiB)": 117.38,
      "step": 42735,
      "train_speed(iter/s)": 1.638371
    },
    {
      "acc": 0.6469192,
      "epoch": 1.0842212075088788,
      "grad_norm": 5.5,
      "learning_rate": 4.717232278140485e-06,
      "loss": 1.67322044,
      "memory(GiB)": 117.38,
      "step": 42740,
      "train_speed(iter/s)": 1.638391
    },
    {
      "acc": 0.67113018,
      "epoch": 1.0843480466768138,
      "grad_norm": 5.21875,
      "learning_rate": 4.716185338938746e-06,
      "loss": 1.53260422,
      "memory(GiB)": 117.38,
      "step": 42745,
      "train_speed(iter/s)": 1.638411
    },
    {
      "acc": 0.64632063,
      "epoch": 1.0844748858447488,
      "grad_norm": 6.65625,
      "learning_rate": 4.7151384122204445e-06,
      "loss": 1.57729111,
      "memory(GiB)": 117.38,
      "step": 42750,
      "train_speed(iter/s)": 1.638433
    },
    {
      "acc": 0.64745231,
      "epoch": 1.0846017250126838,
      "grad_norm": 6.15625,
      "learning_rate": 4.7140914980316254e-06,
      "loss": 1.58687439,
      "memory(GiB)": 117.38,
      "step": 42755,
      "train_speed(iter/s)": 1.638454
    },
    {
      "acc": 0.65624466,
      "epoch": 1.084728564180619,
      "grad_norm": 6.59375,
      "learning_rate": 4.713044596418339e-06,
      "loss": 1.63883438,
      "memory(GiB)": 117.38,
      "step": 42760,
      "train_speed(iter/s)": 1.638475
    },
    {
      "acc": 0.65851412,
      "epoch": 1.084855403348554,
      "grad_norm": 6.03125,
      "learning_rate": 4.711997707426632e-06,
      "loss": 1.60370827,
      "memory(GiB)": 117.38,
      "step": 42765,
      "train_speed(iter/s)": 1.638495
    },
    {
      "acc": 0.65490026,
      "epoch": 1.084982242516489,
      "grad_norm": 5.78125,
      "learning_rate": 4.710950831102555e-06,
      "loss": 1.60263519,
      "memory(GiB)": 117.38,
      "step": 42770,
      "train_speed(iter/s)": 1.638516
    },
    {
      "acc": 0.65141826,
      "epoch": 1.0851090816844242,
      "grad_norm": 6.25,
      "learning_rate": 4.709903967492147e-06,
      "loss": 1.57416229,
      "memory(GiB)": 117.38,
      "step": 42775,
      "train_speed(iter/s)": 1.638538
    },
    {
      "acc": 0.64909539,
      "epoch": 1.0852359208523592,
      "grad_norm": 5.75,
      "learning_rate": 4.7088571166414595e-06,
      "loss": 1.59110508,
      "memory(GiB)": 117.38,
      "step": 42780,
      "train_speed(iter/s)": 1.638558
    },
    {
      "acc": 0.6568511,
      "epoch": 1.0853627600202942,
      "grad_norm": 6.15625,
      "learning_rate": 4.707810278596534e-06,
      "loss": 1.62210732,
      "memory(GiB)": 117.38,
      "step": 42785,
      "train_speed(iter/s)": 1.638579
    },
    {
      "acc": 0.6650569,
      "epoch": 1.0854895991882294,
      "grad_norm": 6.875,
      "learning_rate": 4.7067634534034205e-06,
      "loss": 1.60056992,
      "memory(GiB)": 117.38,
      "step": 42790,
      "train_speed(iter/s)": 1.638599
    },
    {
      "acc": 0.65203667,
      "epoch": 1.0856164383561644,
      "grad_norm": 6.65625,
      "learning_rate": 4.705716641108157e-06,
      "loss": 1.59725084,
      "memory(GiB)": 117.38,
      "step": 42795,
      "train_speed(iter/s)": 1.638621
    },
    {
      "acc": 0.65436087,
      "epoch": 1.0857432775240994,
      "grad_norm": 6.46875,
      "learning_rate": 4.7046698417567894e-06,
      "loss": 1.63516769,
      "memory(GiB)": 117.38,
      "step": 42800,
      "train_speed(iter/s)": 1.638641
    },
    {
      "acc": 0.65082111,
      "epoch": 1.0858701166920346,
      "grad_norm": 5.875,
      "learning_rate": 4.7036230553953616e-06,
      "loss": 1.61057377,
      "memory(GiB)": 117.38,
      "step": 42805,
      "train_speed(iter/s)": 1.638662
    },
    {
      "acc": 0.66845093,
      "epoch": 1.0859969558599696,
      "grad_norm": 6.34375,
      "learning_rate": 4.702576282069916e-06,
      "loss": 1.53074169,
      "memory(GiB)": 117.38,
      "step": 42810,
      "train_speed(iter/s)": 1.638681
    },
    {
      "acc": 0.64389658,
      "epoch": 1.0861237950279046,
      "grad_norm": 6.0,
      "learning_rate": 4.701529521826492e-06,
      "loss": 1.6562664,
      "memory(GiB)": 117.38,
      "step": 42815,
      "train_speed(iter/s)": 1.638703
    },
    {
      "acc": 0.66528811,
      "epoch": 1.0862506341958396,
      "grad_norm": 5.3125,
      "learning_rate": 4.700482774711131e-06,
      "loss": 1.60115089,
      "memory(GiB)": 117.38,
      "step": 42820,
      "train_speed(iter/s)": 1.638725
    },
    {
      "acc": 0.66938047,
      "epoch": 1.0863774733637748,
      "grad_norm": 6.25,
      "learning_rate": 4.699436040769877e-06,
      "loss": 1.49395771,
      "memory(GiB)": 117.38,
      "step": 42825,
      "train_speed(iter/s)": 1.638745
    },
    {
      "acc": 0.66367378,
      "epoch": 1.0865043125317098,
      "grad_norm": 6.15625,
      "learning_rate": 4.698389320048768e-06,
      "loss": 1.59599857,
      "memory(GiB)": 117.38,
      "step": 42830,
      "train_speed(iter/s)": 1.638766
    },
    {
      "acc": 0.63971758,
      "epoch": 1.0866311516996447,
      "grad_norm": 5.40625,
      "learning_rate": 4.697342612593841e-06,
      "loss": 1.63671455,
      "memory(GiB)": 117.38,
      "step": 42835,
      "train_speed(iter/s)": 1.638786
    },
    {
      "acc": 0.65135756,
      "epoch": 1.08675799086758,
      "grad_norm": 5.125,
      "learning_rate": 4.696295918451139e-06,
      "loss": 1.56955032,
      "memory(GiB)": 117.38,
      "step": 42840,
      "train_speed(iter/s)": 1.638806
    },
    {
      "acc": 0.66414266,
      "epoch": 1.086884830035515,
      "grad_norm": 5.25,
      "learning_rate": 4.695249237666697e-06,
      "loss": 1.56630802,
      "memory(GiB)": 117.38,
      "step": 42845,
      "train_speed(iter/s)": 1.638828
    },
    {
      "acc": 0.67779994,
      "epoch": 1.08701166920345,
      "grad_norm": 6.09375,
      "learning_rate": 4.694202570286556e-06,
      "loss": 1.53982239,
      "memory(GiB)": 117.38,
      "step": 42850,
      "train_speed(iter/s)": 1.638848
    },
    {
      "acc": 0.65204387,
      "epoch": 1.0871385083713851,
      "grad_norm": 5.75,
      "learning_rate": 4.693155916356751e-06,
      "loss": 1.65156517,
      "memory(GiB)": 117.38,
      "step": 42855,
      "train_speed(iter/s)": 1.638869
    },
    {
      "acc": 0.63524466,
      "epoch": 1.0872653475393201,
      "grad_norm": 5.71875,
      "learning_rate": 4.692109275923318e-06,
      "loss": 1.68566284,
      "memory(GiB)": 117.38,
      "step": 42860,
      "train_speed(iter/s)": 1.638889
    },
    {
      "acc": 0.65680737,
      "epoch": 1.0873921867072551,
      "grad_norm": 7.625,
      "learning_rate": 4.6910626490322925e-06,
      "loss": 1.65174942,
      "memory(GiB)": 117.38,
      "step": 42865,
      "train_speed(iter/s)": 1.63891
    },
    {
      "acc": 0.65737972,
      "epoch": 1.0875190258751903,
      "grad_norm": 7.375,
      "learning_rate": 4.690016035729714e-06,
      "loss": 1.58031292,
      "memory(GiB)": 117.38,
      "step": 42870,
      "train_speed(iter/s)": 1.638931
    },
    {
      "acc": 0.6521378,
      "epoch": 1.0876458650431253,
      "grad_norm": 7.28125,
      "learning_rate": 4.688969436061612e-06,
      "loss": 1.63118896,
      "memory(GiB)": 117.38,
      "step": 42875,
      "train_speed(iter/s)": 1.638953
    },
    {
      "acc": 0.66498389,
      "epoch": 1.0877727042110603,
      "grad_norm": 6.6875,
      "learning_rate": 4.687922850074022e-06,
      "loss": 1.56488295,
      "memory(GiB)": 117.38,
      "step": 42880,
      "train_speed(iter/s)": 1.638975
    },
    {
      "acc": 0.66012826,
      "epoch": 1.0878995433789955,
      "grad_norm": 5.34375,
      "learning_rate": 4.686876277812981e-06,
      "loss": 1.66019077,
      "memory(GiB)": 117.38,
      "step": 42885,
      "train_speed(iter/s)": 1.638996
    },
    {
      "acc": 0.64414916,
      "epoch": 1.0880263825469305,
      "grad_norm": 5.65625,
      "learning_rate": 4.685829719324519e-06,
      "loss": 1.6694931,
      "memory(GiB)": 117.38,
      "step": 42890,
      "train_speed(iter/s)": 1.639017
    },
    {
      "acc": 0.64602871,
      "epoch": 1.0881532217148655,
      "grad_norm": 5.84375,
      "learning_rate": 4.6847831746546664e-06,
      "loss": 1.68436737,
      "memory(GiB)": 117.38,
      "step": 42895,
      "train_speed(iter/s)": 1.639039
    },
    {
      "acc": 0.65834413,
      "epoch": 1.0882800608828007,
      "grad_norm": 9.0625,
      "learning_rate": 4.683736643849459e-06,
      "loss": 1.58339367,
      "memory(GiB)": 117.38,
      "step": 42900,
      "train_speed(iter/s)": 1.639061
    },
    {
      "acc": 0.65892487,
      "epoch": 1.0884069000507357,
      "grad_norm": 5.8125,
      "learning_rate": 4.6826901269549255e-06,
      "loss": 1.56407623,
      "memory(GiB)": 117.38,
      "step": 42905,
      "train_speed(iter/s)": 1.639082
    },
    {
      "acc": 0.64861221,
      "epoch": 1.0885337392186707,
      "grad_norm": 7.21875,
      "learning_rate": 4.681643624017097e-06,
      "loss": 1.61880608,
      "memory(GiB)": 117.38,
      "step": 42910,
      "train_speed(iter/s)": 1.639104
    },
    {
      "acc": 0.67106977,
      "epoch": 1.0886605783866057,
      "grad_norm": 6.0,
      "learning_rate": 4.680597135082002e-06,
      "loss": 1.51148872,
      "memory(GiB)": 117.38,
      "step": 42915,
      "train_speed(iter/s)": 1.639126
    },
    {
      "acc": 0.67631531,
      "epoch": 1.088787417554541,
      "grad_norm": 6.96875,
      "learning_rate": 4.679550660195673e-06,
      "loss": 1.54404764,
      "memory(GiB)": 117.38,
      "step": 42920,
      "train_speed(iter/s)": 1.639147
    },
    {
      "acc": 0.65046248,
      "epoch": 1.0889142567224759,
      "grad_norm": 8.0,
      "learning_rate": 4.6785041994041345e-06,
      "loss": 1.62483177,
      "memory(GiB)": 117.38,
      "step": 42925,
      "train_speed(iter/s)": 1.63917
    },
    {
      "acc": 0.64308729,
      "epoch": 1.0890410958904109,
      "grad_norm": 5.59375,
      "learning_rate": 4.6774577527534195e-06,
      "loss": 1.64539547,
      "memory(GiB)": 117.38,
      "step": 42930,
      "train_speed(iter/s)": 1.639192
    },
    {
      "acc": 0.66263704,
      "epoch": 1.089167935058346,
      "grad_norm": 5.03125,
      "learning_rate": 4.676411320289551e-06,
      "loss": 1.61424751,
      "memory(GiB)": 117.38,
      "step": 42935,
      "train_speed(iter/s)": 1.639214
    },
    {
      "acc": 0.65977211,
      "epoch": 1.089294774226281,
      "grad_norm": 7.28125,
      "learning_rate": 4.675364902058556e-06,
      "loss": 1.596766,
      "memory(GiB)": 117.38,
      "step": 42940,
      "train_speed(iter/s)": 1.639236
    },
    {
      "acc": 0.68072262,
      "epoch": 1.089421613394216,
      "grad_norm": 6.125,
      "learning_rate": 4.674318498106464e-06,
      "loss": 1.58252411,
      "memory(GiB)": 117.38,
      "step": 42945,
      "train_speed(iter/s)": 1.639258
    },
    {
      "acc": 0.66876469,
      "epoch": 1.0895484525621513,
      "grad_norm": 7.21875,
      "learning_rate": 4.6732721084792985e-06,
      "loss": 1.57067003,
      "memory(GiB)": 117.38,
      "step": 42950,
      "train_speed(iter/s)": 1.63928
    },
    {
      "acc": 0.64119525,
      "epoch": 1.0896752917300863,
      "grad_norm": 6.6875,
      "learning_rate": 4.672225733223084e-06,
      "loss": 1.67180595,
      "memory(GiB)": 117.38,
      "step": 42955,
      "train_speed(iter/s)": 1.639302
    },
    {
      "acc": 0.66922359,
      "epoch": 1.0898021308980212,
      "grad_norm": 6.28125,
      "learning_rate": 4.671179372383844e-06,
      "loss": 1.56107616,
      "memory(GiB)": 117.38,
      "step": 42960,
      "train_speed(iter/s)": 1.639323
    },
    {
      "acc": 0.65076776,
      "epoch": 1.0899289700659565,
      "grad_norm": 5.78125,
      "learning_rate": 4.670133026007604e-06,
      "loss": 1.66118851,
      "memory(GiB)": 117.38,
      "step": 42965,
      "train_speed(iter/s)": 1.639344
    },
    {
      "acc": 0.65551682,
      "epoch": 1.0900558092338914,
      "grad_norm": 6.46875,
      "learning_rate": 4.669086694140388e-06,
      "loss": 1.64479065,
      "memory(GiB)": 117.38,
      "step": 42970,
      "train_speed(iter/s)": 1.639364
    },
    {
      "acc": 0.65045419,
      "epoch": 1.0901826484018264,
      "grad_norm": 4.90625,
      "learning_rate": 4.668040376828214e-06,
      "loss": 1.57832069,
      "memory(GiB)": 117.38,
      "step": 42975,
      "train_speed(iter/s)": 1.639384
    },
    {
      "acc": 0.67459941,
      "epoch": 1.0903094875697614,
      "grad_norm": 5.625,
      "learning_rate": 4.666994074117108e-06,
      "loss": 1.59497089,
      "memory(GiB)": 117.38,
      "step": 42980,
      "train_speed(iter/s)": 1.639405
    },
    {
      "acc": 0.64645042,
      "epoch": 1.0904363267376966,
      "grad_norm": 5.53125,
      "learning_rate": 4.665947786053088e-06,
      "loss": 1.64929523,
      "memory(GiB)": 117.38,
      "step": 42985,
      "train_speed(iter/s)": 1.639424
    },
    {
      "acc": 0.65418391,
      "epoch": 1.0905631659056316,
      "grad_norm": 5.59375,
      "learning_rate": 4.664901512682179e-06,
      "loss": 1.60165024,
      "memory(GiB)": 117.38,
      "step": 42990,
      "train_speed(iter/s)": 1.639446
    },
    {
      "acc": 0.6610569,
      "epoch": 1.0906900050735666,
      "grad_norm": 7.46875,
      "learning_rate": 4.663855254050394e-06,
      "loss": 1.5978941,
      "memory(GiB)": 117.38,
      "step": 42995,
      "train_speed(iter/s)": 1.639466
    },
    {
      "acc": 0.64211078,
      "epoch": 1.0908168442415018,
      "grad_norm": 5.84375,
      "learning_rate": 4.662809010203757e-06,
      "loss": 1.6563961,
      "memory(GiB)": 117.38,
      "step": 43000,
      "train_speed(iter/s)": 1.639487
    },
    {
      "epoch": 1.0908168442415018,
      "eval_acc": 0.6461507297585709,
      "eval_loss": 1.5734246969223022,
      "eval_runtime": 58.6033,
      "eval_samples_per_second": 108.697,
      "eval_steps_per_second": 27.183,
      "step": 43000
    },
    {
      "acc": 0.6504776,
      "epoch": 1.0909436834094368,
      "grad_norm": 5.75,
      "learning_rate": 4.661762781188284e-06,
      "loss": 1.60215588,
      "memory(GiB)": 117.38,
      "step": 43005,
      "train_speed(iter/s)": 1.635586
    },
    {
      "acc": 0.65855298,
      "epoch": 1.0910705225773718,
      "grad_norm": 4.9375,
      "learning_rate": 4.660716567049997e-06,
      "loss": 1.62584953,
      "memory(GiB)": 117.38,
      "step": 43010,
      "train_speed(iter/s)": 1.635607
    },
    {
      "acc": 0.65604277,
      "epoch": 1.091197361745307,
      "grad_norm": 6.78125,
      "learning_rate": 4.659670367834908e-06,
      "loss": 1.61244316,
      "memory(GiB)": 117.38,
      "step": 43015,
      "train_speed(iter/s)": 1.635627
    },
    {
      "acc": 0.65257726,
      "epoch": 1.091324200913242,
      "grad_norm": 6.625,
      "learning_rate": 4.658624183589035e-06,
      "loss": 1.61347485,
      "memory(GiB)": 117.38,
      "step": 43020,
      "train_speed(iter/s)": 1.635649
    },
    {
      "acc": 0.66370211,
      "epoch": 1.091451040081177,
      "grad_norm": 4.96875,
      "learning_rate": 4.657578014358395e-06,
      "loss": 1.58295498,
      "memory(GiB)": 117.38,
      "step": 43025,
      "train_speed(iter/s)": 1.63567
    },
    {
      "acc": 0.64700222,
      "epoch": 1.0915778792491122,
      "grad_norm": 5.3125,
      "learning_rate": 4.656531860189005e-06,
      "loss": 1.60024452,
      "memory(GiB)": 117.38,
      "step": 43030,
      "train_speed(iter/s)": 1.635691
    },
    {
      "acc": 0.65322824,
      "epoch": 1.0917047184170472,
      "grad_norm": 5.625,
      "learning_rate": 4.655485721126875e-06,
      "loss": 1.61180763,
      "memory(GiB)": 117.38,
      "step": 43035,
      "train_speed(iter/s)": 1.635713
    },
    {
      "acc": 0.64594069,
      "epoch": 1.0918315575849822,
      "grad_norm": 8.5625,
      "learning_rate": 4.6544395972180214e-06,
      "loss": 1.66605377,
      "memory(GiB)": 117.38,
      "step": 43040,
      "train_speed(iter/s)": 1.635734
    },
    {
      "acc": 0.64300685,
      "epoch": 1.0919583967529174,
      "grad_norm": 5.4375,
      "learning_rate": 4.653393488508457e-06,
      "loss": 1.63638954,
      "memory(GiB)": 117.38,
      "step": 43045,
      "train_speed(iter/s)": 1.635755
    },
    {
      "acc": 0.64903917,
      "epoch": 1.0920852359208524,
      "grad_norm": 6.3125,
      "learning_rate": 4.652347395044197e-06,
      "loss": 1.59003773,
      "memory(GiB)": 117.38,
      "step": 43050,
      "train_speed(iter/s)": 1.635776
    },
    {
      "acc": 0.65993395,
      "epoch": 1.0922120750887874,
      "grad_norm": 4.90625,
      "learning_rate": 4.651301316871247e-06,
      "loss": 1.61316166,
      "memory(GiB)": 117.38,
      "step": 43055,
      "train_speed(iter/s)": 1.635796
    },
    {
      "acc": 0.6553215,
      "epoch": 1.0923389142567226,
      "grad_norm": 5.4375,
      "learning_rate": 4.6502552540356235e-06,
      "loss": 1.58458881,
      "memory(GiB)": 117.38,
      "step": 43060,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.66177025,
      "epoch": 1.0924657534246576,
      "grad_norm": 5.78125,
      "learning_rate": 4.649209206583335e-06,
      "loss": 1.6367878,
      "memory(GiB)": 117.38,
      "step": 43065,
      "train_speed(iter/s)": 1.635839
    },
    {
      "acc": 0.6725316,
      "epoch": 1.0925925925925926,
      "grad_norm": 5.34375,
      "learning_rate": 4.648163174560393e-06,
      "loss": 1.55871553,
      "memory(GiB)": 117.38,
      "step": 43070,
      "train_speed(iter/s)": 1.635861
    },
    {
      "acc": 0.66052008,
      "epoch": 1.0927194317605275,
      "grad_norm": 5.375,
      "learning_rate": 4.647117158012804e-06,
      "loss": 1.58466578,
      "memory(GiB)": 117.38,
      "step": 43075,
      "train_speed(iter/s)": 1.635881
    },
    {
      "acc": 0.65092411,
      "epoch": 1.0928462709284628,
      "grad_norm": 5.625,
      "learning_rate": 4.646071156986579e-06,
      "loss": 1.55119801,
      "memory(GiB)": 117.38,
      "step": 43080,
      "train_speed(iter/s)": 1.635903
    },
    {
      "acc": 0.6483139,
      "epoch": 1.0929731100963977,
      "grad_norm": 5.96875,
      "learning_rate": 4.645025171527723e-06,
      "loss": 1.64346695,
      "memory(GiB)": 117.38,
      "step": 43085,
      "train_speed(iter/s)": 1.635924
    },
    {
      "acc": 0.64516344,
      "epoch": 1.0930999492643327,
      "grad_norm": 5.46875,
      "learning_rate": 4.643979201682247e-06,
      "loss": 1.62893448,
      "memory(GiB)": 117.38,
      "step": 43090,
      "train_speed(iter/s)": 1.635945
    },
    {
      "acc": 0.64851823,
      "epoch": 1.093226788432268,
      "grad_norm": 7.28125,
      "learning_rate": 4.642933247496155e-06,
      "loss": 1.65581131,
      "memory(GiB)": 117.38,
      "step": 43095,
      "train_speed(iter/s)": 1.635966
    },
    {
      "acc": 0.65348835,
      "epoch": 1.093353627600203,
      "grad_norm": 6.375,
      "learning_rate": 4.641887309015451e-06,
      "loss": 1.63301296,
      "memory(GiB)": 117.38,
      "step": 43100,
      "train_speed(iter/s)": 1.635986
    },
    {
      "acc": 0.66231995,
      "epoch": 1.093480466768138,
      "grad_norm": 4.875,
      "learning_rate": 4.640841386286143e-06,
      "loss": 1.52981148,
      "memory(GiB)": 117.38,
      "step": 43105,
      "train_speed(iter/s)": 1.636006
    },
    {
      "acc": 0.65333753,
      "epoch": 1.0936073059360731,
      "grad_norm": 5.875,
      "learning_rate": 4.639795479354236e-06,
      "loss": 1.65804787,
      "memory(GiB)": 117.38,
      "step": 43110,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.66842785,
      "epoch": 1.0937341451040081,
      "grad_norm": 8.5,
      "learning_rate": 4.6387495882657295e-06,
      "loss": 1.6123127,
      "memory(GiB)": 117.38,
      "step": 43115,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.65867653,
      "epoch": 1.0938609842719431,
      "grad_norm": 5.5,
      "learning_rate": 4.63770371306663e-06,
      "loss": 1.60763702,
      "memory(GiB)": 117.38,
      "step": 43120,
      "train_speed(iter/s)": 1.63607
    },
    {
      "acc": 0.65375547,
      "epoch": 1.0939878234398783,
      "grad_norm": 5.90625,
      "learning_rate": 4.636657853802939e-06,
      "loss": 1.53959503,
      "memory(GiB)": 117.38,
      "step": 43125,
      "train_speed(iter/s)": 1.636091
    },
    {
      "acc": 0.67584362,
      "epoch": 1.0941146626078133,
      "grad_norm": 6.0625,
      "learning_rate": 4.635612010520659e-06,
      "loss": 1.62089291,
      "memory(GiB)": 117.38,
      "step": 43130,
      "train_speed(iter/s)": 1.636113
    },
    {
      "acc": 0.65151534,
      "epoch": 1.0942415017757483,
      "grad_norm": 5.9375,
      "learning_rate": 4.6345661832657866e-06,
      "loss": 1.62925224,
      "memory(GiB)": 117.38,
      "step": 43135,
      "train_speed(iter/s)": 1.636134
    },
    {
      "acc": 0.66400709,
      "epoch": 1.0943683409436833,
      "grad_norm": 5.375,
      "learning_rate": 4.633520372084327e-06,
      "loss": 1.55168896,
      "memory(GiB)": 117.38,
      "step": 43140,
      "train_speed(iter/s)": 1.636155
    },
    {
      "acc": 0.64783411,
      "epoch": 1.0944951801116185,
      "grad_norm": 5.625,
      "learning_rate": 4.632474577022276e-06,
      "loss": 1.66214561,
      "memory(GiB)": 117.38,
      "step": 43145,
      "train_speed(iter/s)": 1.636177
    },
    {
      "acc": 0.66868038,
      "epoch": 1.0946220192795535,
      "grad_norm": 5.46875,
      "learning_rate": 4.631428798125637e-06,
      "loss": 1.58520069,
      "memory(GiB)": 117.38,
      "step": 43150,
      "train_speed(iter/s)": 1.636197
    },
    {
      "acc": 0.65542531,
      "epoch": 1.0947488584474885,
      "grad_norm": 4.75,
      "learning_rate": 4.630383035440403e-06,
      "loss": 1.60485611,
      "memory(GiB)": 117.38,
      "step": 43155,
      "train_speed(iter/s)": 1.636218
    },
    {
      "acc": 0.66994524,
      "epoch": 1.0948756976154237,
      "grad_norm": 5.40625,
      "learning_rate": 4.6293372890125724e-06,
      "loss": 1.59670334,
      "memory(GiB)": 117.38,
      "step": 43160,
      "train_speed(iter/s)": 1.636235
    },
    {
      "acc": 0.65548763,
      "epoch": 1.0950025367833587,
      "grad_norm": 4.78125,
      "learning_rate": 4.628291558888144e-06,
      "loss": 1.63418484,
      "memory(GiB)": 117.38,
      "step": 43165,
      "train_speed(iter/s)": 1.636255
    },
    {
      "acc": 0.66280289,
      "epoch": 1.0951293759512937,
      "grad_norm": 5.9375,
      "learning_rate": 4.627245845113113e-06,
      "loss": 1.59679852,
      "memory(GiB)": 117.38,
      "step": 43170,
      "train_speed(iter/s)": 1.636277
    },
    {
      "acc": 0.65032034,
      "epoch": 1.0952562151192289,
      "grad_norm": 6.21875,
      "learning_rate": 4.626200147733474e-06,
      "loss": 1.59159489,
      "memory(GiB)": 117.38,
      "step": 43175,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.6477489,
      "epoch": 1.0953830542871639,
      "grad_norm": 5.625,
      "learning_rate": 4.62515446679522e-06,
      "loss": 1.66620884,
      "memory(GiB)": 117.38,
      "step": 43180,
      "train_speed(iter/s)": 1.636318
    },
    {
      "acc": 0.67817831,
      "epoch": 1.0955098934550989,
      "grad_norm": 5.375,
      "learning_rate": 4.624108802344347e-06,
      "loss": 1.49521027,
      "memory(GiB)": 117.38,
      "step": 43185,
      "train_speed(iter/s)": 1.63634
    },
    {
      "acc": 0.66555071,
      "epoch": 1.095636732623034,
      "grad_norm": 6.9375,
      "learning_rate": 4.623063154426848e-06,
      "loss": 1.56619835,
      "memory(GiB)": 117.38,
      "step": 43190,
      "train_speed(iter/s)": 1.636361
    },
    {
      "acc": 0.65065045,
      "epoch": 1.095763571790969,
      "grad_norm": 5.78125,
      "learning_rate": 4.622017523088712e-06,
      "loss": 1.59552631,
      "memory(GiB)": 117.38,
      "step": 43195,
      "train_speed(iter/s)": 1.636383
    },
    {
      "acc": 0.65362248,
      "epoch": 1.095890410958904,
      "grad_norm": 6.0,
      "learning_rate": 4.620971908375934e-06,
      "loss": 1.62507133,
      "memory(GiB)": 117.38,
      "step": 43200,
      "train_speed(iter/s)": 1.636404
    },
    {
      "acc": 0.6748086,
      "epoch": 1.0960172501268393,
      "grad_norm": 6.9375,
      "learning_rate": 4.619926310334503e-06,
      "loss": 1.53247128,
      "memory(GiB)": 117.38,
      "step": 43205,
      "train_speed(iter/s)": 1.636425
    },
    {
      "acc": 0.64898806,
      "epoch": 1.0961440892947742,
      "grad_norm": 6.125,
      "learning_rate": 4.618880729010413e-06,
      "loss": 1.59856415,
      "memory(GiB)": 117.38,
      "step": 43210,
      "train_speed(iter/s)": 1.636448
    },
    {
      "acc": 0.66861658,
      "epoch": 1.0962709284627092,
      "grad_norm": 5.59375,
      "learning_rate": 4.617835164449647e-06,
      "loss": 1.52230234,
      "memory(GiB)": 117.38,
      "step": 43215,
      "train_speed(iter/s)": 1.636468
    },
    {
      "acc": 0.66404819,
      "epoch": 1.0963977676306444,
      "grad_norm": 7.21875,
      "learning_rate": 4.616789616698197e-06,
      "loss": 1.57905912,
      "memory(GiB)": 117.38,
      "step": 43220,
      "train_speed(iter/s)": 1.63649
    },
    {
      "acc": 0.63807912,
      "epoch": 1.0965246067985794,
      "grad_norm": 7.59375,
      "learning_rate": 4.61574408580205e-06,
      "loss": 1.66577148,
      "memory(GiB)": 117.38,
      "step": 43225,
      "train_speed(iter/s)": 1.63651
    },
    {
      "acc": 0.66725111,
      "epoch": 1.0966514459665144,
      "grad_norm": 5.78125,
      "learning_rate": 4.614698571807196e-06,
      "loss": 1.54518785,
      "memory(GiB)": 117.38,
      "step": 43230,
      "train_speed(iter/s)": 1.636531
    },
    {
      "acc": 0.65628242,
      "epoch": 1.0967782851344494,
      "grad_norm": 5.71875,
      "learning_rate": 4.6136530747596185e-06,
      "loss": 1.59053173,
      "memory(GiB)": 117.38,
      "step": 43235,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.65458994,
      "epoch": 1.0969051243023846,
      "grad_norm": 5.46875,
      "learning_rate": 4.612607594705301e-06,
      "loss": 1.65995922,
      "memory(GiB)": 117.38,
      "step": 43240,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.6680191,
      "epoch": 1.0970319634703196,
      "grad_norm": 5.96875,
      "learning_rate": 4.611562131690234e-06,
      "loss": 1.56793299,
      "memory(GiB)": 117.38,
      "step": 43245,
      "train_speed(iter/s)": 1.636596
    },
    {
      "acc": 0.65579796,
      "epoch": 1.0971588026382546,
      "grad_norm": 5.78125,
      "learning_rate": 4.610516685760399e-06,
      "loss": 1.57751884,
      "memory(GiB)": 117.38,
      "step": 43250,
      "train_speed(iter/s)": 1.636617
    },
    {
      "acc": 0.66340847,
      "epoch": 1.0972856418061898,
      "grad_norm": 5.5625,
      "learning_rate": 4.6094712569617775e-06,
      "loss": 1.50516148,
      "memory(GiB)": 117.38,
      "step": 43255,
      "train_speed(iter/s)": 1.636637
    },
    {
      "acc": 0.64909182,
      "epoch": 1.0974124809741248,
      "grad_norm": 5.28125,
      "learning_rate": 4.608425845340353e-06,
      "loss": 1.63684902,
      "memory(GiB)": 117.38,
      "step": 43260,
      "train_speed(iter/s)": 1.636659
    },
    {
      "acc": 0.63719807,
      "epoch": 1.0975393201420598,
      "grad_norm": 6.09375,
      "learning_rate": 4.607380450942109e-06,
      "loss": 1.66330795,
      "memory(GiB)": 117.38,
      "step": 43265,
      "train_speed(iter/s)": 1.636682
    },
    {
      "acc": 0.65189552,
      "epoch": 1.097666159309995,
      "grad_norm": 6.15625,
      "learning_rate": 4.606335073813028e-06,
      "loss": 1.64268112,
      "memory(GiB)": 117.38,
      "step": 43270,
      "train_speed(iter/s)": 1.636704
    },
    {
      "acc": 0.64464512,
      "epoch": 1.09779299847793,
      "grad_norm": 5.0625,
      "learning_rate": 4.605289713999085e-06,
      "loss": 1.6694376,
      "memory(GiB)": 117.38,
      "step": 43275,
      "train_speed(iter/s)": 1.636726
    },
    {
      "acc": 0.65150423,
      "epoch": 1.097919837645865,
      "grad_norm": 4.625,
      "learning_rate": 4.604244371546263e-06,
      "loss": 1.60193024,
      "memory(GiB)": 117.38,
      "step": 43280,
      "train_speed(iter/s)": 1.636747
    },
    {
      "acc": 0.64329195,
      "epoch": 1.0980466768138002,
      "grad_norm": 5.34375,
      "learning_rate": 4.603199046500539e-06,
      "loss": 1.6610939,
      "memory(GiB)": 117.38,
      "step": 43285,
      "train_speed(iter/s)": 1.636769
    },
    {
      "acc": 0.65891066,
      "epoch": 1.0981735159817352,
      "grad_norm": 5.75,
      "learning_rate": 4.602153738907896e-06,
      "loss": 1.62164211,
      "memory(GiB)": 117.38,
      "step": 43290,
      "train_speed(iter/s)": 1.63679
    },
    {
      "acc": 0.66268911,
      "epoch": 1.0983003551496702,
      "grad_norm": 5.75,
      "learning_rate": 4.601108448814306e-06,
      "loss": 1.60490112,
      "memory(GiB)": 117.38,
      "step": 43295,
      "train_speed(iter/s)": 1.636812
    },
    {
      "acc": 0.64971757,
      "epoch": 1.0984271943176052,
      "grad_norm": 4.8125,
      "learning_rate": 4.600063176265749e-06,
      "loss": 1.5322319,
      "memory(GiB)": 117.38,
      "step": 43300,
      "train_speed(iter/s)": 1.636833
    },
    {
      "acc": 0.66501102,
      "epoch": 1.0985540334855404,
      "grad_norm": 5.5625,
      "learning_rate": 4.599017921308196e-06,
      "loss": 1.57546043,
      "memory(GiB)": 117.38,
      "step": 43305,
      "train_speed(iter/s)": 1.636854
    },
    {
      "acc": 0.64381585,
      "epoch": 1.0986808726534754,
      "grad_norm": 5.9375,
      "learning_rate": 4.5979726839876285e-06,
      "loss": 1.61664104,
      "memory(GiB)": 117.38,
      "step": 43310,
      "train_speed(iter/s)": 1.636875
    },
    {
      "acc": 0.64273753,
      "epoch": 1.0988077118214103,
      "grad_norm": 5.5,
      "learning_rate": 4.596927464350015e-06,
      "loss": 1.63682766,
      "memory(GiB)": 117.38,
      "step": 43315,
      "train_speed(iter/s)": 1.636897
    },
    {
      "acc": 0.64000797,
      "epoch": 1.0989345509893456,
      "grad_norm": 5.28125,
      "learning_rate": 4.595882262441331e-06,
      "loss": 1.61038322,
      "memory(GiB)": 117.38,
      "step": 43320,
      "train_speed(iter/s)": 1.636588
    },
    {
      "acc": 0.66831989,
      "epoch": 1.0990613901572805,
      "grad_norm": 4.46875,
      "learning_rate": 4.5948370783075505e-06,
      "loss": 1.53623734,
      "memory(GiB)": 117.38,
      "step": 43325,
      "train_speed(iter/s)": 1.636609
    },
    {
      "acc": 0.65933323,
      "epoch": 1.0991882293252155,
      "grad_norm": 6.21875,
      "learning_rate": 4.5937919119946445e-06,
      "loss": 1.630583,
      "memory(GiB)": 117.38,
      "step": 43330,
      "train_speed(iter/s)": 1.63663
    },
    {
      "acc": 0.6712594,
      "epoch": 1.0993150684931507,
      "grad_norm": 6.5,
      "learning_rate": 4.592746763548582e-06,
      "loss": 1.55620079,
      "memory(GiB)": 117.38,
      "step": 43335,
      "train_speed(iter/s)": 1.636651
    },
    {
      "acc": 0.65974803,
      "epoch": 1.0994419076610857,
      "grad_norm": 6.15625,
      "learning_rate": 4.591701633015336e-06,
      "loss": 1.61999149,
      "memory(GiB)": 117.38,
      "step": 43340,
      "train_speed(iter/s)": 1.636672
    },
    {
      "acc": 0.66459017,
      "epoch": 1.0995687468290207,
      "grad_norm": 5.0,
      "learning_rate": 4.590656520440876e-06,
      "loss": 1.62464294,
      "memory(GiB)": 117.38,
      "step": 43345,
      "train_speed(iter/s)": 1.636695
    },
    {
      "acc": 0.67729406,
      "epoch": 1.099695585996956,
      "grad_norm": 5.5625,
      "learning_rate": 4.58961142587117e-06,
      "loss": 1.51495514,
      "memory(GiB)": 117.38,
      "step": 43350,
      "train_speed(iter/s)": 1.636716
    },
    {
      "acc": 0.64763308,
      "epoch": 1.099822425164891,
      "grad_norm": 5.65625,
      "learning_rate": 4.588566349352185e-06,
      "loss": 1.55075903,
      "memory(GiB)": 117.38,
      "step": 43355,
      "train_speed(iter/s)": 1.636736
    },
    {
      "acc": 0.65509777,
      "epoch": 1.099949264332826,
      "grad_norm": 7.375,
      "learning_rate": 4.5875212909298885e-06,
      "loss": 1.56889591,
      "memory(GiB)": 117.38,
      "step": 43360,
      "train_speed(iter/s)": 1.636758
    },
    {
      "acc": 0.66368484,
      "epoch": 1.1000761035007611,
      "grad_norm": 5.34375,
      "learning_rate": 4.586476250650246e-06,
      "loss": 1.55608835,
      "memory(GiB)": 117.38,
      "step": 43365,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.6542634,
      "epoch": 1.1002029426686961,
      "grad_norm": 5.03125,
      "learning_rate": 4.585431228559228e-06,
      "loss": 1.60081081,
      "memory(GiB)": 117.38,
      "step": 43370,
      "train_speed(iter/s)": 1.6368
    },
    {
      "acc": 0.65408893,
      "epoch": 1.100329781836631,
      "grad_norm": 5.40625,
      "learning_rate": 4.584386224702792e-06,
      "loss": 1.55857944,
      "memory(GiB)": 117.38,
      "step": 43375,
      "train_speed(iter/s)": 1.636822
    },
    {
      "acc": 0.6609746,
      "epoch": 1.1004566210045663,
      "grad_norm": 6.53125,
      "learning_rate": 4.583341239126906e-06,
      "loss": 1.60434227,
      "memory(GiB)": 117.38,
      "step": 43380,
      "train_speed(iter/s)": 1.636845
    },
    {
      "acc": 0.65415144,
      "epoch": 1.1005834601725013,
      "grad_norm": 6.375,
      "learning_rate": 4.582296271877534e-06,
      "loss": 1.63812332,
      "memory(GiB)": 117.38,
      "step": 43385,
      "train_speed(iter/s)": 1.636867
    },
    {
      "acc": 0.65546341,
      "epoch": 1.1007102993404363,
      "grad_norm": 6.15625,
      "learning_rate": 4.581251323000636e-06,
      "loss": 1.62109108,
      "memory(GiB)": 117.38,
      "step": 43390,
      "train_speed(iter/s)": 1.636889
    },
    {
      "acc": 0.6368413,
      "epoch": 1.1008371385083713,
      "grad_norm": 5.21875,
      "learning_rate": 4.580206392542175e-06,
      "loss": 1.65205193,
      "memory(GiB)": 117.38,
      "step": 43395,
      "train_speed(iter/s)": 1.636908
    },
    {
      "acc": 0.66631503,
      "epoch": 1.1009639776763065,
      "grad_norm": 5.46875,
      "learning_rate": 4.579161480548109e-06,
      "loss": 1.59916687,
      "memory(GiB)": 117.38,
      "step": 43400,
      "train_speed(iter/s)": 1.63693
    },
    {
      "acc": 0.65419798,
      "epoch": 1.1010908168442415,
      "grad_norm": 4.90625,
      "learning_rate": 4.578116587064402e-06,
      "loss": 1.63244247,
      "memory(GiB)": 117.38,
      "step": 43405,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.66090374,
      "epoch": 1.1012176560121765,
      "grad_norm": 9.3125,
      "learning_rate": 4.577071712137012e-06,
      "loss": 1.61473446,
      "memory(GiB)": 117.38,
      "step": 43410,
      "train_speed(iter/s)": 1.636972
    },
    {
      "acc": 0.65870848,
      "epoch": 1.1013444951801117,
      "grad_norm": 5.9375,
      "learning_rate": 4.576026855811893e-06,
      "loss": 1.61457329,
      "memory(GiB)": 117.38,
      "step": 43415,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.66705761,
      "epoch": 1.1014713343480467,
      "grad_norm": 9.875,
      "learning_rate": 4.5749820181350095e-06,
      "loss": 1.58334522,
      "memory(GiB)": 117.38,
      "step": 43420,
      "train_speed(iter/s)": 1.637015
    },
    {
      "acc": 0.64460411,
      "epoch": 1.1015981735159817,
      "grad_norm": 5.1875,
      "learning_rate": 4.57393719915231e-06,
      "loss": 1.61328239,
      "memory(GiB)": 117.38,
      "step": 43425,
      "train_speed(iter/s)": 1.637036
    },
    {
      "acc": 0.65223541,
      "epoch": 1.1017250126839169,
      "grad_norm": 5.46875,
      "learning_rate": 4.5728923989097604e-06,
      "loss": 1.57075558,
      "memory(GiB)": 117.38,
      "step": 43430,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.65929437,
      "epoch": 1.1018518518518519,
      "grad_norm": 5.5625,
      "learning_rate": 4.571847617453306e-06,
      "loss": 1.53379965,
      "memory(GiB)": 117.38,
      "step": 43435,
      "train_speed(iter/s)": 1.637078
    },
    {
      "acc": 0.65777717,
      "epoch": 1.1019786910197868,
      "grad_norm": 6.09375,
      "learning_rate": 4.570802854828906e-06,
      "loss": 1.62695885,
      "memory(GiB)": 117.38,
      "step": 43440,
      "train_speed(iter/s)": 1.6371
    },
    {
      "acc": 0.66201434,
      "epoch": 1.102105530187722,
      "grad_norm": 5.8125,
      "learning_rate": 4.569758111082512e-06,
      "loss": 1.55977707,
      "memory(GiB)": 117.38,
      "step": 43445,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.66806774,
      "epoch": 1.102232369355657,
      "grad_norm": 6.53125,
      "learning_rate": 4.568713386260078e-06,
      "loss": 1.51805706,
      "memory(GiB)": 117.38,
      "step": 43450,
      "train_speed(iter/s)": 1.637143
    },
    {
      "acc": 0.67697258,
      "epoch": 1.102359208523592,
      "grad_norm": 6.8125,
      "learning_rate": 4.567668680407555e-06,
      "loss": 1.55169106,
      "memory(GiB)": 117.38,
      "step": 43455,
      "train_speed(iter/s)": 1.637165
    },
    {
      "acc": 0.64767594,
      "epoch": 1.102486047691527,
      "grad_norm": 6.53125,
      "learning_rate": 4.566623993570893e-06,
      "loss": 1.64335442,
      "memory(GiB)": 117.38,
      "step": 43460,
      "train_speed(iter/s)": 1.637187
    },
    {
      "acc": 0.63292551,
      "epoch": 1.1026128868594622,
      "grad_norm": 5.71875,
      "learning_rate": 4.565579325796043e-06,
      "loss": 1.69831314,
      "memory(GiB)": 117.38,
      "step": 43465,
      "train_speed(iter/s)": 1.637209
    },
    {
      "acc": 0.66303062,
      "epoch": 1.1027397260273972,
      "grad_norm": 5.1875,
      "learning_rate": 4.564534677128954e-06,
      "loss": 1.5953743,
      "memory(GiB)": 117.38,
      "step": 43470,
      "train_speed(iter/s)": 1.637232
    },
    {
      "acc": 0.65839834,
      "epoch": 1.1028665651953322,
      "grad_norm": 4.84375,
      "learning_rate": 4.563490047615574e-06,
      "loss": 1.6028698,
      "memory(GiB)": 117.38,
      "step": 43475,
      "train_speed(iter/s)": 1.637253
    },
    {
      "acc": 0.65589886,
      "epoch": 1.1029934043632674,
      "grad_norm": 6.1875,
      "learning_rate": 4.56244543730185e-06,
      "loss": 1.56749649,
      "memory(GiB)": 117.38,
      "step": 43480,
      "train_speed(iter/s)": 1.637274
    },
    {
      "acc": 0.65689411,
      "epoch": 1.1031202435312024,
      "grad_norm": 5.1875,
      "learning_rate": 4.561400846233729e-06,
      "loss": 1.54747581,
      "memory(GiB)": 117.38,
      "step": 43485,
      "train_speed(iter/s)": 1.637295
    },
    {
      "acc": 0.65407953,
      "epoch": 1.1032470826991374,
      "grad_norm": 6.03125,
      "learning_rate": 4.56035627445716e-06,
      "loss": 1.55124359,
      "memory(GiB)": 117.38,
      "step": 43490,
      "train_speed(iter/s)": 1.637316
    },
    {
      "acc": 0.64014273,
      "epoch": 1.1033739218670726,
      "grad_norm": 6.375,
      "learning_rate": 4.55931172201808e-06,
      "loss": 1.70528984,
      "memory(GiB)": 117.38,
      "step": 43495,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.64971676,
      "epoch": 1.1035007610350076,
      "grad_norm": 5.78125,
      "learning_rate": 4.558267188962441e-06,
      "loss": 1.56202087,
      "memory(GiB)": 117.38,
      "step": 43500,
      "train_speed(iter/s)": 1.637358
    },
    {
      "acc": 0.65679598,
      "epoch": 1.1036276002029426,
      "grad_norm": 4.875,
      "learning_rate": 4.557222675336182e-06,
      "loss": 1.63533001,
      "memory(GiB)": 117.38,
      "step": 43505,
      "train_speed(iter/s)": 1.637379
    },
    {
      "acc": 0.65194511,
      "epoch": 1.1037544393708778,
      "grad_norm": 6.15625,
      "learning_rate": 4.556178181185249e-06,
      "loss": 1.57605553,
      "memory(GiB)": 117.38,
      "step": 43510,
      "train_speed(iter/s)": 1.637399
    },
    {
      "acc": 0.64925585,
      "epoch": 1.1038812785388128,
      "grad_norm": 5.53125,
      "learning_rate": 4.555133706555579e-06,
      "loss": 1.67549248,
      "memory(GiB)": 117.38,
      "step": 43515,
      "train_speed(iter/s)": 1.637421
    },
    {
      "acc": 0.64526606,
      "epoch": 1.1040081177067478,
      "grad_norm": 6.0,
      "learning_rate": 4.554089251493115e-06,
      "loss": 1.64196129,
      "memory(GiB)": 117.38,
      "step": 43520,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.66295462,
      "epoch": 1.104134956874683,
      "grad_norm": 6.25,
      "learning_rate": 4.553044816043796e-06,
      "loss": 1.55494499,
      "memory(GiB)": 117.38,
      "step": 43525,
      "train_speed(iter/s)": 1.637464
    },
    {
      "acc": 0.65765047,
      "epoch": 1.104261796042618,
      "grad_norm": 6.75,
      "learning_rate": 4.552000400253563e-06,
      "loss": 1.59547186,
      "memory(GiB)": 117.38,
      "step": 43530,
      "train_speed(iter/s)": 1.637486
    },
    {
      "acc": 0.66380029,
      "epoch": 1.104388635210553,
      "grad_norm": 6.15625,
      "learning_rate": 4.550956004168352e-06,
      "loss": 1.61445999,
      "memory(GiB)": 117.38,
      "step": 43535,
      "train_speed(iter/s)": 1.637506
    },
    {
      "acc": 0.65298853,
      "epoch": 1.1045154743784882,
      "grad_norm": 4.78125,
      "learning_rate": 4.5499116278341e-06,
      "loss": 1.64223251,
      "memory(GiB)": 117.38,
      "step": 43540,
      "train_speed(iter/s)": 1.637525
    },
    {
      "acc": 0.65271883,
      "epoch": 1.1046423135464232,
      "grad_norm": 8.125,
      "learning_rate": 4.548867271296745e-06,
      "loss": 1.63382072,
      "memory(GiB)": 117.38,
      "step": 43545,
      "train_speed(iter/s)": 1.637547
    },
    {
      "acc": 0.67285204,
      "epoch": 1.1047691527143582,
      "grad_norm": 5.71875,
      "learning_rate": 4.547822934602222e-06,
      "loss": 1.54750843,
      "memory(GiB)": 117.38,
      "step": 43550,
      "train_speed(iter/s)": 1.637567
    },
    {
      "acc": 0.6586154,
      "epoch": 1.1048959918822931,
      "grad_norm": 6.4375,
      "learning_rate": 4.5467786177964635e-06,
      "loss": 1.59689074,
      "memory(GiB)": 117.38,
      "step": 43555,
      "train_speed(iter/s)": 1.637587
    },
    {
      "acc": 0.65430512,
      "epoch": 1.1050228310502284,
      "grad_norm": 5.53125,
      "learning_rate": 4.545734320925406e-06,
      "loss": 1.61495438,
      "memory(GiB)": 117.38,
      "step": 43560,
      "train_speed(iter/s)": 1.637608
    },
    {
      "acc": 0.66548595,
      "epoch": 1.1051496702181633,
      "grad_norm": 6.4375,
      "learning_rate": 4.544690044034981e-06,
      "loss": 1.60419312,
      "memory(GiB)": 117.38,
      "step": 43565,
      "train_speed(iter/s)": 1.637629
    },
    {
      "acc": 0.65906448,
      "epoch": 1.1052765093860983,
      "grad_norm": 5.3125,
      "learning_rate": 4.543645787171122e-06,
      "loss": 1.59676704,
      "memory(GiB)": 117.38,
      "step": 43570,
      "train_speed(iter/s)": 1.637649
    },
    {
      "acc": 0.64783249,
      "epoch": 1.1054033485540335,
      "grad_norm": 5.40625,
      "learning_rate": 4.5426015503797565e-06,
      "loss": 1.64254856,
      "memory(GiB)": 117.38,
      "step": 43575,
      "train_speed(iter/s)": 1.637669
    },
    {
      "acc": 0.65406318,
      "epoch": 1.1055301877219685,
      "grad_norm": 5.96875,
      "learning_rate": 4.5415573337068185e-06,
      "loss": 1.62269363,
      "memory(GiB)": 117.38,
      "step": 43580,
      "train_speed(iter/s)": 1.63769
    },
    {
      "acc": 0.65632648,
      "epoch": 1.1056570268899035,
      "grad_norm": 6.34375,
      "learning_rate": 4.540513137198233e-06,
      "loss": 1.63387146,
      "memory(GiB)": 117.38,
      "step": 43585,
      "train_speed(iter/s)": 1.63771
    },
    {
      "acc": 0.67137933,
      "epoch": 1.1057838660578387,
      "grad_norm": 5.75,
      "learning_rate": 4.539468960899936e-06,
      "loss": 1.53223467,
      "memory(GiB)": 117.38,
      "step": 43590,
      "train_speed(iter/s)": 1.637731
    },
    {
      "acc": 0.64252062,
      "epoch": 1.1059107052257737,
      "grad_norm": 7.1875,
      "learning_rate": 4.538424804857847e-06,
      "loss": 1.6505209,
      "memory(GiB)": 117.38,
      "step": 43595,
      "train_speed(iter/s)": 1.63775
    },
    {
      "acc": 0.65709219,
      "epoch": 1.1060375443937087,
      "grad_norm": 5.375,
      "learning_rate": 4.537380669117896e-06,
      "loss": 1.62953682,
      "memory(GiB)": 117.38,
      "step": 43600,
      "train_speed(iter/s)": 1.637771
    },
    {
      "acc": 0.65931711,
      "epoch": 1.106164383561644,
      "grad_norm": 7.5625,
      "learning_rate": 4.536336553726008e-06,
      "loss": 1.56825047,
      "memory(GiB)": 117.38,
      "step": 43605,
      "train_speed(iter/s)": 1.637791
    },
    {
      "acc": 0.66197777,
      "epoch": 1.106291222729579,
      "grad_norm": 5.03125,
      "learning_rate": 4.535292458728112e-06,
      "loss": 1.59539576,
      "memory(GiB)": 117.38,
      "step": 43610,
      "train_speed(iter/s)": 1.637812
    },
    {
      "acc": 0.64459934,
      "epoch": 1.106418061897514,
      "grad_norm": 6.21875,
      "learning_rate": 4.534248384170126e-06,
      "loss": 1.61581249,
      "memory(GiB)": 117.38,
      "step": 43615,
      "train_speed(iter/s)": 1.637832
    },
    {
      "acc": 0.65564461,
      "epoch": 1.106544901065449,
      "grad_norm": 6.0,
      "learning_rate": 4.533204330097974e-06,
      "loss": 1.6060112,
      "memory(GiB)": 117.38,
      "step": 43620,
      "train_speed(iter/s)": 1.637852
    },
    {
      "acc": 0.66526031,
      "epoch": 1.106671740233384,
      "grad_norm": 4.59375,
      "learning_rate": 4.532160296557581e-06,
      "loss": 1.60432014,
      "memory(GiB)": 117.38,
      "step": 43625,
      "train_speed(iter/s)": 1.637873
    },
    {
      "acc": 0.67201915,
      "epoch": 1.106798579401319,
      "grad_norm": 8.125,
      "learning_rate": 4.531116283594868e-06,
      "loss": 1.53097973,
      "memory(GiB)": 117.38,
      "step": 43630,
      "train_speed(iter/s)": 1.637895
    },
    {
      "acc": 0.67831602,
      "epoch": 1.106925418569254,
      "grad_norm": 6.21875,
      "learning_rate": 4.530072291255753e-06,
      "loss": 1.55634899,
      "memory(GiB)": 117.38,
      "step": 43635,
      "train_speed(iter/s)": 1.637915
    },
    {
      "acc": 0.6624465,
      "epoch": 1.1070522577371893,
      "grad_norm": 5.5625,
      "learning_rate": 4.529028319586157e-06,
      "loss": 1.59423561,
      "memory(GiB)": 117.38,
      "step": 43640,
      "train_speed(iter/s)": 1.637936
    },
    {
      "acc": 0.64880672,
      "epoch": 1.1071790969051243,
      "grad_norm": 6.34375,
      "learning_rate": 4.527984368631997e-06,
      "loss": 1.69873047,
      "memory(GiB)": 117.38,
      "step": 43645,
      "train_speed(iter/s)": 1.637956
    },
    {
      "acc": 0.65363159,
      "epoch": 1.1073059360730593,
      "grad_norm": 5.5625,
      "learning_rate": 4.526940438439196e-06,
      "loss": 1.5803566,
      "memory(GiB)": 117.38,
      "step": 43650,
      "train_speed(iter/s)": 1.637976
    },
    {
      "acc": 0.66382847,
      "epoch": 1.1074327752409945,
      "grad_norm": 7.03125,
      "learning_rate": 4.525896529053662e-06,
      "loss": 1.61647263,
      "memory(GiB)": 117.38,
      "step": 43655,
      "train_speed(iter/s)": 1.637996
    },
    {
      "acc": 0.65083418,
      "epoch": 1.1075596144089295,
      "grad_norm": 5.28125,
      "learning_rate": 4.524852640521318e-06,
      "loss": 1.57888641,
      "memory(GiB)": 117.38,
      "step": 43660,
      "train_speed(iter/s)": 1.638017
    },
    {
      "acc": 0.65918074,
      "epoch": 1.1076864535768645,
      "grad_norm": 5.40625,
      "learning_rate": 4.523808772888073e-06,
      "loss": 1.53898945,
      "memory(GiB)": 117.38,
      "step": 43665,
      "train_speed(iter/s)": 1.638038
    },
    {
      "acc": 0.65505533,
      "epoch": 1.1078132927447997,
      "grad_norm": 4.90625,
      "learning_rate": 4.522764926199848e-06,
      "loss": 1.61178856,
      "memory(GiB)": 117.38,
      "step": 43670,
      "train_speed(iter/s)": 1.638058
    },
    {
      "acc": 0.64763803,
      "epoch": 1.1079401319127347,
      "grad_norm": 6.46875,
      "learning_rate": 4.5217211005025516e-06,
      "loss": 1.59003477,
      "memory(GiB)": 117.38,
      "step": 43675,
      "train_speed(iter/s)": 1.638079
    },
    {
      "acc": 0.65308132,
      "epoch": 1.1080669710806696,
      "grad_norm": 7.125,
      "learning_rate": 4.520677295842095e-06,
      "loss": 1.55530643,
      "memory(GiB)": 117.38,
      "step": 43680,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.6806448,
      "epoch": 1.1081938102486049,
      "grad_norm": 6.5625,
      "learning_rate": 4.5196335122643915e-06,
      "loss": 1.46125841,
      "memory(GiB)": 117.38,
      "step": 43685,
      "train_speed(iter/s)": 1.63812
    },
    {
      "acc": 0.65832281,
      "epoch": 1.1083206494165398,
      "grad_norm": 5.9375,
      "learning_rate": 4.518589749815352e-06,
      "loss": 1.54655991,
      "memory(GiB)": 117.38,
      "step": 43690,
      "train_speed(iter/s)": 1.638142
    },
    {
      "acc": 0.65218477,
      "epoch": 1.1084474885844748,
      "grad_norm": 5.8125,
      "learning_rate": 4.517546008540884e-06,
      "loss": 1.59491978,
      "memory(GiB)": 117.38,
      "step": 43695,
      "train_speed(iter/s)": 1.638163
    },
    {
      "acc": 0.65494366,
      "epoch": 1.10857432775241,
      "grad_norm": 5.375,
      "learning_rate": 4.5165022884868946e-06,
      "loss": 1.5815609,
      "memory(GiB)": 117.38,
      "step": 43700,
      "train_speed(iter/s)": 1.638184
    },
    {
      "acc": 0.63820348,
      "epoch": 1.108701166920345,
      "grad_norm": 5.90625,
      "learning_rate": 4.515458589699295e-06,
      "loss": 1.6875267,
      "memory(GiB)": 117.38,
      "step": 43705,
      "train_speed(iter/s)": 1.638205
    },
    {
      "acc": 0.66521559,
      "epoch": 1.10882800608828,
      "grad_norm": 6.75,
      "learning_rate": 4.514414912223991e-06,
      "loss": 1.58573027,
      "memory(GiB)": 117.38,
      "step": 43710,
      "train_speed(iter/s)": 1.638226
    },
    {
      "acc": 0.66810308,
      "epoch": 1.108954845256215,
      "grad_norm": 6.0,
      "learning_rate": 4.513371256106885e-06,
      "loss": 1.52283573,
      "memory(GiB)": 117.38,
      "step": 43715,
      "train_speed(iter/s)": 1.638249
    },
    {
      "acc": 0.64753456,
      "epoch": 1.1090816844241502,
      "grad_norm": 6.125,
      "learning_rate": 4.512327621393885e-06,
      "loss": 1.61174202,
      "memory(GiB)": 117.38,
      "step": 43720,
      "train_speed(iter/s)": 1.638269
    },
    {
      "acc": 0.6785759,
      "epoch": 1.1092085235920852,
      "grad_norm": 5.96875,
      "learning_rate": 4.511284008130892e-06,
      "loss": 1.5328721,
      "memory(GiB)": 117.38,
      "step": 43725,
      "train_speed(iter/s)": 1.638291
    },
    {
      "acc": 0.65266247,
      "epoch": 1.1093353627600202,
      "grad_norm": 5.375,
      "learning_rate": 4.510240416363813e-06,
      "loss": 1.59236279,
      "memory(GiB)": 117.38,
      "step": 43730,
      "train_speed(iter/s)": 1.638313
    },
    {
      "acc": 0.64590249,
      "epoch": 1.1094622019279554,
      "grad_norm": 5.15625,
      "learning_rate": 4.5091968461385455e-06,
      "loss": 1.66553326,
      "memory(GiB)": 117.38,
      "step": 43735,
      "train_speed(iter/s)": 1.638334
    },
    {
      "acc": 0.65689449,
      "epoch": 1.1095890410958904,
      "grad_norm": 5.375,
      "learning_rate": 4.508153297500993e-06,
      "loss": 1.58050213,
      "memory(GiB)": 117.38,
      "step": 43740,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.65717506,
      "epoch": 1.1097158802638254,
      "grad_norm": 5.5625,
      "learning_rate": 4.507109770497052e-06,
      "loss": 1.57905378,
      "memory(GiB)": 117.38,
      "step": 43745,
      "train_speed(iter/s)": 1.638377
    },
    {
      "acc": 0.66607375,
      "epoch": 1.1098427194317606,
      "grad_norm": 5.46875,
      "learning_rate": 4.506066265172626e-06,
      "loss": 1.54343758,
      "memory(GiB)": 117.38,
      "step": 43750,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.62807961,
      "epoch": 1.1099695585996956,
      "grad_norm": 5.78125,
      "learning_rate": 4.505022781573611e-06,
      "loss": 1.67012405,
      "memory(GiB)": 117.38,
      "step": 43755,
      "train_speed(iter/s)": 1.638421
    },
    {
      "acc": 0.66743011,
      "epoch": 1.1100963977676306,
      "grad_norm": 5.5,
      "learning_rate": 4.503979319745902e-06,
      "loss": 1.61075878,
      "memory(GiB)": 117.38,
      "step": 43760,
      "train_speed(iter/s)": 1.638442
    },
    {
      "acc": 0.65517569,
      "epoch": 1.1102232369355658,
      "grad_norm": 5.4375,
      "learning_rate": 4.502935879735398e-06,
      "loss": 1.5934227,
      "memory(GiB)": 117.38,
      "step": 43765,
      "train_speed(iter/s)": 1.638463
    },
    {
      "acc": 0.65846057,
      "epoch": 1.1103500761035008,
      "grad_norm": 5.15625,
      "learning_rate": 4.5018924615879956e-06,
      "loss": 1.60735435,
      "memory(GiB)": 117.38,
      "step": 43770,
      "train_speed(iter/s)": 1.638484
    },
    {
      "acc": 0.66134844,
      "epoch": 1.1104769152714358,
      "grad_norm": 5.15625,
      "learning_rate": 4.500849065349584e-06,
      "loss": 1.57719936,
      "memory(GiB)": 117.38,
      "step": 43775,
      "train_speed(iter/s)": 1.638504
    },
    {
      "acc": 0.64201336,
      "epoch": 1.1106037544393708,
      "grad_norm": 6.53125,
      "learning_rate": 4.499805691066059e-06,
      "loss": 1.5882637,
      "memory(GiB)": 117.38,
      "step": 43780,
      "train_speed(iter/s)": 1.638525
    },
    {
      "acc": 0.65040393,
      "epoch": 1.110730593607306,
      "grad_norm": 7.4375,
      "learning_rate": 4.498762338783314e-06,
      "loss": 1.61698227,
      "memory(GiB)": 117.38,
      "step": 43785,
      "train_speed(iter/s)": 1.638546
    },
    {
      "acc": 0.67040863,
      "epoch": 1.110857432775241,
      "grad_norm": 6.15625,
      "learning_rate": 4.49771900854724e-06,
      "loss": 1.55735893,
      "memory(GiB)": 117.38,
      "step": 43790,
      "train_speed(iter/s)": 1.638567
    },
    {
      "acc": 0.64874072,
      "epoch": 1.110984271943176,
      "grad_norm": 7.40625,
      "learning_rate": 4.496675700403724e-06,
      "loss": 1.68709431,
      "memory(GiB)": 117.38,
      "step": 43795,
      "train_speed(iter/s)": 1.638588
    },
    {
      "acc": 0.65509071,
      "epoch": 1.1111111111111112,
      "grad_norm": 5.78125,
      "learning_rate": 4.495632414398659e-06,
      "loss": 1.66659393,
      "memory(GiB)": 117.38,
      "step": 43800,
      "train_speed(iter/s)": 1.638611
    },
    {
      "acc": 0.64641938,
      "epoch": 1.1112379502790461,
      "grad_norm": 5.59375,
      "learning_rate": 4.494589150577932e-06,
      "loss": 1.70870705,
      "memory(GiB)": 117.38,
      "step": 43805,
      "train_speed(iter/s)": 1.638632
    },
    {
      "acc": 0.66988783,
      "epoch": 1.1113647894469811,
      "grad_norm": 6.125,
      "learning_rate": 4.493545908987432e-06,
      "loss": 1.6110878,
      "memory(GiB)": 117.38,
      "step": 43810,
      "train_speed(iter/s)": 1.638654
    },
    {
      "acc": 0.63878813,
      "epoch": 1.1114916286149163,
      "grad_norm": 5.03125,
      "learning_rate": 4.492502689673044e-06,
      "loss": 1.69132462,
      "memory(GiB)": 117.38,
      "step": 43815,
      "train_speed(iter/s)": 1.638675
    },
    {
      "acc": 0.67498226,
      "epoch": 1.1116184677828513,
      "grad_norm": 6.03125,
      "learning_rate": 4.491459492680651e-06,
      "loss": 1.57180214,
      "memory(GiB)": 117.38,
      "step": 43820,
      "train_speed(iter/s)": 1.638696
    },
    {
      "acc": 0.66567869,
      "epoch": 1.1117453069507863,
      "grad_norm": 5.90625,
      "learning_rate": 4.4904163180561425e-06,
      "loss": 1.52184563,
      "memory(GiB)": 117.38,
      "step": 43825,
      "train_speed(iter/s)": 1.638718
    },
    {
      "acc": 0.6372015,
      "epoch": 1.1118721461187215,
      "grad_norm": 5.375,
      "learning_rate": 4.4893731658453996e-06,
      "loss": 1.60513268,
      "memory(GiB)": 117.38,
      "step": 43830,
      "train_speed(iter/s)": 1.63874
    },
    {
      "acc": 0.66937075,
      "epoch": 1.1119989852866565,
      "grad_norm": 5.375,
      "learning_rate": 4.4883300360943035e-06,
      "loss": 1.5709199,
      "memory(GiB)": 117.38,
      "step": 43835,
      "train_speed(iter/s)": 1.638762
    },
    {
      "acc": 0.64969788,
      "epoch": 1.1121258244545915,
      "grad_norm": 5.84375,
      "learning_rate": 4.4872869288487366e-06,
      "loss": 1.60509033,
      "memory(GiB)": 117.38,
      "step": 43840,
      "train_speed(iter/s)": 1.638782
    },
    {
      "acc": 0.6528121,
      "epoch": 1.1122526636225267,
      "grad_norm": 5.0,
      "learning_rate": 4.48624384415458e-06,
      "loss": 1.53557501,
      "memory(GiB)": 117.38,
      "step": 43845,
      "train_speed(iter/s)": 1.638803
    },
    {
      "acc": 0.65376334,
      "epoch": 1.1123795027904617,
      "grad_norm": 5.96875,
      "learning_rate": 4.485200782057715e-06,
      "loss": 1.64941502,
      "memory(GiB)": 117.38,
      "step": 43850,
      "train_speed(iter/s)": 1.638474
    },
    {
      "acc": 0.64608474,
      "epoch": 1.1125063419583967,
      "grad_norm": 6.125,
      "learning_rate": 4.4841577426040145e-06,
      "loss": 1.62714024,
      "memory(GiB)": 117.38,
      "step": 43855,
      "train_speed(iter/s)": 1.638494
    },
    {
      "acc": 0.66031342,
      "epoch": 1.112633181126332,
      "grad_norm": 5.375,
      "learning_rate": 4.483114725839361e-06,
      "loss": 1.5913106,
      "memory(GiB)": 117.38,
      "step": 43860,
      "train_speed(iter/s)": 1.638516
    },
    {
      "acc": 0.6516191,
      "epoch": 1.112760020294267,
      "grad_norm": 6.3125,
      "learning_rate": 4.482071731809629e-06,
      "loss": 1.65043297,
      "memory(GiB)": 117.38,
      "step": 43865,
      "train_speed(iter/s)": 1.638537
    },
    {
      "acc": 0.64568548,
      "epoch": 1.112886859462202,
      "grad_norm": 6.28125,
      "learning_rate": 4.481028760560697e-06,
      "loss": 1.66282692,
      "memory(GiB)": 117.38,
      "step": 43870,
      "train_speed(iter/s)": 1.638557
    },
    {
      "acc": 0.66215296,
      "epoch": 1.1130136986301369,
      "grad_norm": 5.90625,
      "learning_rate": 4.479985812138435e-06,
      "loss": 1.60832958,
      "memory(GiB)": 117.38,
      "step": 43875,
      "train_speed(iter/s)": 1.638579
    },
    {
      "acc": 0.65164309,
      "epoch": 1.113140537798072,
      "grad_norm": 6.0625,
      "learning_rate": 4.478942886588719e-06,
      "loss": 1.58625851,
      "memory(GiB)": 117.38,
      "step": 43880,
      "train_speed(iter/s)": 1.6386
    },
    {
      "acc": 0.65018668,
      "epoch": 1.113267376966007,
      "grad_norm": 8.8125,
      "learning_rate": 4.47789998395742e-06,
      "loss": 1.58408499,
      "memory(GiB)": 117.38,
      "step": 43885,
      "train_speed(iter/s)": 1.638621
    },
    {
      "acc": 0.66350298,
      "epoch": 1.113394216133942,
      "grad_norm": 7.3125,
      "learning_rate": 4.476857104290413e-06,
      "loss": 1.63399525,
      "memory(GiB)": 117.38,
      "step": 43890,
      "train_speed(iter/s)": 1.638642
    },
    {
      "acc": 0.67013855,
      "epoch": 1.1135210553018773,
      "grad_norm": 6.5,
      "learning_rate": 4.4758142476335655e-06,
      "loss": 1.59673882,
      "memory(GiB)": 117.38,
      "step": 43895,
      "train_speed(iter/s)": 1.638662
    },
    {
      "acc": 0.66996784,
      "epoch": 1.1136478944698123,
      "grad_norm": 4.6875,
      "learning_rate": 4.474771414032747e-06,
      "loss": 1.64837151,
      "memory(GiB)": 117.38,
      "step": 43900,
      "train_speed(iter/s)": 1.638683
    },
    {
      "acc": 0.65942335,
      "epoch": 1.1137747336377473,
      "grad_norm": 6.3125,
      "learning_rate": 4.473728603533827e-06,
      "loss": 1.60491714,
      "memory(GiB)": 117.38,
      "step": 43905,
      "train_speed(iter/s)": 1.638364
    },
    {
      "acc": 0.64758348,
      "epoch": 1.1139015728056825,
      "grad_norm": 6.75,
      "learning_rate": 4.472685816182674e-06,
      "loss": 1.62763023,
      "memory(GiB)": 117.38,
      "step": 43910,
      "train_speed(iter/s)": 1.638385
    },
    {
      "acc": 0.65678248,
      "epoch": 1.1140284119736175,
      "grad_norm": 7.21875,
      "learning_rate": 4.471643052025152e-06,
      "loss": 1.64759941,
      "memory(GiB)": 117.38,
      "step": 43915,
      "train_speed(iter/s)": 1.638407
    },
    {
      "acc": 0.64520226,
      "epoch": 1.1141552511415524,
      "grad_norm": 5.78125,
      "learning_rate": 4.470600311107127e-06,
      "loss": 1.67286606,
      "memory(GiB)": 117.38,
      "step": 43920,
      "train_speed(iter/s)": 1.638428
    },
    {
      "acc": 0.6421525,
      "epoch": 1.1142820903094877,
      "grad_norm": 5.6875,
      "learning_rate": 4.469557593474464e-06,
      "loss": 1.57976723,
      "memory(GiB)": 117.38,
      "step": 43925,
      "train_speed(iter/s)": 1.63845
    },
    {
      "acc": 0.64846869,
      "epoch": 1.1144089294774226,
      "grad_norm": 6.0,
      "learning_rate": 4.468514899173027e-06,
      "loss": 1.6131958,
      "memory(GiB)": 117.38,
      "step": 43930,
      "train_speed(iter/s)": 1.638471
    },
    {
      "acc": 0.66371565,
      "epoch": 1.1145357686453576,
      "grad_norm": 5.875,
      "learning_rate": 4.4674722282486775e-06,
      "loss": 1.56924372,
      "memory(GiB)": 117.38,
      "step": 43935,
      "train_speed(iter/s)": 1.638491
    },
    {
      "acc": 0.66313229,
      "epoch": 1.1146626078132926,
      "grad_norm": 5.625,
      "learning_rate": 4.4664295807472765e-06,
      "loss": 1.56594429,
      "memory(GiB)": 117.38,
      "step": 43940,
      "train_speed(iter/s)": 1.638512
    },
    {
      "acc": 0.65459232,
      "epoch": 1.1147894469812278,
      "grad_norm": 7.0,
      "learning_rate": 4.465386956714684e-06,
      "loss": 1.60682793,
      "memory(GiB)": 117.38,
      "step": 43945,
      "train_speed(iter/s)": 1.638532
    },
    {
      "acc": 0.65105801,
      "epoch": 1.1149162861491628,
      "grad_norm": 5.9375,
      "learning_rate": 4.4643443561967625e-06,
      "loss": 1.59797335,
      "memory(GiB)": 117.38,
      "step": 43950,
      "train_speed(iter/s)": 1.638553
    },
    {
      "acc": 0.66231003,
      "epoch": 1.1150431253170978,
      "grad_norm": 6.40625,
      "learning_rate": 4.463301779239366e-06,
      "loss": 1.50973549,
      "memory(GiB)": 117.38,
      "step": 43955,
      "train_speed(iter/s)": 1.638572
    },
    {
      "acc": 0.67661629,
      "epoch": 1.115169964485033,
      "grad_norm": 5.40625,
      "learning_rate": 4.462259225888354e-06,
      "loss": 1.53829308,
      "memory(GiB)": 117.38,
      "step": 43960,
      "train_speed(iter/s)": 1.638594
    },
    {
      "acc": 0.64776192,
      "epoch": 1.115296803652968,
      "grad_norm": 5.6875,
      "learning_rate": 4.4612166961895805e-06,
      "loss": 1.6595377,
      "memory(GiB)": 117.38,
      "step": 43965,
      "train_speed(iter/s)": 1.638615
    },
    {
      "acc": 0.65194893,
      "epoch": 1.115423642820903,
      "grad_norm": 5.0625,
      "learning_rate": 4.460174190188905e-06,
      "loss": 1.59635143,
      "memory(GiB)": 117.38,
      "step": 43970,
      "train_speed(iter/s)": 1.638635
    },
    {
      "acc": 0.64870939,
      "epoch": 1.1155504819888382,
      "grad_norm": 6.0,
      "learning_rate": 4.459131707932177e-06,
      "loss": 1.6761734,
      "memory(GiB)": 117.38,
      "step": 43975,
      "train_speed(iter/s)": 1.638656
    },
    {
      "acc": 0.64989128,
      "epoch": 1.1156773211567732,
      "grad_norm": 5.59375,
      "learning_rate": 4.458089249465251e-06,
      "loss": 1.64217873,
      "memory(GiB)": 117.38,
      "step": 43980,
      "train_speed(iter/s)": 1.638676
    },
    {
      "acc": 0.64520297,
      "epoch": 1.1158041603247082,
      "grad_norm": 4.75,
      "learning_rate": 4.45704681483398e-06,
      "loss": 1.64760551,
      "memory(GiB)": 117.38,
      "step": 43985,
      "train_speed(iter/s)": 1.638696
    },
    {
      "acc": 0.6482317,
      "epoch": 1.1159309994926434,
      "grad_norm": 5.375,
      "learning_rate": 4.456004404084215e-06,
      "loss": 1.62359486,
      "memory(GiB)": 117.38,
      "step": 43990,
      "train_speed(iter/s)": 1.638718
    },
    {
      "acc": 0.64194298,
      "epoch": 1.1160578386605784,
      "grad_norm": 6.96875,
      "learning_rate": 4.454962017261803e-06,
      "loss": 1.67395,
      "memory(GiB)": 117.38,
      "step": 43995,
      "train_speed(iter/s)": 1.638739
    },
    {
      "acc": 0.64670744,
      "epoch": 1.1161846778285134,
      "grad_norm": 6.5,
      "learning_rate": 4.453919654412596e-06,
      "loss": 1.60647163,
      "memory(GiB)": 117.38,
      "step": 44000,
      "train_speed(iter/s)": 1.638761
    },
    {
      "epoch": 1.1161846778285134,
      "eval_acc": 0.646246791839735,
      "eval_loss": 1.573931336402893,
      "eval_runtime": 58.639,
      "eval_samples_per_second": 108.631,
      "eval_steps_per_second": 27.166,
      "step": 44000
    },
    {
      "acc": 0.66043444,
      "epoch": 1.1163115169964486,
      "grad_norm": 5.53125,
      "learning_rate": 4.45287731558244e-06,
      "loss": 1.55361366,
      "memory(GiB)": 117.38,
      "step": 44005,
      "train_speed(iter/s)": 1.634948
    },
    {
      "acc": 0.6583395,
      "epoch": 1.1164383561643836,
      "grad_norm": 4.6875,
      "learning_rate": 4.451835000817185e-06,
      "loss": 1.56284523,
      "memory(GiB)": 117.38,
      "step": 44010,
      "train_speed(iter/s)": 1.634968
    },
    {
      "acc": 0.66295152,
      "epoch": 1.1165651953323186,
      "grad_norm": 6.5625,
      "learning_rate": 4.450792710162672e-06,
      "loss": 1.62032051,
      "memory(GiB)": 117.38,
      "step": 44015,
      "train_speed(iter/s)": 1.634986
    },
    {
      "acc": 0.65016289,
      "epoch": 1.1166920345002538,
      "grad_norm": 5.875,
      "learning_rate": 4.449750443664747e-06,
      "loss": 1.58196907,
      "memory(GiB)": 117.38,
      "step": 44020,
      "train_speed(iter/s)": 1.635006
    },
    {
      "acc": 0.66230268,
      "epoch": 1.1168188736681888,
      "grad_norm": 4.9375,
      "learning_rate": 4.448708201369254e-06,
      "loss": 1.57816582,
      "memory(GiB)": 117.38,
      "step": 44025,
      "train_speed(iter/s)": 1.635026
    },
    {
      "acc": 0.65436254,
      "epoch": 1.1169457128361238,
      "grad_norm": 5.65625,
      "learning_rate": 4.4476659833220374e-06,
      "loss": 1.59933891,
      "memory(GiB)": 117.38,
      "step": 44030,
      "train_speed(iter/s)": 1.635045
    },
    {
      "acc": 0.66842809,
      "epoch": 1.1170725520040587,
      "grad_norm": 5.8125,
      "learning_rate": 4.4466237895689365e-06,
      "loss": 1.56417561,
      "memory(GiB)": 117.38,
      "step": 44035,
      "train_speed(iter/s)": 1.635063
    },
    {
      "acc": 0.65718164,
      "epoch": 1.117199391171994,
      "grad_norm": 6.75,
      "learning_rate": 4.44558162015579e-06,
      "loss": 1.59644241,
      "memory(GiB)": 117.38,
      "step": 44040,
      "train_speed(iter/s)": 1.635076
    },
    {
      "acc": 0.65321913,
      "epoch": 1.117326230339929,
      "grad_norm": 6.34375,
      "learning_rate": 4.444539475128441e-06,
      "loss": 1.63150883,
      "memory(GiB)": 117.38,
      "step": 44045,
      "train_speed(iter/s)": 1.635094
    },
    {
      "acc": 0.65699592,
      "epoch": 1.117453069507864,
      "grad_norm": 5.4375,
      "learning_rate": 4.443497354532726e-06,
      "loss": 1.62423592,
      "memory(GiB)": 117.38,
      "step": 44050,
      "train_speed(iter/s)": 1.635114
    },
    {
      "acc": 0.65642576,
      "epoch": 1.1175799086757991,
      "grad_norm": 6.0,
      "learning_rate": 4.442455258414482e-06,
      "loss": 1.53582563,
      "memory(GiB)": 117.38,
      "step": 44055,
      "train_speed(iter/s)": 1.635135
    },
    {
      "acc": 0.64688973,
      "epoch": 1.1177067478437341,
      "grad_norm": 5.75,
      "learning_rate": 4.441413186819543e-06,
      "loss": 1.59591398,
      "memory(GiB)": 117.38,
      "step": 44060,
      "train_speed(iter/s)": 1.635153
    },
    {
      "acc": 0.65907316,
      "epoch": 1.1178335870116691,
      "grad_norm": 6.78125,
      "learning_rate": 4.440371139793747e-06,
      "loss": 1.54488697,
      "memory(GiB)": 117.38,
      "step": 44065,
      "train_speed(iter/s)": 1.635173
    },
    {
      "acc": 0.65992684,
      "epoch": 1.1179604261796043,
      "grad_norm": 5.375,
      "learning_rate": 4.43932911738293e-06,
      "loss": 1.59141226,
      "memory(GiB)": 117.38,
      "step": 44070,
      "train_speed(iter/s)": 1.635192
    },
    {
      "acc": 0.64759164,
      "epoch": 1.1180872653475393,
      "grad_norm": 7.09375,
      "learning_rate": 4.438287119632917e-06,
      "loss": 1.61874142,
      "memory(GiB)": 117.38,
      "step": 44075,
      "train_speed(iter/s)": 1.635213
    },
    {
      "acc": 0.65413737,
      "epoch": 1.1182141045154743,
      "grad_norm": 9.5,
      "learning_rate": 4.4372451465895465e-06,
      "loss": 1.59834528,
      "memory(GiB)": 117.38,
      "step": 44080,
      "train_speed(iter/s)": 1.635233
    },
    {
      "acc": 0.64904704,
      "epoch": 1.1183409436834095,
      "grad_norm": 5.625,
      "learning_rate": 4.436203198298645e-06,
      "loss": 1.63180313,
      "memory(GiB)": 117.38,
      "step": 44085,
      "train_speed(iter/s)": 1.635254
    },
    {
      "acc": 0.64984627,
      "epoch": 1.1184677828513445,
      "grad_norm": 4.96875,
      "learning_rate": 4.435161274806049e-06,
      "loss": 1.58911619,
      "memory(GiB)": 117.38,
      "step": 44090,
      "train_speed(iter/s)": 1.635274
    },
    {
      "acc": 0.64327478,
      "epoch": 1.1185946220192795,
      "grad_norm": 5.25,
      "learning_rate": 4.4341193761575765e-06,
      "loss": 1.64164276,
      "memory(GiB)": 117.38,
      "step": 44095,
      "train_speed(iter/s)": 1.635294
    },
    {
      "acc": 0.67385597,
      "epoch": 1.1187214611872145,
      "grad_norm": 6.375,
      "learning_rate": 4.433077502399063e-06,
      "loss": 1.61192093,
      "memory(GiB)": 117.38,
      "step": 44100,
      "train_speed(iter/s)": 1.635314
    },
    {
      "acc": 0.6531714,
      "epoch": 1.1188483003551497,
      "grad_norm": 5.75,
      "learning_rate": 4.43203565357633e-06,
      "loss": 1.54096785,
      "memory(GiB)": 117.38,
      "step": 44105,
      "train_speed(iter/s)": 1.635334
    },
    {
      "acc": 0.65698619,
      "epoch": 1.1189751395230847,
      "grad_norm": 6.46875,
      "learning_rate": 4.430993829735208e-06,
      "loss": 1.57295332,
      "memory(GiB)": 117.38,
      "step": 44110,
      "train_speed(iter/s)": 1.635354
    },
    {
      "acc": 0.66614261,
      "epoch": 1.1191019786910197,
      "grad_norm": 5.21875,
      "learning_rate": 4.429952030921516e-06,
      "loss": 1.51668005,
      "memory(GiB)": 117.38,
      "step": 44115,
      "train_speed(iter/s)": 1.635375
    },
    {
      "acc": 0.67182083,
      "epoch": 1.119228817858955,
      "grad_norm": 7.125,
      "learning_rate": 4.428910257181077e-06,
      "loss": 1.61970291,
      "memory(GiB)": 117.38,
      "step": 44120,
      "train_speed(iter/s)": 1.635396
    },
    {
      "acc": 0.66382823,
      "epoch": 1.1193556570268899,
      "grad_norm": 7.25,
      "learning_rate": 4.427868508559717e-06,
      "loss": 1.55633621,
      "memory(GiB)": 117.38,
      "step": 44125,
      "train_speed(iter/s)": 1.635416
    },
    {
      "acc": 0.67406707,
      "epoch": 1.1194824961948249,
      "grad_norm": 6.34375,
      "learning_rate": 4.426826785103256e-06,
      "loss": 1.51318779,
      "memory(GiB)": 117.38,
      "step": 44130,
      "train_speed(iter/s)": 1.635435
    },
    {
      "acc": 0.65523081,
      "epoch": 1.11960933536276,
      "grad_norm": 7.8125,
      "learning_rate": 4.425785086857509e-06,
      "loss": 1.59485798,
      "memory(GiB)": 117.38,
      "step": 44135,
      "train_speed(iter/s)": 1.635456
    },
    {
      "acc": 0.65386038,
      "epoch": 1.119736174530695,
      "grad_norm": 6.09375,
      "learning_rate": 4.424743413868298e-06,
      "loss": 1.62073269,
      "memory(GiB)": 117.38,
      "step": 44140,
      "train_speed(iter/s)": 1.635476
    },
    {
      "acc": 0.66281242,
      "epoch": 1.11986301369863,
      "grad_norm": 6.40625,
      "learning_rate": 4.42370176618144e-06,
      "loss": 1.55912018,
      "memory(GiB)": 117.38,
      "step": 44145,
      "train_speed(iter/s)": 1.635496
    },
    {
      "acc": 0.63929787,
      "epoch": 1.1199898528665653,
      "grad_norm": 5.25,
      "learning_rate": 4.422660143842753e-06,
      "loss": 1.60670071,
      "memory(GiB)": 117.38,
      "step": 44150,
      "train_speed(iter/s)": 1.635517
    },
    {
      "acc": 0.65457773,
      "epoch": 1.1201166920345003,
      "grad_norm": 9.0625,
      "learning_rate": 4.421618546898048e-06,
      "loss": 1.65196648,
      "memory(GiB)": 117.38,
      "step": 44155,
      "train_speed(iter/s)": 1.635535
    },
    {
      "acc": 0.65805984,
      "epoch": 1.1202435312024352,
      "grad_norm": 6.59375,
      "learning_rate": 4.420576975393143e-06,
      "loss": 1.50358963,
      "memory(GiB)": 117.38,
      "step": 44160,
      "train_speed(iter/s)": 1.635555
    },
    {
      "acc": 0.66869898,
      "epoch": 1.1203703703703705,
      "grad_norm": 5.40625,
      "learning_rate": 4.4195354293738484e-06,
      "loss": 1.52080555,
      "memory(GiB)": 117.38,
      "step": 44165,
      "train_speed(iter/s)": 1.635576
    },
    {
      "acc": 0.6514441,
      "epoch": 1.1204972095383054,
      "grad_norm": 6.0,
      "learning_rate": 4.418493908885979e-06,
      "loss": 1.5425602,
      "memory(GiB)": 117.38,
      "step": 44170,
      "train_speed(iter/s)": 1.635596
    },
    {
      "acc": 0.65485201,
      "epoch": 1.1206240487062404,
      "grad_norm": 5.03125,
      "learning_rate": 4.417452413975343e-06,
      "loss": 1.5841114,
      "memory(GiB)": 117.38,
      "step": 44175,
      "train_speed(iter/s)": 1.635617
    },
    {
      "acc": 0.66331625,
      "epoch": 1.1207508878741756,
      "grad_norm": 5.0,
      "learning_rate": 4.4164109446877514e-06,
      "loss": 1.58154135,
      "memory(GiB)": 117.38,
      "step": 44180,
      "train_speed(iter/s)": 1.635638
    },
    {
      "acc": 0.66471424,
      "epoch": 1.1208777270421106,
      "grad_norm": 4.8125,
      "learning_rate": 4.41536950106901e-06,
      "loss": 1.5538332,
      "memory(GiB)": 117.38,
      "step": 44185,
      "train_speed(iter/s)": 1.635658
    },
    {
      "acc": 0.66929636,
      "epoch": 1.1210045662100456,
      "grad_norm": 6.21875,
      "learning_rate": 4.414328083164931e-06,
      "loss": 1.56172695,
      "memory(GiB)": 117.38,
      "step": 44190,
      "train_speed(iter/s)": 1.635677
    },
    {
      "acc": 0.64653416,
      "epoch": 1.1211314053779806,
      "grad_norm": 5.46875,
      "learning_rate": 4.4132866910213154e-06,
      "loss": 1.65548344,
      "memory(GiB)": 117.38,
      "step": 44195,
      "train_speed(iter/s)": 1.635698
    },
    {
      "acc": 0.66073055,
      "epoch": 1.1212582445459158,
      "grad_norm": 5.5,
      "learning_rate": 4.41224532468397e-06,
      "loss": 1.49708891,
      "memory(GiB)": 117.38,
      "step": 44200,
      "train_speed(iter/s)": 1.635718
    },
    {
      "acc": 0.65340195,
      "epoch": 1.1213850837138508,
      "grad_norm": 5.6875,
      "learning_rate": 4.411203984198701e-06,
      "loss": 1.61638412,
      "memory(GiB)": 117.38,
      "step": 44205,
      "train_speed(iter/s)": 1.635737
    },
    {
      "acc": 0.64651895,
      "epoch": 1.1215119228817858,
      "grad_norm": 4.5625,
      "learning_rate": 4.41016266961131e-06,
      "loss": 1.59872532,
      "memory(GiB)": 117.38,
      "step": 44210,
      "train_speed(iter/s)": 1.635757
    },
    {
      "acc": 0.66024466,
      "epoch": 1.121638762049721,
      "grad_norm": 8.1875,
      "learning_rate": 4.409121380967597e-06,
      "loss": 1.63750057,
      "memory(GiB)": 117.38,
      "step": 44215,
      "train_speed(iter/s)": 1.635776
    },
    {
      "acc": 0.66965275,
      "epoch": 1.121765601217656,
      "grad_norm": 9.25,
      "learning_rate": 4.408080118313364e-06,
      "loss": 1.5953516,
      "memory(GiB)": 117.38,
      "step": 44220,
      "train_speed(iter/s)": 1.635797
    },
    {
      "acc": 0.65244808,
      "epoch": 1.121892440385591,
      "grad_norm": 6.53125,
      "learning_rate": 4.40703888169441e-06,
      "loss": 1.63399849,
      "memory(GiB)": 117.38,
      "step": 44225,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.66125412,
      "epoch": 1.1220192795535262,
      "grad_norm": 9.25,
      "learning_rate": 4.4059976711565355e-06,
      "loss": 1.54146194,
      "memory(GiB)": 117.38,
      "step": 44230,
      "train_speed(iter/s)": 1.63584
    },
    {
      "acc": 0.65405045,
      "epoch": 1.1221461187214612,
      "grad_norm": 6.125,
      "learning_rate": 4.404956486745532e-06,
      "loss": 1.56526117,
      "memory(GiB)": 117.38,
      "step": 44235,
      "train_speed(iter/s)": 1.63586
    },
    {
      "acc": 0.65093994,
      "epoch": 1.1222729578893962,
      "grad_norm": 7.0,
      "learning_rate": 4.403915328507201e-06,
      "loss": 1.63668766,
      "memory(GiB)": 117.38,
      "step": 44240,
      "train_speed(iter/s)": 1.635882
    },
    {
      "acc": 0.65194454,
      "epoch": 1.1223997970573314,
      "grad_norm": 5.28125,
      "learning_rate": 4.4028741964873334e-06,
      "loss": 1.5779974,
      "memory(GiB)": 117.38,
      "step": 44245,
      "train_speed(iter/s)": 1.635902
    },
    {
      "acc": 0.64303694,
      "epoch": 1.1225266362252664,
      "grad_norm": 7.625,
      "learning_rate": 4.4018330907317275e-06,
      "loss": 1.61261368,
      "memory(GiB)": 117.38,
      "step": 44250,
      "train_speed(iter/s)": 1.635922
    },
    {
      "acc": 0.64048562,
      "epoch": 1.1226534753932014,
      "grad_norm": 5.96875,
      "learning_rate": 4.400792011286171e-06,
      "loss": 1.72979774,
      "memory(GiB)": 117.38,
      "step": 44255,
      "train_speed(iter/s)": 1.635942
    },
    {
      "acc": 0.66565065,
      "epoch": 1.1227803145611364,
      "grad_norm": 5.65625,
      "learning_rate": 4.3997509581964566e-06,
      "loss": 1.55208635,
      "memory(GiB)": 117.38,
      "step": 44260,
      "train_speed(iter/s)": 1.635961
    },
    {
      "acc": 0.65540981,
      "epoch": 1.1229071537290716,
      "grad_norm": 5.75,
      "learning_rate": 4.398709931508376e-06,
      "loss": 1.54899044,
      "memory(GiB)": 117.38,
      "step": 44265,
      "train_speed(iter/s)": 1.635982
    },
    {
      "acc": 0.65252419,
      "epoch": 1.1230339928970066,
      "grad_norm": 6.9375,
      "learning_rate": 4.397668931267718e-06,
      "loss": 1.55044785,
      "memory(GiB)": 117.38,
      "step": 44270,
      "train_speed(iter/s)": 1.636001
    },
    {
      "acc": 0.65328288,
      "epoch": 1.1231608320649416,
      "grad_norm": 5.40625,
      "learning_rate": 4.396627957520269e-06,
      "loss": 1.63892097,
      "memory(GiB)": 117.38,
      "step": 44275,
      "train_speed(iter/s)": 1.636021
    },
    {
      "acc": 0.63981328,
      "epoch": 1.1232876712328768,
      "grad_norm": 5.90625,
      "learning_rate": 4.395587010311815e-06,
      "loss": 1.63370705,
      "memory(GiB)": 117.38,
      "step": 44280,
      "train_speed(iter/s)": 1.636041
    },
    {
      "acc": 0.66041307,
      "epoch": 1.1234145104008117,
      "grad_norm": 6.28125,
      "learning_rate": 4.394546089688143e-06,
      "loss": 1.58444443,
      "memory(GiB)": 117.38,
      "step": 44285,
      "train_speed(iter/s)": 1.63606
    },
    {
      "acc": 0.65751848,
      "epoch": 1.1235413495687467,
      "grad_norm": 7.0,
      "learning_rate": 4.3935051956950395e-06,
      "loss": 1.6225771,
      "memory(GiB)": 117.38,
      "step": 44290,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.6475522,
      "epoch": 1.123668188736682,
      "grad_norm": 5.46875,
      "learning_rate": 4.3924643283782824e-06,
      "loss": 1.62086678,
      "memory(GiB)": 117.38,
      "step": 44295,
      "train_speed(iter/s)": 1.636103
    },
    {
      "acc": 0.66249762,
      "epoch": 1.123795027904617,
      "grad_norm": 7.4375,
      "learning_rate": 4.391423487783657e-06,
      "loss": 1.59110508,
      "memory(GiB)": 117.38,
      "step": 44300,
      "train_speed(iter/s)": 1.636124
    },
    {
      "acc": 0.66288376,
      "epoch": 1.123921867072552,
      "grad_norm": 6.09375,
      "learning_rate": 4.3903826739569444e-06,
      "loss": 1.58983974,
      "memory(GiB)": 117.38,
      "step": 44305,
      "train_speed(iter/s)": 1.636144
    },
    {
      "acc": 0.66546173,
      "epoch": 1.1240487062404871,
      "grad_norm": 6.34375,
      "learning_rate": 4.389341886943926e-06,
      "loss": 1.56044073,
      "memory(GiB)": 117.38,
      "step": 44310,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.64793382,
      "epoch": 1.1241755454084221,
      "grad_norm": 5.90625,
      "learning_rate": 4.388301126790374e-06,
      "loss": 1.63326569,
      "memory(GiB)": 117.38,
      "step": 44315,
      "train_speed(iter/s)": 1.636183
    },
    {
      "acc": 0.66180568,
      "epoch": 1.1243023845763571,
      "grad_norm": 5.96875,
      "learning_rate": 4.387260393542071e-06,
      "loss": 1.60197639,
      "memory(GiB)": 117.38,
      "step": 44320,
      "train_speed(iter/s)": 1.636202
    },
    {
      "acc": 0.64380836,
      "epoch": 1.1244292237442923,
      "grad_norm": 5.75,
      "learning_rate": 4.38621968724479e-06,
      "loss": 1.6886879,
      "memory(GiB)": 117.38,
      "step": 44325,
      "train_speed(iter/s)": 1.636222
    },
    {
      "acc": 0.65574608,
      "epoch": 1.1245560629122273,
      "grad_norm": 6.0,
      "learning_rate": 4.385179007944311e-06,
      "loss": 1.59371214,
      "memory(GiB)": 117.38,
      "step": 44330,
      "train_speed(iter/s)": 1.636241
    },
    {
      "acc": 0.65164137,
      "epoch": 1.1246829020801623,
      "grad_norm": 8.0625,
      "learning_rate": 4.384138355686402e-06,
      "loss": 1.65738449,
      "memory(GiB)": 117.38,
      "step": 44335,
      "train_speed(iter/s)": 1.636261
    },
    {
      "acc": 0.64402332,
      "epoch": 1.1248097412480975,
      "grad_norm": 5.84375,
      "learning_rate": 4.383097730516837e-06,
      "loss": 1.64249401,
      "memory(GiB)": 117.38,
      "step": 44340,
      "train_speed(iter/s)": 1.636282
    },
    {
      "acc": 0.67568874,
      "epoch": 1.1249365804160325,
      "grad_norm": 6.34375,
      "learning_rate": 4.382057132481389e-06,
      "loss": 1.56319075,
      "memory(GiB)": 117.38,
      "step": 44345,
      "train_speed(iter/s)": 1.636302
    },
    {
      "acc": 0.65463829,
      "epoch": 1.1250634195839675,
      "grad_norm": 6.59375,
      "learning_rate": 4.381016561625829e-06,
      "loss": 1.54930124,
      "memory(GiB)": 117.38,
      "step": 44350,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.6481185,
      "epoch": 1.1251902587519025,
      "grad_norm": 6.1875,
      "learning_rate": 4.379976017995922e-06,
      "loss": 1.68179588,
      "memory(GiB)": 117.38,
      "step": 44355,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.66424837,
      "epoch": 1.1253170979198377,
      "grad_norm": 8.125,
      "learning_rate": 4.378935501637438e-06,
      "loss": 1.62036667,
      "memory(GiB)": 117.38,
      "step": 44360,
      "train_speed(iter/s)": 1.636362
    },
    {
      "acc": 0.65817924,
      "epoch": 1.1254439370877727,
      "grad_norm": 6.75,
      "learning_rate": 4.377895012596144e-06,
      "loss": 1.5880394,
      "memory(GiB)": 117.38,
      "step": 44365,
      "train_speed(iter/s)": 1.636382
    },
    {
      "acc": 0.64536557,
      "epoch": 1.1255707762557077,
      "grad_norm": 5.53125,
      "learning_rate": 4.376854550917805e-06,
      "loss": 1.65076275,
      "memory(GiB)": 117.38,
      "step": 44370,
      "train_speed(iter/s)": 1.636402
    },
    {
      "acc": 0.65267496,
      "epoch": 1.1256976154236429,
      "grad_norm": 5.375,
      "learning_rate": 4.375814116648184e-06,
      "loss": 1.64900436,
      "memory(GiB)": 117.38,
      "step": 44375,
      "train_speed(iter/s)": 1.636422
    },
    {
      "acc": 0.65974469,
      "epoch": 1.1258244545915779,
      "grad_norm": 4.90625,
      "learning_rate": 4.374773709833045e-06,
      "loss": 1.58544083,
      "memory(GiB)": 117.38,
      "step": 44380,
      "train_speed(iter/s)": 1.636442
    },
    {
      "acc": 0.66633081,
      "epoch": 1.1259512937595129,
      "grad_norm": 7.15625,
      "learning_rate": 4.37373333051815e-06,
      "loss": 1.56389456,
      "memory(GiB)": 117.38,
      "step": 44385,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.63738928,
      "epoch": 1.126078132927448,
      "grad_norm": 5.90625,
      "learning_rate": 4.37269297874926e-06,
      "loss": 1.67483444,
      "memory(GiB)": 117.38,
      "step": 44390,
      "train_speed(iter/s)": 1.636483
    },
    {
      "acc": 0.64450359,
      "epoch": 1.126204972095383,
      "grad_norm": 5.5625,
      "learning_rate": 4.371652654572134e-06,
      "loss": 1.68096046,
      "memory(GiB)": 117.38,
      "step": 44395,
      "train_speed(iter/s)": 1.636503
    },
    {
      "acc": 0.65925589,
      "epoch": 1.126331811263318,
      "grad_norm": 6.46875,
      "learning_rate": 4.370612358032529e-06,
      "loss": 1.54828854,
      "memory(GiB)": 117.38,
      "step": 44400,
      "train_speed(iter/s)": 1.636523
    },
    {
      "acc": 0.66103125,
      "epoch": 1.1264586504312533,
      "grad_norm": 6.6875,
      "learning_rate": 4.369572089176201e-06,
      "loss": 1.54554634,
      "memory(GiB)": 117.38,
      "step": 44405,
      "train_speed(iter/s)": 1.636543
    },
    {
      "acc": 0.64993038,
      "epoch": 1.1265854895991883,
      "grad_norm": 6.40625,
      "learning_rate": 4.3685318480489095e-06,
      "loss": 1.60171547,
      "memory(GiB)": 117.38,
      "step": 44410,
      "train_speed(iter/s)": 1.636563
    },
    {
      "acc": 0.66563597,
      "epoch": 1.1267123287671232,
      "grad_norm": 6.1875,
      "learning_rate": 4.367491634696405e-06,
      "loss": 1.55331097,
      "memory(GiB)": 117.38,
      "step": 44415,
      "train_speed(iter/s)": 1.636583
    },
    {
      "acc": 0.64933901,
      "epoch": 1.1268391679350582,
      "grad_norm": 6.75,
      "learning_rate": 4.366451449164442e-06,
      "loss": 1.58267803,
      "memory(GiB)": 117.38,
      "step": 44420,
      "train_speed(iter/s)": 1.636603
    },
    {
      "acc": 0.65524464,
      "epoch": 1.1269660071029934,
      "grad_norm": 5.28125,
      "learning_rate": 4.365411291498774e-06,
      "loss": 1.61520157,
      "memory(GiB)": 117.38,
      "step": 44425,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.64920053,
      "epoch": 1.1270928462709284,
      "grad_norm": 5.78125,
      "learning_rate": 4.364371161745151e-06,
      "loss": 1.63367004,
      "memory(GiB)": 117.38,
      "step": 44430,
      "train_speed(iter/s)": 1.636643
    },
    {
      "acc": 0.64144926,
      "epoch": 1.1272196854388636,
      "grad_norm": 5.5,
      "learning_rate": 4.363331059949321e-06,
      "loss": 1.61619301,
      "memory(GiB)": 117.38,
      "step": 44435,
      "train_speed(iter/s)": 1.636663
    },
    {
      "acc": 0.66180468,
      "epoch": 1.1273465246067986,
      "grad_norm": 5.3125,
      "learning_rate": 4.362290986157034e-06,
      "loss": 1.58090649,
      "memory(GiB)": 117.38,
      "step": 44440,
      "train_speed(iter/s)": 1.636683
    },
    {
      "acc": 0.65206394,
      "epoch": 1.1274733637747336,
      "grad_norm": 6.0625,
      "learning_rate": 4.361250940414036e-06,
      "loss": 1.61410484,
      "memory(GiB)": 117.38,
      "step": 44445,
      "train_speed(iter/s)": 1.636701
    },
    {
      "acc": 0.65910387,
      "epoch": 1.1276002029426686,
      "grad_norm": 6.71875,
      "learning_rate": 4.360210922766076e-06,
      "loss": 1.59003572,
      "memory(GiB)": 117.38,
      "step": 44450,
      "train_speed(iter/s)": 1.636722
    },
    {
      "acc": 0.63463645,
      "epoch": 1.1277270421106038,
      "grad_norm": 5.75,
      "learning_rate": 4.359170933258893e-06,
      "loss": 1.68063622,
      "memory(GiB)": 117.38,
      "step": 44455,
      "train_speed(iter/s)": 1.636741
    },
    {
      "acc": 0.66906643,
      "epoch": 1.1278538812785388,
      "grad_norm": 5.59375,
      "learning_rate": 4.358130971938235e-06,
      "loss": 1.56312609,
      "memory(GiB)": 117.38,
      "step": 44460,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.64765439,
      "epoch": 1.1279807204464738,
      "grad_norm": 5.34375,
      "learning_rate": 4.357091038849841e-06,
      "loss": 1.64965096,
      "memory(GiB)": 117.38,
      "step": 44465,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.67054591,
      "epoch": 1.128107559614409,
      "grad_norm": 5.6875,
      "learning_rate": 4.356051134039455e-06,
      "loss": 1.56810436,
      "memory(GiB)": 117.38,
      "step": 44470,
      "train_speed(iter/s)": 1.6368
    },
    {
      "acc": 0.66476593,
      "epoch": 1.128234398782344,
      "grad_norm": 5.0,
      "learning_rate": 4.3550112575528155e-06,
      "loss": 1.56342506,
      "memory(GiB)": 117.38,
      "step": 44475,
      "train_speed(iter/s)": 1.636819
    },
    {
      "acc": 0.66813889,
      "epoch": 1.128361237950279,
      "grad_norm": 5.375,
      "learning_rate": 4.353971409435659e-06,
      "loss": 1.48909473,
      "memory(GiB)": 117.38,
      "step": 44480,
      "train_speed(iter/s)": 1.636839
    },
    {
      "acc": 0.6668561,
      "epoch": 1.1284880771182142,
      "grad_norm": 6.53125,
      "learning_rate": 4.352931589733725e-06,
      "loss": 1.5471199,
      "memory(GiB)": 117.38,
      "step": 44485,
      "train_speed(iter/s)": 1.636857
    },
    {
      "acc": 0.6454442,
      "epoch": 1.1286149162861492,
      "grad_norm": 7.65625,
      "learning_rate": 4.35189179849275e-06,
      "loss": 1.64262047,
      "memory(GiB)": 117.38,
      "step": 44490,
      "train_speed(iter/s)": 1.636878
    },
    {
      "acc": 0.65689402,
      "epoch": 1.1287417554540842,
      "grad_norm": 6.15625,
      "learning_rate": 4.350852035758466e-06,
      "loss": 1.69725113,
      "memory(GiB)": 117.38,
      "step": 44495,
      "train_speed(iter/s)": 1.636899
    },
    {
      "acc": 0.64023209,
      "epoch": 1.1288685946220194,
      "grad_norm": 4.8125,
      "learning_rate": 4.3498123015766066e-06,
      "loss": 1.61459503,
      "memory(GiB)": 117.38,
      "step": 44500,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.66754217,
      "epoch": 1.1289954337899544,
      "grad_norm": 7.4375,
      "learning_rate": 4.348772595992906e-06,
      "loss": 1.58678741,
      "memory(GiB)": 117.38,
      "step": 44505,
      "train_speed(iter/s)": 1.636939
    },
    {
      "acc": 0.64727921,
      "epoch": 1.1291222729578894,
      "grad_norm": 5.96875,
      "learning_rate": 4.347732919053096e-06,
      "loss": 1.58723469,
      "memory(GiB)": 117.38,
      "step": 44510,
      "train_speed(iter/s)": 1.636959
    },
    {
      "acc": 0.6448493,
      "epoch": 1.1292491121258244,
      "grad_norm": 5.4375,
      "learning_rate": 4.346693270802902e-06,
      "loss": 1.62992878,
      "memory(GiB)": 117.38,
      "step": 44515,
      "train_speed(iter/s)": 1.63698
    },
    {
      "acc": 0.65734653,
      "epoch": 1.1293759512937596,
      "grad_norm": 5.40625,
      "learning_rate": 4.345653651288055e-06,
      "loss": 1.5425087,
      "memory(GiB)": 117.38,
      "step": 44520,
      "train_speed(iter/s)": 1.637
    },
    {
      "acc": 0.64768391,
      "epoch": 1.1295027904616946,
      "grad_norm": 7.375,
      "learning_rate": 4.344614060554281e-06,
      "loss": 1.58433056,
      "memory(GiB)": 117.38,
      "step": 44525,
      "train_speed(iter/s)": 1.637021
    },
    {
      "acc": 0.6630897,
      "epoch": 1.1296296296296295,
      "grad_norm": 5.78125,
      "learning_rate": 4.343574498647311e-06,
      "loss": 1.59602089,
      "memory(GiB)": 117.38,
      "step": 44530,
      "train_speed(iter/s)": 1.637041
    },
    {
      "acc": 0.66031818,
      "epoch": 1.1297564687975648,
      "grad_norm": 5.21875,
      "learning_rate": 4.342534965612861e-06,
      "loss": 1.55400171,
      "memory(GiB)": 117.38,
      "step": 44535,
      "train_speed(iter/s)": 1.637061
    },
    {
      "acc": 0.65776958,
      "epoch": 1.1298833079654997,
      "grad_norm": 5.84375,
      "learning_rate": 4.34149546149666e-06,
      "loss": 1.57825499,
      "memory(GiB)": 117.38,
      "step": 44540,
      "train_speed(iter/s)": 1.63708
    },
    {
      "acc": 0.65936079,
      "epoch": 1.1300101471334347,
      "grad_norm": 5.4375,
      "learning_rate": 4.340455986344428e-06,
      "loss": 1.6558918,
      "memory(GiB)": 117.38,
      "step": 44545,
      "train_speed(iter/s)": 1.6371
    },
    {
      "acc": 0.65897188,
      "epoch": 1.13013698630137,
      "grad_norm": 5.8125,
      "learning_rate": 4.3394165402018875e-06,
      "loss": 1.61213932,
      "memory(GiB)": 117.38,
      "step": 44550,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.65728865,
      "epoch": 1.130263825469305,
      "grad_norm": 5.625,
      "learning_rate": 4.338377123114757e-06,
      "loss": 1.59036427,
      "memory(GiB)": 117.38,
      "step": 44555,
      "train_speed(iter/s)": 1.637139
    },
    {
      "acc": 0.64370222,
      "epoch": 1.13039066463724,
      "grad_norm": 5.15625,
      "learning_rate": 4.337337735128752e-06,
      "loss": 1.64670715,
      "memory(GiB)": 117.38,
      "step": 44560,
      "train_speed(iter/s)": 1.637159
    },
    {
      "acc": 0.65626535,
      "epoch": 1.1305175038051751,
      "grad_norm": 4.46875,
      "learning_rate": 4.336298376289594e-06,
      "loss": 1.60992241,
      "memory(GiB)": 117.38,
      "step": 44565,
      "train_speed(iter/s)": 1.637178
    },
    {
      "acc": 0.64314036,
      "epoch": 1.1306443429731101,
      "grad_norm": 9.1875,
      "learning_rate": 4.335259046642998e-06,
      "loss": 1.67879658,
      "memory(GiB)": 117.38,
      "step": 44570,
      "train_speed(iter/s)": 1.637197
    },
    {
      "acc": 0.65109367,
      "epoch": 1.130771182141045,
      "grad_norm": 4.96875,
      "learning_rate": 4.334219746234675e-06,
      "loss": 1.60179787,
      "memory(GiB)": 117.38,
      "step": 44575,
      "train_speed(iter/s)": 1.637217
    },
    {
      "acc": 0.64804864,
      "epoch": 1.13089802130898,
      "grad_norm": 5.78125,
      "learning_rate": 4.3331804751103395e-06,
      "loss": 1.61155281,
      "memory(GiB)": 117.38,
      "step": 44580,
      "train_speed(iter/s)": 1.637236
    },
    {
      "acc": 0.65168433,
      "epoch": 1.1310248604769153,
      "grad_norm": 7.875,
      "learning_rate": 4.332141233315705e-06,
      "loss": 1.61855316,
      "memory(GiB)": 117.38,
      "step": 44585,
      "train_speed(iter/s)": 1.637255
    },
    {
      "acc": 0.66715488,
      "epoch": 1.1311516996448503,
      "grad_norm": 5.75,
      "learning_rate": 4.331102020896482e-06,
      "loss": 1.59727602,
      "memory(GiB)": 117.38,
      "step": 44590,
      "train_speed(iter/s)": 1.637276
    },
    {
      "acc": 0.66982784,
      "epoch": 1.1312785388127855,
      "grad_norm": 6.4375,
      "learning_rate": 4.330062837898376e-06,
      "loss": 1.52059727,
      "memory(GiB)": 117.38,
      "step": 44595,
      "train_speed(iter/s)": 1.637296
    },
    {
      "acc": 0.65167823,
      "epoch": 1.1314053779807205,
      "grad_norm": 5.90625,
      "learning_rate": 4.3290236843670985e-06,
      "loss": 1.61874866,
      "memory(GiB)": 117.38,
      "step": 44600,
      "train_speed(iter/s)": 1.637317
    },
    {
      "acc": 0.65669365,
      "epoch": 1.1315322171486555,
      "grad_norm": 5.6875,
      "learning_rate": 4.327984560348354e-06,
      "loss": 1.62331429,
      "memory(GiB)": 117.38,
      "step": 44605,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.6701004,
      "epoch": 1.1316590563165905,
      "grad_norm": 5.28125,
      "learning_rate": 4.3269454658878516e-06,
      "loss": 1.57906294,
      "memory(GiB)": 117.38,
      "step": 44610,
      "train_speed(iter/s)": 1.637357
    },
    {
      "acc": 0.66720662,
      "epoch": 1.1317858954845257,
      "grad_norm": 5.09375,
      "learning_rate": 4.325906401031291e-06,
      "loss": 1.60992165,
      "memory(GiB)": 117.38,
      "step": 44615,
      "train_speed(iter/s)": 1.637377
    },
    {
      "acc": 0.65735626,
      "epoch": 1.1319127346524607,
      "grad_norm": 6.09375,
      "learning_rate": 4.324867365824376e-06,
      "loss": 1.5744648,
      "memory(GiB)": 117.38,
      "step": 44620,
      "train_speed(iter/s)": 1.637397
    },
    {
      "acc": 0.6614768,
      "epoch": 1.1320395738203957,
      "grad_norm": 6.65625,
      "learning_rate": 4.323828360312809e-06,
      "loss": 1.59246025,
      "memory(GiB)": 117.38,
      "step": 44625,
      "train_speed(iter/s)": 1.637418
    },
    {
      "acc": 0.65896254,
      "epoch": 1.1321664129883309,
      "grad_norm": 5.6875,
      "learning_rate": 4.32278938454229e-06,
      "loss": 1.55731621,
      "memory(GiB)": 117.38,
      "step": 44630,
      "train_speed(iter/s)": 1.637437
    },
    {
      "acc": 0.66510549,
      "epoch": 1.1322932521562659,
      "grad_norm": 7.6875,
      "learning_rate": 4.321750438558517e-06,
      "loss": 1.52425518,
      "memory(GiB)": 117.38,
      "step": 44635,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.65005569,
      "epoch": 1.1324200913242009,
      "grad_norm": 5.40625,
      "learning_rate": 4.3207115224071874e-06,
      "loss": 1.5399312,
      "memory(GiB)": 117.38,
      "step": 44640,
      "train_speed(iter/s)": 1.637477
    },
    {
      "acc": 0.64005647,
      "epoch": 1.132546930492136,
      "grad_norm": 5.5,
      "learning_rate": 4.319672636133998e-06,
      "loss": 1.67645168,
      "memory(GiB)": 117.38,
      "step": 44645,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.66499214,
      "epoch": 1.132673769660071,
      "grad_norm": 7.59375,
      "learning_rate": 4.318633779784646e-06,
      "loss": 1.62051659,
      "memory(GiB)": 117.38,
      "step": 44650,
      "train_speed(iter/s)": 1.637518
    },
    {
      "acc": 0.65457191,
      "epoch": 1.132800608828006,
      "grad_norm": 5.65625,
      "learning_rate": 4.317594953404818e-06,
      "loss": 1.64559517,
      "memory(GiB)": 117.38,
      "step": 44655,
      "train_speed(iter/s)": 1.637538
    },
    {
      "acc": 0.65417347,
      "epoch": 1.1329274479959413,
      "grad_norm": 4.78125,
      "learning_rate": 4.316556157040213e-06,
      "loss": 1.57564392,
      "memory(GiB)": 117.38,
      "step": 44660,
      "train_speed(iter/s)": 1.637558
    },
    {
      "acc": 0.63608503,
      "epoch": 1.1330542871638762,
      "grad_norm": 5.59375,
      "learning_rate": 4.315517390736519e-06,
      "loss": 1.66183567,
      "memory(GiB)": 117.38,
      "step": 44665,
      "train_speed(iter/s)": 1.637578
    },
    {
      "acc": 0.64407768,
      "epoch": 1.1331811263318112,
      "grad_norm": 4.4375,
      "learning_rate": 4.314478654539429e-06,
      "loss": 1.59842501,
      "memory(GiB)": 117.38,
      "step": 44670,
      "train_speed(iter/s)": 1.637597
    },
    {
      "acc": 0.65451493,
      "epoch": 1.1333079654997462,
      "grad_norm": 5.34375,
      "learning_rate": 4.313439948494625e-06,
      "loss": 1.62349586,
      "memory(GiB)": 117.38,
      "step": 44675,
      "train_speed(iter/s)": 1.637616
    },
    {
      "acc": 0.65971632,
      "epoch": 1.1334348046676814,
      "grad_norm": 7.34375,
      "learning_rate": 4.312401272647799e-06,
      "loss": 1.56838341,
      "memory(GiB)": 117.38,
      "step": 44680,
      "train_speed(iter/s)": 1.637633
    },
    {
      "acc": 0.65503941,
      "epoch": 1.1335616438356164,
      "grad_norm": 5.59375,
      "learning_rate": 4.311362627044633e-06,
      "loss": 1.60436687,
      "memory(GiB)": 117.38,
      "step": 44685,
      "train_speed(iter/s)": 1.637652
    },
    {
      "acc": 0.65877519,
      "epoch": 1.1336884830035514,
      "grad_norm": 5.9375,
      "learning_rate": 4.310324011730816e-06,
      "loss": 1.61761551,
      "memory(GiB)": 117.38,
      "step": 44690,
      "train_speed(iter/s)": 1.637671
    },
    {
      "acc": 0.65009527,
      "epoch": 1.1338153221714866,
      "grad_norm": 5.65625,
      "learning_rate": 4.309285426752027e-06,
      "loss": 1.65259628,
      "memory(GiB)": 117.38,
      "step": 44695,
      "train_speed(iter/s)": 1.63769
    },
    {
      "acc": 0.65982046,
      "epoch": 1.1339421613394216,
      "grad_norm": 5.625,
      "learning_rate": 4.308246872153947e-06,
      "loss": 1.61216545,
      "memory(GiB)": 117.38,
      "step": 44700,
      "train_speed(iter/s)": 1.63771
    },
    {
      "acc": 0.66697783,
      "epoch": 1.1340690005073566,
      "grad_norm": 5.1875,
      "learning_rate": 4.307208347982259e-06,
      "loss": 1.48208141,
      "memory(GiB)": 117.38,
      "step": 44705,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.66580992,
      "epoch": 1.1341958396752918,
      "grad_norm": 6.09375,
      "learning_rate": 4.306169854282643e-06,
      "loss": 1.53322449,
      "memory(GiB)": 117.38,
      "step": 44710,
      "train_speed(iter/s)": 1.637751
    },
    {
      "acc": 0.66776094,
      "epoch": 1.1343226788432268,
      "grad_norm": 7.25,
      "learning_rate": 4.305131391100773e-06,
      "loss": 1.5979948,
      "memory(GiB)": 117.38,
      "step": 44715,
      "train_speed(iter/s)": 1.63777
    },
    {
      "acc": 0.67849522,
      "epoch": 1.1344495180111618,
      "grad_norm": 5.25,
      "learning_rate": 4.304092958482325e-06,
      "loss": 1.5401968,
      "memory(GiB)": 117.38,
      "step": 44720,
      "train_speed(iter/s)": 1.63779
    },
    {
      "acc": 0.65658293,
      "epoch": 1.134576357179097,
      "grad_norm": 7.53125,
      "learning_rate": 4.303054556472978e-06,
      "loss": 1.59548969,
      "memory(GiB)": 117.38,
      "step": 44725,
      "train_speed(iter/s)": 1.63781
    },
    {
      "acc": 0.65887961,
      "epoch": 1.134703196347032,
      "grad_norm": 5.53125,
      "learning_rate": 4.3020161851184036e-06,
      "loss": 1.60931931,
      "memory(GiB)": 117.38,
      "step": 44730,
      "train_speed(iter/s)": 1.637828
    },
    {
      "acc": 0.64138188,
      "epoch": 1.134830035514967,
      "grad_norm": 7.34375,
      "learning_rate": 4.300977844464273e-06,
      "loss": 1.6142828,
      "memory(GiB)": 117.38,
      "step": 44735,
      "train_speed(iter/s)": 1.637846
    },
    {
      "acc": 0.6602459,
      "epoch": 1.134956874682902,
      "grad_norm": 6.125,
      "learning_rate": 4.2999395345562564e-06,
      "loss": 1.55534592,
      "memory(GiB)": 117.38,
      "step": 44740,
      "train_speed(iter/s)": 1.637866
    },
    {
      "acc": 0.64283967,
      "epoch": 1.1350837138508372,
      "grad_norm": 9.125,
      "learning_rate": 4.298901255440025e-06,
      "loss": 1.62984886,
      "memory(GiB)": 117.38,
      "step": 44745,
      "train_speed(iter/s)": 1.637886
    },
    {
      "acc": 0.64952207,
      "epoch": 1.1352105530187722,
      "grad_norm": 8.25,
      "learning_rate": 4.297863007161249e-06,
      "loss": 1.62485161,
      "memory(GiB)": 117.38,
      "step": 44750,
      "train_speed(iter/s)": 1.637906
    },
    {
      "acc": 0.65533204,
      "epoch": 1.1353373921867074,
      "grad_norm": 4.875,
      "learning_rate": 4.29682478976559e-06,
      "loss": 1.54883928,
      "memory(GiB)": 117.38,
      "step": 44755,
      "train_speed(iter/s)": 1.637925
    },
    {
      "acc": 0.65837083,
      "epoch": 1.1354642313546424,
      "grad_norm": 6.15625,
      "learning_rate": 4.295786603298717e-06,
      "loss": 1.56862326,
      "memory(GiB)": 117.38,
      "step": 44760,
      "train_speed(iter/s)": 1.637944
    },
    {
      "acc": 0.65649004,
      "epoch": 1.1355910705225774,
      "grad_norm": 4.96875,
      "learning_rate": 4.294748447806293e-06,
      "loss": 1.58333549,
      "memory(GiB)": 117.38,
      "step": 44765,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.66233821,
      "epoch": 1.1357179096905123,
      "grad_norm": 6.59375,
      "learning_rate": 4.293710323333983e-06,
      "loss": 1.59596853,
      "memory(GiB)": 117.38,
      "step": 44770,
      "train_speed(iter/s)": 1.637982
    },
    {
      "acc": 0.65775008,
      "epoch": 1.1358447488584476,
      "grad_norm": 5.875,
      "learning_rate": 4.292672229927445e-06,
      "loss": 1.57593956,
      "memory(GiB)": 117.38,
      "step": 44775,
      "train_speed(iter/s)": 1.638
    },
    {
      "acc": 0.67099819,
      "epoch": 1.1359715880263825,
      "grad_norm": 5.8125,
      "learning_rate": 4.2916341676323386e-06,
      "loss": 1.56701775,
      "memory(GiB)": 117.38,
      "step": 44780,
      "train_speed(iter/s)": 1.63802
    },
    {
      "acc": 0.66356764,
      "epoch": 1.1360984271943175,
      "grad_norm": 6.03125,
      "learning_rate": 4.290596136494326e-06,
      "loss": 1.54503708,
      "memory(GiB)": 117.38,
      "step": 44785,
      "train_speed(iter/s)": 1.63804
    },
    {
      "acc": 0.64491863,
      "epoch": 1.1362252663622527,
      "grad_norm": 7.15625,
      "learning_rate": 4.289558136559063e-06,
      "loss": 1.67876015,
      "memory(GiB)": 117.38,
      "step": 44790,
      "train_speed(iter/s)": 1.63806
    },
    {
      "acc": 0.65235085,
      "epoch": 1.1363521055301877,
      "grad_norm": 5.5625,
      "learning_rate": 4.288520167872203e-06,
      "loss": 1.58397017,
      "memory(GiB)": 117.38,
      "step": 44795,
      "train_speed(iter/s)": 1.638079
    },
    {
      "acc": 0.64608803,
      "epoch": 1.1364789446981227,
      "grad_norm": 5.21875,
      "learning_rate": 4.287482230479404e-06,
      "loss": 1.70430908,
      "memory(GiB)": 117.38,
      "step": 44800,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.64419556,
      "epoch": 1.136605783866058,
      "grad_norm": 6.0625,
      "learning_rate": 4.286444324426318e-06,
      "loss": 1.63320675,
      "memory(GiB)": 117.38,
      "step": 44805,
      "train_speed(iter/s)": 1.638118
    },
    {
      "acc": 0.65612288,
      "epoch": 1.136732623033993,
      "grad_norm": 5.125,
      "learning_rate": 4.2854064497585964e-06,
      "loss": 1.58696957,
      "memory(GiB)": 117.38,
      "step": 44810,
      "train_speed(iter/s)": 1.638136
    },
    {
      "acc": 0.65463748,
      "epoch": 1.136859462201928,
      "grad_norm": 5.5625,
      "learning_rate": 4.284368606521888e-06,
      "loss": 1.64917889,
      "memory(GiB)": 117.38,
      "step": 44815,
      "train_speed(iter/s)": 1.638155
    },
    {
      "acc": 0.68417792,
      "epoch": 1.1369863013698631,
      "grad_norm": 4.96875,
      "learning_rate": 4.283330794761845e-06,
      "loss": 1.57826042,
      "memory(GiB)": 117.38,
      "step": 44820,
      "train_speed(iter/s)": 1.638175
    },
    {
      "acc": 0.65448837,
      "epoch": 1.137113140537798,
      "grad_norm": 5.875,
      "learning_rate": 4.282293014524112e-06,
      "loss": 1.58738117,
      "memory(GiB)": 117.38,
      "step": 44825,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.65662594,
      "epoch": 1.137239979705733,
      "grad_norm": 5.375,
      "learning_rate": 4.281255265854338e-06,
      "loss": 1.59228792,
      "memory(GiB)": 117.38,
      "step": 44830,
      "train_speed(iter/s)": 1.638213
    },
    {
      "acc": 0.66493349,
      "epoch": 1.137366818873668,
      "grad_norm": 5.1875,
      "learning_rate": 4.280217548798166e-06,
      "loss": 1.54698172,
      "memory(GiB)": 117.38,
      "step": 44835,
      "train_speed(iter/s)": 1.638233
    },
    {
      "acc": 0.6514761,
      "epoch": 1.1374936580416033,
      "grad_norm": 6.9375,
      "learning_rate": 4.279179863401239e-06,
      "loss": 1.6118803,
      "memory(GiB)": 117.38,
      "step": 44840,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.67122731,
      "epoch": 1.1376204972095383,
      "grad_norm": 6.75,
      "learning_rate": 4.278142209709199e-06,
      "loss": 1.53611288,
      "memory(GiB)": 117.38,
      "step": 44845,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.65977073,
      "epoch": 1.1377473363774733,
      "grad_norm": 6.34375,
      "learning_rate": 4.277104587767691e-06,
      "loss": 1.62168236,
      "memory(GiB)": 117.38,
      "step": 44850,
      "train_speed(iter/s)": 1.63829
    },
    {
      "acc": 0.64796233,
      "epoch": 1.1378741755454085,
      "grad_norm": 7.8125,
      "learning_rate": 4.276066997622348e-06,
      "loss": 1.62023373,
      "memory(GiB)": 117.38,
      "step": 44855,
      "train_speed(iter/s)": 1.638311
    },
    {
      "acc": 0.67169561,
      "epoch": 1.1380010147133435,
      "grad_norm": 6.8125,
      "learning_rate": 4.27502943931881e-06,
      "loss": 1.54391232,
      "memory(GiB)": 117.38,
      "step": 44860,
      "train_speed(iter/s)": 1.638332
    },
    {
      "acc": 0.64612837,
      "epoch": 1.1381278538812785,
      "grad_norm": 5.21875,
      "learning_rate": 4.273991912902716e-06,
      "loss": 1.63414402,
      "memory(GiB)": 117.38,
      "step": 44865,
      "train_speed(iter/s)": 1.638353
    },
    {
      "acc": 0.64558144,
      "epoch": 1.1382546930492137,
      "grad_norm": 5.8125,
      "learning_rate": 4.272954418419699e-06,
      "loss": 1.60734348,
      "memory(GiB)": 117.38,
      "step": 44870,
      "train_speed(iter/s)": 1.638373
    },
    {
      "acc": 0.66381435,
      "epoch": 1.1383815322171487,
      "grad_norm": 5.53125,
      "learning_rate": 4.2719169559153905e-06,
      "loss": 1.53794365,
      "memory(GiB)": 117.38,
      "step": 44875,
      "train_speed(iter/s)": 1.638394
    },
    {
      "acc": 0.67569208,
      "epoch": 1.1385083713850837,
      "grad_norm": 5.1875,
      "learning_rate": 4.270879525435426e-06,
      "loss": 1.56903572,
      "memory(GiB)": 117.38,
      "step": 44880,
      "train_speed(iter/s)": 1.638414
    },
    {
      "acc": 0.66070075,
      "epoch": 1.1386352105530189,
      "grad_norm": 5.15625,
      "learning_rate": 4.269842127025435e-06,
      "loss": 1.59961758,
      "memory(GiB)": 117.38,
      "step": 44885,
      "train_speed(iter/s)": 1.638433
    },
    {
      "acc": 0.65325918,
      "epoch": 1.1387620497209539,
      "grad_norm": 5.84375,
      "learning_rate": 4.2688047607310504e-06,
      "loss": 1.6062355,
      "memory(GiB)": 117.38,
      "step": 44890,
      "train_speed(iter/s)": 1.638453
    },
    {
      "acc": 0.64345627,
      "epoch": 1.1388888888888888,
      "grad_norm": 5.625,
      "learning_rate": 4.267767426597893e-06,
      "loss": 1.65838737,
      "memory(GiB)": 117.38,
      "step": 44895,
      "train_speed(iter/s)": 1.638473
    },
    {
      "acc": 0.65645738,
      "epoch": 1.1390157280568238,
      "grad_norm": 4.90625,
      "learning_rate": 4.266730124671594e-06,
      "loss": 1.58498802,
      "memory(GiB)": 117.38,
      "step": 44900,
      "train_speed(iter/s)": 1.638492
    },
    {
      "acc": 0.66112752,
      "epoch": 1.139142567224759,
      "grad_norm": 5.875,
      "learning_rate": 4.265692854997778e-06,
      "loss": 1.57835989,
      "memory(GiB)": 117.38,
      "step": 44905,
      "train_speed(iter/s)": 1.638511
    },
    {
      "acc": 0.66965404,
      "epoch": 1.139269406392694,
      "grad_norm": 5.25,
      "learning_rate": 4.2646556176220714e-06,
      "loss": 1.54745722,
      "memory(GiB)": 117.38,
      "step": 44910,
      "train_speed(iter/s)": 1.638532
    },
    {
      "acc": 0.66337614,
      "epoch": 1.1393962455606292,
      "grad_norm": 5.3125,
      "learning_rate": 4.263618412590092e-06,
      "loss": 1.54854355,
      "memory(GiB)": 117.38,
      "step": 44915,
      "train_speed(iter/s)": 1.63855
    },
    {
      "acc": 0.64248281,
      "epoch": 1.1395230847285642,
      "grad_norm": 5.4375,
      "learning_rate": 4.2625812399474604e-06,
      "loss": 1.64810219,
      "memory(GiB)": 117.38,
      "step": 44920,
      "train_speed(iter/s)": 1.63857
    },
    {
      "acc": 0.66312189,
      "epoch": 1.1396499238964992,
      "grad_norm": 5.21875,
      "learning_rate": 4.2615440997398e-06,
      "loss": 1.6174427,
      "memory(GiB)": 117.38,
      "step": 44925,
      "train_speed(iter/s)": 1.638589
    },
    {
      "acc": 0.66680937,
      "epoch": 1.1397767630644342,
      "grad_norm": 6.125,
      "learning_rate": 4.2605069920127284e-06,
      "loss": 1.52570429,
      "memory(GiB)": 117.38,
      "step": 44930,
      "train_speed(iter/s)": 1.638609
    },
    {
      "acc": 0.66472435,
      "epoch": 1.1399036022323694,
      "grad_norm": 6.09375,
      "learning_rate": 4.25946991681186e-06,
      "loss": 1.56732063,
      "memory(GiB)": 117.38,
      "step": 44935,
      "train_speed(iter/s)": 1.638629
    },
    {
      "acc": 0.65112996,
      "epoch": 1.1400304414003044,
      "grad_norm": 5.4375,
      "learning_rate": 4.258432874182809e-06,
      "loss": 1.61382751,
      "memory(GiB)": 117.38,
      "step": 44940,
      "train_speed(iter/s)": 1.638648
    },
    {
      "acc": 0.65448222,
      "epoch": 1.1401572805682394,
      "grad_norm": 5.53125,
      "learning_rate": 4.2573958641711925e-06,
      "loss": 1.63212433,
      "memory(GiB)": 117.38,
      "step": 44945,
      "train_speed(iter/s)": 1.638668
    },
    {
      "acc": 0.65999322,
      "epoch": 1.1402841197361746,
      "grad_norm": 4.71875,
      "learning_rate": 4.256358886822622e-06,
      "loss": 1.52996912,
      "memory(GiB)": 117.38,
      "step": 44950,
      "train_speed(iter/s)": 1.638687
    },
    {
      "acc": 0.66640196,
      "epoch": 1.1404109589041096,
      "grad_norm": 6.53125,
      "learning_rate": 4.255321942182707e-06,
      "loss": 1.55350838,
      "memory(GiB)": 117.38,
      "step": 44955,
      "train_speed(iter/s)": 1.638707
    },
    {
      "acc": 0.65074863,
      "epoch": 1.1405377980720446,
      "grad_norm": 5.53125,
      "learning_rate": 4.254285030297058e-06,
      "loss": 1.64202518,
      "memory(GiB)": 117.38,
      "step": 44960,
      "train_speed(iter/s)": 1.638725
    },
    {
      "acc": 0.66520491,
      "epoch": 1.1406646372399798,
      "grad_norm": 5.0,
      "learning_rate": 4.2532481512112814e-06,
      "loss": 1.63261967,
      "memory(GiB)": 117.38,
      "step": 44965,
      "train_speed(iter/s)": 1.638744
    },
    {
      "acc": 0.64974203,
      "epoch": 1.1407914764079148,
      "grad_norm": 8.8125,
      "learning_rate": 4.252211304970988e-06,
      "loss": 1.61925392,
      "memory(GiB)": 117.38,
      "step": 44970,
      "train_speed(iter/s)": 1.638765
    },
    {
      "acc": 0.65617852,
      "epoch": 1.1409183155758498,
      "grad_norm": 6.625,
      "learning_rate": 4.251174491621778e-06,
      "loss": 1.59221592,
      "memory(GiB)": 117.38,
      "step": 44975,
      "train_speed(iter/s)": 1.638784
    },
    {
      "acc": 0.65519772,
      "epoch": 1.141045154743785,
      "grad_norm": 5.15625,
      "learning_rate": 4.250137711209258e-06,
      "loss": 1.64379768,
      "memory(GiB)": 117.38,
      "step": 44980,
      "train_speed(iter/s)": 1.638804
    },
    {
      "acc": 0.65432749,
      "epoch": 1.14117199391172,
      "grad_norm": 5.875,
      "learning_rate": 4.249100963779028e-06,
      "loss": 1.69992599,
      "memory(GiB)": 117.38,
      "step": 44985,
      "train_speed(iter/s)": 1.638824
    },
    {
      "acc": 0.66140928,
      "epoch": 1.141298833079655,
      "grad_norm": 6.75,
      "learning_rate": 4.248064249376692e-06,
      "loss": 1.58639002,
      "memory(GiB)": 117.38,
      "step": 44990,
      "train_speed(iter/s)": 1.638842
    },
    {
      "acc": 0.66274366,
      "epoch": 1.14142567224759,
      "grad_norm": 5.34375,
      "learning_rate": 4.2470275680478466e-06,
      "loss": 1.58178024,
      "memory(GiB)": 117.38,
      "step": 44995,
      "train_speed(iter/s)": 1.638861
    },
    {
      "acc": 0.65980358,
      "epoch": 1.1415525114155252,
      "grad_norm": 6.53125,
      "learning_rate": 4.2459909198380886e-06,
      "loss": 1.5451066,
      "memory(GiB)": 117.38,
      "step": 45000,
      "train_speed(iter/s)": 1.63888
    },
    {
      "epoch": 1.1415525114155252,
      "eval_acc": 0.6462634982886332,
      "eval_loss": 1.5736206769943237,
      "eval_runtime": 58.592,
      "eval_samples_per_second": 108.718,
      "eval_steps_per_second": 27.188,
      "step": 45000
    },
    {
      "acc": 0.67257562,
      "epoch": 1.1416793505834602,
      "grad_norm": 5.5625,
      "learning_rate": 4.244954304793019e-06,
      "loss": 1.54084663,
      "memory(GiB)": 117.38,
      "step": 45005,
      "train_speed(iter/s)": 1.635155
    },
    {
      "acc": 0.65244265,
      "epoch": 1.1418061897513951,
      "grad_norm": 5.625,
      "learning_rate": 4.2439177229582304e-06,
      "loss": 1.62374191,
      "memory(GiB)": 117.38,
      "step": 45010,
      "train_speed(iter/s)": 1.635174
    },
    {
      "acc": 0.66174579,
      "epoch": 1.1419330289193304,
      "grad_norm": 5.59375,
      "learning_rate": 4.242881174379313e-06,
      "loss": 1.59536924,
      "memory(GiB)": 117.38,
      "step": 45015,
      "train_speed(iter/s)": 1.635192
    },
    {
      "acc": 0.64588933,
      "epoch": 1.1420598680872653,
      "grad_norm": 6.0,
      "learning_rate": 4.241844659101865e-06,
      "loss": 1.60734749,
      "memory(GiB)": 117.38,
      "step": 45020,
      "train_speed(iter/s)": 1.635212
    },
    {
      "acc": 0.64297466,
      "epoch": 1.1421867072552003,
      "grad_norm": 4.90625,
      "learning_rate": 4.240808177171472e-06,
      "loss": 1.70465965,
      "memory(GiB)": 117.38,
      "step": 45025,
      "train_speed(iter/s)": 1.635231
    },
    {
      "acc": 0.65627313,
      "epoch": 1.1423135464231355,
      "grad_norm": 5.5625,
      "learning_rate": 4.239771728633727e-06,
      "loss": 1.56444607,
      "memory(GiB)": 117.38,
      "step": 45030,
      "train_speed(iter/s)": 1.635251
    },
    {
      "acc": 0.62234554,
      "epoch": 1.1424403855910705,
      "grad_norm": 5.0,
      "learning_rate": 4.238735313534213e-06,
      "loss": 1.70080528,
      "memory(GiB)": 117.38,
      "step": 45035,
      "train_speed(iter/s)": 1.63527
    },
    {
      "acc": 0.65512896,
      "epoch": 1.1425672247590055,
      "grad_norm": 6.25,
      "learning_rate": 4.23769893191852e-06,
      "loss": 1.64819946,
      "memory(GiB)": 117.38,
      "step": 45040,
      "train_speed(iter/s)": 1.63529
    },
    {
      "acc": 0.66054111,
      "epoch": 1.1426940639269407,
      "grad_norm": 4.96875,
      "learning_rate": 4.236662583832229e-06,
      "loss": 1.52770672,
      "memory(GiB)": 117.38,
      "step": 45045,
      "train_speed(iter/s)": 1.635309
    },
    {
      "acc": 0.64748535,
      "epoch": 1.1428209030948757,
      "grad_norm": 5.5,
      "learning_rate": 4.23562626932093e-06,
      "loss": 1.73409805,
      "memory(GiB)": 117.38,
      "step": 45050,
      "train_speed(iter/s)": 1.635329
    },
    {
      "acc": 0.64391332,
      "epoch": 1.1429477422628107,
      "grad_norm": 5.5625,
      "learning_rate": 4.234589988430198e-06,
      "loss": 1.63327446,
      "memory(GiB)": 117.38,
      "step": 45055,
      "train_speed(iter/s)": 1.635349
    },
    {
      "acc": 0.65088005,
      "epoch": 1.1430745814307457,
      "grad_norm": 5.5625,
      "learning_rate": 4.233553741205615e-06,
      "loss": 1.64944038,
      "memory(GiB)": 117.38,
      "step": 45060,
      "train_speed(iter/s)": 1.635367
    },
    {
      "acc": 0.64193854,
      "epoch": 1.143201420598681,
      "grad_norm": 6.0,
      "learning_rate": 4.2325175276927614e-06,
      "loss": 1.59619846,
      "memory(GiB)": 117.38,
      "step": 45065,
      "train_speed(iter/s)": 1.635388
    },
    {
      "acc": 0.6534915,
      "epoch": 1.143328259766616,
      "grad_norm": 6.0625,
      "learning_rate": 4.231481347937214e-06,
      "loss": 1.56089849,
      "memory(GiB)": 117.38,
      "step": 45070,
      "train_speed(iter/s)": 1.635407
    },
    {
      "acc": 0.65843577,
      "epoch": 1.143455098934551,
      "grad_norm": 5.53125,
      "learning_rate": 4.230445201984547e-06,
      "loss": 1.54402351,
      "memory(GiB)": 117.38,
      "step": 45075,
      "train_speed(iter/s)": 1.635428
    },
    {
      "acc": 0.65084324,
      "epoch": 1.143581938102486,
      "grad_norm": 5.03125,
      "learning_rate": 4.229409089880336e-06,
      "loss": 1.61615219,
      "memory(GiB)": 117.38,
      "step": 45080,
      "train_speed(iter/s)": 1.635449
    },
    {
      "acc": 0.64929686,
      "epoch": 1.143708777270421,
      "grad_norm": 6.15625,
      "learning_rate": 4.2283730116701535e-06,
      "loss": 1.64138031,
      "memory(GiB)": 117.38,
      "step": 45085,
      "train_speed(iter/s)": 1.635469
    },
    {
      "acc": 0.66016579,
      "epoch": 1.143835616438356,
      "grad_norm": 6.25,
      "learning_rate": 4.227336967399573e-06,
      "loss": 1.61218834,
      "memory(GiB)": 117.38,
      "step": 45090,
      "train_speed(iter/s)": 1.63549
    },
    {
      "acc": 0.66132369,
      "epoch": 1.1439624556062913,
      "grad_norm": 5.46875,
      "learning_rate": 4.2263009571141585e-06,
      "loss": 1.5713686,
      "memory(GiB)": 117.38,
      "step": 45095,
      "train_speed(iter/s)": 1.63551
    },
    {
      "acc": 0.66328506,
      "epoch": 1.1440892947742263,
      "grad_norm": 5.21875,
      "learning_rate": 4.225264980859485e-06,
      "loss": 1.57846813,
      "memory(GiB)": 117.38,
      "step": 45100,
      "train_speed(iter/s)": 1.635531
    },
    {
      "acc": 0.64998884,
      "epoch": 1.1442161339421613,
      "grad_norm": 5.8125,
      "learning_rate": 4.224229038681115e-06,
      "loss": 1.66005554,
      "memory(GiB)": 117.38,
      "step": 45105,
      "train_speed(iter/s)": 1.635552
    },
    {
      "acc": 0.64826536,
      "epoch": 1.1443429731100965,
      "grad_norm": 5.3125,
      "learning_rate": 4.223193130624619e-06,
      "loss": 1.65065155,
      "memory(GiB)": 117.38,
      "step": 45110,
      "train_speed(iter/s)": 1.635571
    },
    {
      "acc": 0.67079296,
      "epoch": 1.1444698122780315,
      "grad_norm": 5.4375,
      "learning_rate": 4.222157256735553e-06,
      "loss": 1.50241985,
      "memory(GiB)": 117.38,
      "step": 45115,
      "train_speed(iter/s)": 1.635589
    },
    {
      "acc": 0.66567268,
      "epoch": 1.1445966514459665,
      "grad_norm": 6.1875,
      "learning_rate": 4.2211214170594865e-06,
      "loss": 1.59095192,
      "memory(GiB)": 117.38,
      "step": 45120,
      "train_speed(iter/s)": 1.63561
    },
    {
      "acc": 0.65066247,
      "epoch": 1.1447234906139017,
      "grad_norm": 8.1875,
      "learning_rate": 4.220085611641976e-06,
      "loss": 1.6233942,
      "memory(GiB)": 117.38,
      "step": 45125,
      "train_speed(iter/s)": 1.635629
    },
    {
      "acc": 0.66051474,
      "epoch": 1.1448503297818367,
      "grad_norm": 5.125,
      "learning_rate": 4.2190498405285826e-06,
      "loss": 1.64969177,
      "memory(GiB)": 117.38,
      "step": 45130,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.66293893,
      "epoch": 1.1449771689497716,
      "grad_norm": 5.75,
      "learning_rate": 4.218014103764865e-06,
      "loss": 1.57484598,
      "memory(GiB)": 117.38,
      "step": 45135,
      "train_speed(iter/s)": 1.635668
    },
    {
      "acc": 0.63383932,
      "epoch": 1.1451040081177069,
      "grad_norm": 5.53125,
      "learning_rate": 4.216978401396376e-06,
      "loss": 1.65815392,
      "memory(GiB)": 117.38,
      "step": 45140,
      "train_speed(iter/s)": 1.635687
    },
    {
      "acc": 0.64647589,
      "epoch": 1.1452308472856418,
      "grad_norm": 6.3125,
      "learning_rate": 4.215942733468675e-06,
      "loss": 1.63710175,
      "memory(GiB)": 117.38,
      "step": 45145,
      "train_speed(iter/s)": 1.635709
    },
    {
      "acc": 0.67577429,
      "epoch": 1.1453576864535768,
      "grad_norm": 6.09375,
      "learning_rate": 4.2149071000273134e-06,
      "loss": 1.49446564,
      "memory(GiB)": 117.38,
      "step": 45150,
      "train_speed(iter/s)": 1.635729
    },
    {
      "acc": 0.65749898,
      "epoch": 1.1454845256215118,
      "grad_norm": 7.15625,
      "learning_rate": 4.213871501117842e-06,
      "loss": 1.54127226,
      "memory(GiB)": 117.38,
      "step": 45155,
      "train_speed(iter/s)": 1.635749
    },
    {
      "acc": 0.64806232,
      "epoch": 1.145611364789447,
      "grad_norm": 5.625,
      "learning_rate": 4.212835936785811e-06,
      "loss": 1.69376068,
      "memory(GiB)": 117.38,
      "step": 45160,
      "train_speed(iter/s)": 1.635769
    },
    {
      "acc": 0.671523,
      "epoch": 1.145738203957382,
      "grad_norm": 6.03125,
      "learning_rate": 4.21180040707677e-06,
      "loss": 1.51248293,
      "memory(GiB)": 117.38,
      "step": 45165,
      "train_speed(iter/s)": 1.63579
    },
    {
      "acc": 0.6596693,
      "epoch": 1.145865043125317,
      "grad_norm": 5.4375,
      "learning_rate": 4.2107649120362684e-06,
      "loss": 1.64692001,
      "memory(GiB)": 117.38,
      "step": 45170,
      "train_speed(iter/s)": 1.63581
    },
    {
      "acc": 0.64200983,
      "epoch": 1.1459918822932522,
      "grad_norm": 4.71875,
      "learning_rate": 4.2097294517098465e-06,
      "loss": 1.65302429,
      "memory(GiB)": 117.38,
      "step": 45175,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.65853386,
      "epoch": 1.1461187214611872,
      "grad_norm": 7.09375,
      "learning_rate": 4.208694026143054e-06,
      "loss": 1.66464615,
      "memory(GiB)": 117.38,
      "step": 45180,
      "train_speed(iter/s)": 1.635843
    },
    {
      "acc": 0.66068316,
      "epoch": 1.1462455606291222,
      "grad_norm": 6.46875,
      "learning_rate": 4.2076586353814295e-06,
      "loss": 1.68183842,
      "memory(GiB)": 117.38,
      "step": 45185,
      "train_speed(iter/s)": 1.635862
    },
    {
      "acc": 0.66107907,
      "epoch": 1.1463723997970574,
      "grad_norm": 5.25,
      "learning_rate": 4.2066232794705174e-06,
      "loss": 1.58206558,
      "memory(GiB)": 117.38,
      "step": 45190,
      "train_speed(iter/s)": 1.635881
    },
    {
      "acc": 0.64624529,
      "epoch": 1.1464992389649924,
      "grad_norm": 5.65625,
      "learning_rate": 4.205587958455854e-06,
      "loss": 1.60086403,
      "memory(GiB)": 117.38,
      "step": 45195,
      "train_speed(iter/s)": 1.635902
    },
    {
      "acc": 0.65240498,
      "epoch": 1.1466260781329274,
      "grad_norm": 5.34375,
      "learning_rate": 4.204552672382981e-06,
      "loss": 1.65291519,
      "memory(GiB)": 117.38,
      "step": 45200,
      "train_speed(iter/s)": 1.635922
    },
    {
      "acc": 0.65476499,
      "epoch": 1.1467529173008626,
      "grad_norm": 5.03125,
      "learning_rate": 4.203517421297431e-06,
      "loss": 1.5894145,
      "memory(GiB)": 117.38,
      "step": 45205,
      "train_speed(iter/s)": 1.635943
    },
    {
      "acc": 0.64782419,
      "epoch": 1.1468797564687976,
      "grad_norm": 5.40625,
      "learning_rate": 4.202482205244742e-06,
      "loss": 1.57690372,
      "memory(GiB)": 117.38,
      "step": 45210,
      "train_speed(iter/s)": 1.635963
    },
    {
      "acc": 0.67050691,
      "epoch": 1.1470065956367326,
      "grad_norm": 6.78125,
      "learning_rate": 4.201447024270446e-06,
      "loss": 1.55274944,
      "memory(GiB)": 117.38,
      "step": 45215,
      "train_speed(iter/s)": 1.635984
    },
    {
      "acc": 0.65082684,
      "epoch": 1.1471334348046676,
      "grad_norm": 5.375,
      "learning_rate": 4.200411878420074e-06,
      "loss": 1.65268574,
      "memory(GiB)": 117.38,
      "step": 45220,
      "train_speed(iter/s)": 1.636004
    },
    {
      "acc": 0.64993443,
      "epoch": 1.1472602739726028,
      "grad_norm": 6.875,
      "learning_rate": 4.199376767739158e-06,
      "loss": 1.58830147,
      "memory(GiB)": 117.38,
      "step": 45225,
      "train_speed(iter/s)": 1.636024
    },
    {
      "acc": 0.67012959,
      "epoch": 1.1473871131405378,
      "grad_norm": 6.3125,
      "learning_rate": 4.1983416922732276e-06,
      "loss": 1.56773262,
      "memory(GiB)": 117.38,
      "step": 45230,
      "train_speed(iter/s)": 1.636043
    },
    {
      "acc": 0.65392261,
      "epoch": 1.147513952308473,
      "grad_norm": 5.25,
      "learning_rate": 4.197306652067807e-06,
      "loss": 1.54658403,
      "memory(GiB)": 117.38,
      "step": 45235,
      "train_speed(iter/s)": 1.636063
    },
    {
      "acc": 0.66988049,
      "epoch": 1.147640791476408,
      "grad_norm": 6.25,
      "learning_rate": 4.196271647168425e-06,
      "loss": 1.52350998,
      "memory(GiB)": 117.38,
      "step": 45240,
      "train_speed(iter/s)": 1.636083
    },
    {
      "acc": 0.63933067,
      "epoch": 1.147767630644343,
      "grad_norm": 5.71875,
      "learning_rate": 4.195236677620604e-06,
      "loss": 1.66917629,
      "memory(GiB)": 117.38,
      "step": 45245,
      "train_speed(iter/s)": 1.636104
    },
    {
      "acc": 0.67018023,
      "epoch": 1.147894469812278,
      "grad_norm": 6.90625,
      "learning_rate": 4.1942017434698675e-06,
      "loss": 1.46641054,
      "memory(GiB)": 117.38,
      "step": 45250,
      "train_speed(iter/s)": 1.636124
    },
    {
      "acc": 0.65379639,
      "epoch": 1.1480213089802132,
      "grad_norm": 4.96875,
      "learning_rate": 4.1931668447617346e-06,
      "loss": 1.57654285,
      "memory(GiB)": 117.38,
      "step": 45255,
      "train_speed(iter/s)": 1.636143
    },
    {
      "acc": 0.66897969,
      "epoch": 1.1481481481481481,
      "grad_norm": 6.0,
      "learning_rate": 4.192131981541727e-06,
      "loss": 1.64674473,
      "memory(GiB)": 117.38,
      "step": 45260,
      "train_speed(iter/s)": 1.636163
    },
    {
      "acc": 0.65962706,
      "epoch": 1.1482749873160831,
      "grad_norm": 6.84375,
      "learning_rate": 4.19109715385536e-06,
      "loss": 1.58185654,
      "memory(GiB)": 117.38,
      "step": 45265,
      "train_speed(iter/s)": 1.636183
    },
    {
      "acc": 0.63951559,
      "epoch": 1.1484018264840183,
      "grad_norm": 5.25,
      "learning_rate": 4.190062361748154e-06,
      "loss": 1.61705322,
      "memory(GiB)": 117.38,
      "step": 45270,
      "train_speed(iter/s)": 1.636203
    },
    {
      "acc": 0.65071244,
      "epoch": 1.1485286656519533,
      "grad_norm": 7.34375,
      "learning_rate": 4.189027605265621e-06,
      "loss": 1.6297657,
      "memory(GiB)": 117.38,
      "step": 45275,
      "train_speed(iter/s)": 1.636223
    },
    {
      "acc": 0.65401368,
      "epoch": 1.1486555048198883,
      "grad_norm": 6.84375,
      "learning_rate": 4.187992884453273e-06,
      "loss": 1.61136131,
      "memory(GiB)": 117.38,
      "step": 45280,
      "train_speed(iter/s)": 1.636243
    },
    {
      "acc": 0.65580549,
      "epoch": 1.1487823439878235,
      "grad_norm": 5.125,
      "learning_rate": 4.186958199356624e-06,
      "loss": 1.61955166,
      "memory(GiB)": 117.38,
      "step": 45285,
      "train_speed(iter/s)": 1.636264
    },
    {
      "acc": 0.64785986,
      "epoch": 1.1489091831557585,
      "grad_norm": 7.46875,
      "learning_rate": 4.185923550021185e-06,
      "loss": 1.63833656,
      "memory(GiB)": 117.38,
      "step": 45290,
      "train_speed(iter/s)": 1.636284
    },
    {
      "acc": 0.65836873,
      "epoch": 1.1490360223236935,
      "grad_norm": 4.75,
      "learning_rate": 4.1848889364924625e-06,
      "loss": 1.58048573,
      "memory(GiB)": 117.38,
      "step": 45295,
      "train_speed(iter/s)": 1.636303
    },
    {
      "acc": 0.65235901,
      "epoch": 1.1491628614916287,
      "grad_norm": 6.34375,
      "learning_rate": 4.183854358815962e-06,
      "loss": 1.65781555,
      "memory(GiB)": 117.38,
      "step": 45300,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.6562737,
      "epoch": 1.1492897006595637,
      "grad_norm": 6.59375,
      "learning_rate": 4.182819817037192e-06,
      "loss": 1.58850946,
      "memory(GiB)": 117.38,
      "step": 45305,
      "train_speed(iter/s)": 1.636343
    },
    {
      "acc": 0.65125065,
      "epoch": 1.1494165398274987,
      "grad_norm": 5.71875,
      "learning_rate": 4.181785311201655e-06,
      "loss": 1.62749519,
      "memory(GiB)": 117.38,
      "step": 45310,
      "train_speed(iter/s)": 1.636364
    },
    {
      "acc": 0.65248361,
      "epoch": 1.1495433789954337,
      "grad_norm": 5.4375,
      "learning_rate": 4.1807508413548515e-06,
      "loss": 1.64515991,
      "memory(GiB)": 117.38,
      "step": 45315,
      "train_speed(iter/s)": 1.636383
    },
    {
      "acc": 0.65627723,
      "epoch": 1.149670218163369,
      "grad_norm": 6.6875,
      "learning_rate": 4.179716407542285e-06,
      "loss": 1.56477146,
      "memory(GiB)": 117.38,
      "step": 45320,
      "train_speed(iter/s)": 1.636403
    },
    {
      "acc": 0.64873562,
      "epoch": 1.1497970573313039,
      "grad_norm": 5.625,
      "learning_rate": 4.178682009809452e-06,
      "loss": 1.64755135,
      "memory(GiB)": 117.38,
      "step": 45325,
      "train_speed(iter/s)": 1.636425
    },
    {
      "acc": 0.65606441,
      "epoch": 1.1499238964992389,
      "grad_norm": 4.75,
      "learning_rate": 4.177647648201854e-06,
      "loss": 1.59035797,
      "memory(GiB)": 117.38,
      "step": 45330,
      "train_speed(iter/s)": 1.636446
    },
    {
      "acc": 0.64957151,
      "epoch": 1.150050735667174,
      "grad_norm": 5.4375,
      "learning_rate": 4.1766133227649815e-06,
      "loss": 1.58410587,
      "memory(GiB)": 117.38,
      "step": 45335,
      "train_speed(iter/s)": 1.636466
    },
    {
      "acc": 0.66110668,
      "epoch": 1.150177574835109,
      "grad_norm": 5.53125,
      "learning_rate": 4.175579033544332e-06,
      "loss": 1.55576954,
      "memory(GiB)": 117.38,
      "step": 45340,
      "train_speed(iter/s)": 1.636487
    },
    {
      "acc": 0.64153652,
      "epoch": 1.150304414003044,
      "grad_norm": 5.78125,
      "learning_rate": 4.174544780585395e-06,
      "loss": 1.61314507,
      "memory(GiB)": 117.38,
      "step": 45345,
      "train_speed(iter/s)": 1.636508
    },
    {
      "acc": 0.64415364,
      "epoch": 1.1504312531709793,
      "grad_norm": 10.3125,
      "learning_rate": 4.1735105639336686e-06,
      "loss": 1.71888123,
      "memory(GiB)": 117.38,
      "step": 45350,
      "train_speed(iter/s)": 1.636528
    },
    {
      "acc": 0.6536171,
      "epoch": 1.1505580923389143,
      "grad_norm": 5.28125,
      "learning_rate": 4.172476383634635e-06,
      "loss": 1.64055214,
      "memory(GiB)": 117.38,
      "step": 45355,
      "train_speed(iter/s)": 1.636548
    },
    {
      "acc": 0.65930958,
      "epoch": 1.1506849315068493,
      "grad_norm": 6.03125,
      "learning_rate": 4.171442239733783e-06,
      "loss": 1.67506371,
      "memory(GiB)": 117.38,
      "step": 45360,
      "train_speed(iter/s)": 1.636567
    },
    {
      "acc": 0.64076943,
      "epoch": 1.1508117706747845,
      "grad_norm": 5.625,
      "learning_rate": 4.170408132276603e-06,
      "loss": 1.68986702,
      "memory(GiB)": 117.38,
      "step": 45365,
      "train_speed(iter/s)": 1.636586
    },
    {
      "acc": 0.64943233,
      "epoch": 1.1509386098427195,
      "grad_norm": 7.53125,
      "learning_rate": 4.1693740613085776e-06,
      "loss": 1.6353241,
      "memory(GiB)": 117.38,
      "step": 45370,
      "train_speed(iter/s)": 1.636606
    },
    {
      "acc": 0.67001419,
      "epoch": 1.1510654490106544,
      "grad_norm": 5.59375,
      "learning_rate": 4.168340026875188e-06,
      "loss": 1.55925875,
      "memory(GiB)": 117.38,
      "step": 45375,
      "train_speed(iter/s)": 1.636628
    },
    {
      "acc": 0.65950322,
      "epoch": 1.1511922881785894,
      "grad_norm": 4.9375,
      "learning_rate": 4.167306029021917e-06,
      "loss": 1.59508457,
      "memory(GiB)": 117.38,
      "step": 45380,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.65052991,
      "epoch": 1.1513191273465246,
      "grad_norm": 6.375,
      "learning_rate": 4.166272067794246e-06,
      "loss": 1.6451601,
      "memory(GiB)": 117.38,
      "step": 45385,
      "train_speed(iter/s)": 1.636669
    },
    {
      "acc": 0.64905534,
      "epoch": 1.1514459665144596,
      "grad_norm": 4.96875,
      "learning_rate": 4.165238143237651e-06,
      "loss": 1.64910812,
      "memory(GiB)": 117.38,
      "step": 45390,
      "train_speed(iter/s)": 1.63669
    },
    {
      "acc": 0.64131794,
      "epoch": 1.1515728056823948,
      "grad_norm": 5.125,
      "learning_rate": 4.164204255397608e-06,
      "loss": 1.61529942,
      "memory(GiB)": 117.38,
      "step": 45395,
      "train_speed(iter/s)": 1.63671
    },
    {
      "acc": 0.64460678,
      "epoch": 1.1516996448503298,
      "grad_norm": 4.75,
      "learning_rate": 4.163170404319596e-06,
      "loss": 1.61353378,
      "memory(GiB)": 117.38,
      "step": 45400,
      "train_speed(iter/s)": 1.636731
    },
    {
      "acc": 0.66338181,
      "epoch": 1.1518264840182648,
      "grad_norm": 5.375,
      "learning_rate": 4.1621365900490825e-06,
      "loss": 1.61284351,
      "memory(GiB)": 117.38,
      "step": 45405,
      "train_speed(iter/s)": 1.636752
    },
    {
      "acc": 0.66150312,
      "epoch": 1.1519533231861998,
      "grad_norm": 6.4375,
      "learning_rate": 4.1611028126315455e-06,
      "loss": 1.56076374,
      "memory(GiB)": 117.38,
      "step": 45410,
      "train_speed(iter/s)": 1.636772
    },
    {
      "acc": 0.6500288,
      "epoch": 1.152080162354135,
      "grad_norm": 5.5625,
      "learning_rate": 4.160069072112451e-06,
      "loss": 1.59573555,
      "memory(GiB)": 117.38,
      "step": 45415,
      "train_speed(iter/s)": 1.636792
    },
    {
      "acc": 0.66158094,
      "epoch": 1.15220700152207,
      "grad_norm": 4.78125,
      "learning_rate": 4.1590353685372695e-06,
      "loss": 1.53414373,
      "memory(GiB)": 117.38,
      "step": 45420,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.66660585,
      "epoch": 1.152333840690005,
      "grad_norm": 6.5,
      "learning_rate": 4.158001701951465e-06,
      "loss": 1.52418327,
      "memory(GiB)": 117.38,
      "step": 45425,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.64957085,
      "epoch": 1.1524606798579402,
      "grad_norm": 5.9375,
      "learning_rate": 4.156968072400508e-06,
      "loss": 1.62925873,
      "memory(GiB)": 117.38,
      "step": 45430,
      "train_speed(iter/s)": 1.636853
    },
    {
      "acc": 0.65010386,
      "epoch": 1.1525875190258752,
      "grad_norm": 6.40625,
      "learning_rate": 4.155934479929858e-06,
      "loss": 1.62804775,
      "memory(GiB)": 117.38,
      "step": 45435,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.68676844,
      "epoch": 1.1527143581938102,
      "grad_norm": 9.25,
      "learning_rate": 4.154900924584976e-06,
      "loss": 1.46937981,
      "memory(GiB)": 117.38,
      "step": 45440,
      "train_speed(iter/s)": 1.636893
    },
    {
      "acc": 0.64122305,
      "epoch": 1.1528411973617454,
      "grad_norm": 5.3125,
      "learning_rate": 4.153867406411327e-06,
      "loss": 1.68359184,
      "memory(GiB)": 117.38,
      "step": 45445,
      "train_speed(iter/s)": 1.636914
    },
    {
      "acc": 0.66693935,
      "epoch": 1.1529680365296804,
      "grad_norm": 6.53125,
      "learning_rate": 4.152833925454367e-06,
      "loss": 1.53952808,
      "memory(GiB)": 117.38,
      "step": 45450,
      "train_speed(iter/s)": 1.636934
    },
    {
      "acc": 0.66180687,
      "epoch": 1.1530948756976154,
      "grad_norm": 5.21875,
      "learning_rate": 4.1518004817595515e-06,
      "loss": 1.54953289,
      "memory(GiB)": 117.38,
      "step": 45455,
      "train_speed(iter/s)": 1.636952
    },
    {
      "acc": 0.63008494,
      "epoch": 1.1532217148655506,
      "grad_norm": 7.21875,
      "learning_rate": 4.150767075372338e-06,
      "loss": 1.70316315,
      "memory(GiB)": 117.38,
      "step": 45460,
      "train_speed(iter/s)": 1.636973
    },
    {
      "acc": 0.65845613,
      "epoch": 1.1533485540334856,
      "grad_norm": 6.125,
      "learning_rate": 4.149733706338182e-06,
      "loss": 1.55905695,
      "memory(GiB)": 117.38,
      "step": 45465,
      "train_speed(iter/s)": 1.636993
    },
    {
      "acc": 0.65770588,
      "epoch": 1.1534753932014206,
      "grad_norm": 5.875,
      "learning_rate": 4.148700374702533e-06,
      "loss": 1.58927193,
      "memory(GiB)": 117.38,
      "step": 45470,
      "train_speed(iter/s)": 1.637011
    },
    {
      "acc": 0.66058807,
      "epoch": 1.1536022323693556,
      "grad_norm": 5.0,
      "learning_rate": 4.147667080510841e-06,
      "loss": 1.58783236,
      "memory(GiB)": 117.38,
      "step": 45475,
      "train_speed(iter/s)": 1.637031
    },
    {
      "acc": 0.6663022,
      "epoch": 1.1537290715372908,
      "grad_norm": 5.25,
      "learning_rate": 4.146633823808557e-06,
      "loss": 1.54864883,
      "memory(GiB)": 117.38,
      "step": 45480,
      "train_speed(iter/s)": 1.637051
    },
    {
      "acc": 0.64306774,
      "epoch": 1.1538559107052258,
      "grad_norm": 5.875,
      "learning_rate": 4.145600604641127e-06,
      "loss": 1.62109756,
      "memory(GiB)": 117.38,
      "step": 45485,
      "train_speed(iter/s)": 1.637071
    },
    {
      "acc": 0.66459589,
      "epoch": 1.1539827498731607,
      "grad_norm": 6.375,
      "learning_rate": 4.1445674230539985e-06,
      "loss": 1.58660212,
      "memory(GiB)": 117.38,
      "step": 45490,
      "train_speed(iter/s)": 1.63709
    },
    {
      "acc": 0.6673192,
      "epoch": 1.154109589041096,
      "grad_norm": 6.25,
      "learning_rate": 4.143534279092613e-06,
      "loss": 1.58046036,
      "memory(GiB)": 117.38,
      "step": 45495,
      "train_speed(iter/s)": 1.63711
    },
    {
      "acc": 0.65245991,
      "epoch": 1.154236428209031,
      "grad_norm": 6.09375,
      "learning_rate": 4.142501172802412e-06,
      "loss": 1.68155785,
      "memory(GiB)": 117.38,
      "step": 45500,
      "train_speed(iter/s)": 1.637129
    },
    {
      "acc": 0.64704008,
      "epoch": 1.154363267376966,
      "grad_norm": 5.9375,
      "learning_rate": 4.14146810422884e-06,
      "loss": 1.65030556,
      "memory(GiB)": 117.38,
      "step": 45505,
      "train_speed(iter/s)": 1.637148
    },
    {
      "acc": 0.65180693,
      "epoch": 1.1544901065449011,
      "grad_norm": 7.0,
      "learning_rate": 4.140435073417335e-06,
      "loss": 1.57812862,
      "memory(GiB)": 117.38,
      "step": 45510,
      "train_speed(iter/s)": 1.637168
    },
    {
      "acc": 0.65421929,
      "epoch": 1.1546169457128361,
      "grad_norm": 5.125,
      "learning_rate": 4.139402080413331e-06,
      "loss": 1.63333874,
      "memory(GiB)": 117.38,
      "step": 45515,
      "train_speed(iter/s)": 1.637187
    },
    {
      "acc": 0.66482153,
      "epoch": 1.1547437848807711,
      "grad_norm": 5.5,
      "learning_rate": 4.138369125262266e-06,
      "loss": 1.60764771,
      "memory(GiB)": 117.38,
      "step": 45520,
      "train_speed(iter/s)": 1.637208
    },
    {
      "acc": 0.65694504,
      "epoch": 1.1548706240487063,
      "grad_norm": 5.53125,
      "learning_rate": 4.137336208009574e-06,
      "loss": 1.59771414,
      "memory(GiB)": 117.38,
      "step": 45525,
      "train_speed(iter/s)": 1.637227
    },
    {
      "acc": 0.64919019,
      "epoch": 1.1549974632166413,
      "grad_norm": 5.59375,
      "learning_rate": 4.136303328700688e-06,
      "loss": 1.64983845,
      "memory(GiB)": 117.38,
      "step": 45530,
      "train_speed(iter/s)": 1.637248
    },
    {
      "acc": 0.63951387,
      "epoch": 1.1551243023845763,
      "grad_norm": 5.78125,
      "learning_rate": 4.135270487381037e-06,
      "loss": 1.66697044,
      "memory(GiB)": 117.38,
      "step": 45535,
      "train_speed(iter/s)": 1.637269
    },
    {
      "acc": 0.66064291,
      "epoch": 1.1552511415525113,
      "grad_norm": 5.15625,
      "learning_rate": 4.13423768409605e-06,
      "loss": 1.56094532,
      "memory(GiB)": 117.38,
      "step": 45540,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.6514194,
      "epoch": 1.1553779807204465,
      "grad_norm": 5.21875,
      "learning_rate": 4.133204918891155e-06,
      "loss": 1.63216496,
      "memory(GiB)": 117.38,
      "step": 45545,
      "train_speed(iter/s)": 1.63731
    },
    {
      "acc": 0.65630569,
      "epoch": 1.1555048198883815,
      "grad_norm": 6.21875,
      "learning_rate": 4.132172191811781e-06,
      "loss": 1.63021584,
      "memory(GiB)": 117.38,
      "step": 45550,
      "train_speed(iter/s)": 1.63733
    },
    {
      "acc": 0.65527201,
      "epoch": 1.1556316590563167,
      "grad_norm": 5.21875,
      "learning_rate": 4.131139502903345e-06,
      "loss": 1.61451988,
      "memory(GiB)": 117.38,
      "step": 45555,
      "train_speed(iter/s)": 1.63735
    },
    {
      "acc": 0.64712048,
      "epoch": 1.1557584982242517,
      "grad_norm": 5.125,
      "learning_rate": 4.130106852211273e-06,
      "loss": 1.64628601,
      "memory(GiB)": 117.38,
      "step": 45560,
      "train_speed(iter/s)": 1.63737
    },
    {
      "acc": 0.65945153,
      "epoch": 1.1558853373921867,
      "grad_norm": 5.6875,
      "learning_rate": 4.129074239780986e-06,
      "loss": 1.55537014,
      "memory(GiB)": 117.38,
      "step": 45565,
      "train_speed(iter/s)": 1.63739
    },
    {
      "acc": 0.67395611,
      "epoch": 1.1560121765601217,
      "grad_norm": 5.9375,
      "learning_rate": 4.128041665657903e-06,
      "loss": 1.58240671,
      "memory(GiB)": 117.38,
      "step": 45570,
      "train_speed(iter/s)": 1.63741
    },
    {
      "acc": 0.66048293,
      "epoch": 1.1561390157280569,
      "grad_norm": 5.5,
      "learning_rate": 4.127009129887441e-06,
      "loss": 1.58171101,
      "memory(GiB)": 117.38,
      "step": 45575,
      "train_speed(iter/s)": 1.637431
    },
    {
      "acc": 0.6499372,
      "epoch": 1.1562658548959919,
      "grad_norm": 5.65625,
      "learning_rate": 4.125976632515013e-06,
      "loss": 1.60724125,
      "memory(GiB)": 117.38,
      "step": 45580,
      "train_speed(iter/s)": 1.637452
    },
    {
      "acc": 0.65003896,
      "epoch": 1.1563926940639269,
      "grad_norm": 5.34375,
      "learning_rate": 4.124944173586036e-06,
      "loss": 1.62497673,
      "memory(GiB)": 117.38,
      "step": 45585,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.64358754,
      "epoch": 1.156519533231862,
      "grad_norm": 6.09375,
      "learning_rate": 4.123911753145922e-06,
      "loss": 1.64446964,
      "memory(GiB)": 117.38,
      "step": 45590,
      "train_speed(iter/s)": 1.637492
    },
    {
      "acc": 0.66175175,
      "epoch": 1.156646372399797,
      "grad_norm": 7.25,
      "learning_rate": 4.12287937124008e-06,
      "loss": 1.61964569,
      "memory(GiB)": 117.38,
      "step": 45595,
      "train_speed(iter/s)": 1.637513
    },
    {
      "acc": 0.65238795,
      "epoch": 1.156773211567732,
      "grad_norm": 6.78125,
      "learning_rate": 4.121847027913918e-06,
      "loss": 1.67598457,
      "memory(GiB)": 117.38,
      "step": 45600,
      "train_speed(iter/s)": 1.637535
    },
    {
      "acc": 0.66461244,
      "epoch": 1.1569000507356673,
      "grad_norm": 6.09375,
      "learning_rate": 4.1208147232128456e-06,
      "loss": 1.58861532,
      "memory(GiB)": 117.38,
      "step": 45605,
      "train_speed(iter/s)": 1.637555
    },
    {
      "acc": 0.64696674,
      "epoch": 1.1570268899036023,
      "grad_norm": 5.71875,
      "learning_rate": 4.119782457182267e-06,
      "loss": 1.65146809,
      "memory(GiB)": 117.38,
      "step": 45610,
      "train_speed(iter/s)": 1.637575
    },
    {
      "acc": 0.6603157,
      "epoch": 1.1571537290715372,
      "grad_norm": 5.625,
      "learning_rate": 4.118750229867585e-06,
      "loss": 1.59788628,
      "memory(GiB)": 117.38,
      "step": 45615,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.64708014,
      "epoch": 1.1572805682394725,
      "grad_norm": 6.125,
      "learning_rate": 4.117718041314204e-06,
      "loss": 1.63051414,
      "memory(GiB)": 117.38,
      "step": 45620,
      "train_speed(iter/s)": 1.637617
    },
    {
      "acc": 0.65227494,
      "epoch": 1.1574074074074074,
      "grad_norm": 5.28125,
      "learning_rate": 4.11668589156752e-06,
      "loss": 1.6279808,
      "memory(GiB)": 117.38,
      "step": 45625,
      "train_speed(iter/s)": 1.637638
    },
    {
      "acc": 0.67124414,
      "epoch": 1.1575342465753424,
      "grad_norm": 5.6875,
      "learning_rate": 4.115653780672937e-06,
      "loss": 1.53081388,
      "memory(GiB)": 117.38,
      "step": 45630,
      "train_speed(iter/s)": 1.637658
    },
    {
      "acc": 0.65506029,
      "epoch": 1.1576610857432774,
      "grad_norm": 6.09375,
      "learning_rate": 4.1146217086758475e-06,
      "loss": 1.60845337,
      "memory(GiB)": 117.38,
      "step": 45635,
      "train_speed(iter/s)": 1.637678
    },
    {
      "acc": 0.67593899,
      "epoch": 1.1577879249112126,
      "grad_norm": 6.03125,
      "learning_rate": 4.113589675621649e-06,
      "loss": 1.50124655,
      "memory(GiB)": 117.38,
      "step": 45640,
      "train_speed(iter/s)": 1.637698
    },
    {
      "acc": 0.66740408,
      "epoch": 1.1579147640791476,
      "grad_norm": 4.8125,
      "learning_rate": 4.112557681555733e-06,
      "loss": 1.5421032,
      "memory(GiB)": 117.38,
      "step": 45645,
      "train_speed(iter/s)": 1.637717
    },
    {
      "acc": 0.65874996,
      "epoch": 1.1580416032470826,
      "grad_norm": 5.75,
      "learning_rate": 4.111525726523494e-06,
      "loss": 1.6162529,
      "memory(GiB)": 117.38,
      "step": 45650,
      "train_speed(iter/s)": 1.637738
    },
    {
      "acc": 0.66456437,
      "epoch": 1.1581684424150178,
      "grad_norm": 4.71875,
      "learning_rate": 4.110493810570319e-06,
      "loss": 1.50876808,
      "memory(GiB)": 117.38,
      "step": 45655,
      "train_speed(iter/s)": 1.637758
    },
    {
      "acc": 0.6542655,
      "epoch": 1.1582952815829528,
      "grad_norm": 6.59375,
      "learning_rate": 4.109461933741598e-06,
      "loss": 1.63389492,
      "memory(GiB)": 117.38,
      "step": 45660,
      "train_speed(iter/s)": 1.637778
    },
    {
      "acc": 0.64341602,
      "epoch": 1.1584221207508878,
      "grad_norm": 7.0625,
      "learning_rate": 4.108430096082716e-06,
      "loss": 1.66366501,
      "memory(GiB)": 117.38,
      "step": 45665,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.65244751,
      "epoch": 1.158548959918823,
      "grad_norm": 5.3125,
      "learning_rate": 4.107398297639062e-06,
      "loss": 1.63552132,
      "memory(GiB)": 117.38,
      "step": 45670,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.64832549,
      "epoch": 1.158675799086758,
      "grad_norm": 5.875,
      "learning_rate": 4.106366538456013e-06,
      "loss": 1.64582176,
      "memory(GiB)": 117.38,
      "step": 45675,
      "train_speed(iter/s)": 1.63784
    },
    {
      "acc": 0.67131143,
      "epoch": 1.158802638254693,
      "grad_norm": 5.375,
      "learning_rate": 4.105334818578954e-06,
      "loss": 1.53551464,
      "memory(GiB)": 117.38,
      "step": 45680,
      "train_speed(iter/s)": 1.637861
    },
    {
      "acc": 0.65744576,
      "epoch": 1.1589294774226282,
      "grad_norm": 5.4375,
      "learning_rate": 4.104303138053265e-06,
      "loss": 1.63645992,
      "memory(GiB)": 117.38,
      "step": 45685,
      "train_speed(iter/s)": 1.637882
    },
    {
      "acc": 0.67203264,
      "epoch": 1.1590563165905632,
      "grad_norm": 7.0,
      "learning_rate": 4.103271496924323e-06,
      "loss": 1.55208626,
      "memory(GiB)": 117.38,
      "step": 45690,
      "train_speed(iter/s)": 1.637902
    },
    {
      "acc": 0.67174592,
      "epoch": 1.1591831557584982,
      "grad_norm": 8.25,
      "learning_rate": 4.102239895237503e-06,
      "loss": 1.63320484,
      "memory(GiB)": 117.38,
      "step": 45695,
      "train_speed(iter/s)": 1.637923
    },
    {
      "acc": 0.67263622,
      "epoch": 1.1593099949264332,
      "grad_norm": 5.59375,
      "learning_rate": 4.101208333038181e-06,
      "loss": 1.54915209,
      "memory(GiB)": 117.38,
      "step": 45700,
      "train_speed(iter/s)": 1.637943
    },
    {
      "acc": 0.65530787,
      "epoch": 1.1594368340943684,
      "grad_norm": 6.78125,
      "learning_rate": 4.1001768103717285e-06,
      "loss": 1.62030373,
      "memory(GiB)": 117.38,
      "step": 45705,
      "train_speed(iter/s)": 1.637962
    },
    {
      "acc": 0.64375172,
      "epoch": 1.1595636732623034,
      "grad_norm": 6.21875,
      "learning_rate": 4.09914532728352e-06,
      "loss": 1.54258194,
      "memory(GiB)": 117.38,
      "step": 45710,
      "train_speed(iter/s)": 1.637982
    },
    {
      "acc": 0.64644041,
      "epoch": 1.1596905124302386,
      "grad_norm": 6.125,
      "learning_rate": 4.09811388381892e-06,
      "loss": 1.65020676,
      "memory(GiB)": 117.38,
      "step": 45715,
      "train_speed(iter/s)": 1.638001
    },
    {
      "acc": 0.66290736,
      "epoch": 1.1598173515981736,
      "grad_norm": 6.78125,
      "learning_rate": 4.097082480023298e-06,
      "loss": 1.62647362,
      "memory(GiB)": 117.38,
      "step": 45720,
      "train_speed(iter/s)": 1.638021
    },
    {
      "acc": 0.63748798,
      "epoch": 1.1599441907661086,
      "grad_norm": 5.46875,
      "learning_rate": 4.09605111594202e-06,
      "loss": 1.67567177,
      "memory(GiB)": 117.38,
      "step": 45725,
      "train_speed(iter/s)": 1.638041
    },
    {
      "acc": 0.65290003,
      "epoch": 1.1600710299340435,
      "grad_norm": 5.84375,
      "learning_rate": 4.095019791620451e-06,
      "loss": 1.61052761,
      "memory(GiB)": 117.38,
      "step": 45730,
      "train_speed(iter/s)": 1.638061
    },
    {
      "acc": 0.65383158,
      "epoch": 1.1601978691019788,
      "grad_norm": 5.1875,
      "learning_rate": 4.093988507103951e-06,
      "loss": 1.56698151,
      "memory(GiB)": 117.38,
      "step": 45735,
      "train_speed(iter/s)": 1.63808
    },
    {
      "acc": 0.66335864,
      "epoch": 1.1603247082699137,
      "grad_norm": 6.21875,
      "learning_rate": 4.09295726243788e-06,
      "loss": 1.54548235,
      "memory(GiB)": 117.38,
      "step": 45740,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.67101846,
      "epoch": 1.1604515474378487,
      "grad_norm": 7.34375,
      "learning_rate": 4.091926057667601e-06,
      "loss": 1.60606804,
      "memory(GiB)": 117.38,
      "step": 45745,
      "train_speed(iter/s)": 1.638121
    },
    {
      "acc": 0.64621401,
      "epoch": 1.160578386605784,
      "grad_norm": 6.34375,
      "learning_rate": 4.0908948928384675e-06,
      "loss": 1.5893981,
      "memory(GiB)": 117.38,
      "step": 45750,
      "train_speed(iter/s)": 1.63814
    },
    {
      "acc": 0.6651525,
      "epoch": 1.160705225773719,
      "grad_norm": 6.75,
      "learning_rate": 4.089863767995835e-06,
      "loss": 1.60778904,
      "memory(GiB)": 117.38,
      "step": 45755,
      "train_speed(iter/s)": 1.638162
    },
    {
      "acc": 0.64678073,
      "epoch": 1.160832064941654,
      "grad_norm": 5.8125,
      "learning_rate": 4.088832683185057e-06,
      "loss": 1.62014313,
      "memory(GiB)": 117.38,
      "step": 45760,
      "train_speed(iter/s)": 1.638181
    },
    {
      "acc": 0.63979187,
      "epoch": 1.1609589041095891,
      "grad_norm": 6.28125,
      "learning_rate": 4.087801638451485e-06,
      "loss": 1.65764751,
      "memory(GiB)": 117.38,
      "step": 45765,
      "train_speed(iter/s)": 1.6382
    },
    {
      "acc": 0.65871515,
      "epoch": 1.1610857432775241,
      "grad_norm": 5.15625,
      "learning_rate": 4.086770633840472e-06,
      "loss": 1.60395947,
      "memory(GiB)": 117.38,
      "step": 45770,
      "train_speed(iter/s)": 1.638218
    },
    {
      "acc": 0.65753632,
      "epoch": 1.161212582445459,
      "grad_norm": 6.53125,
      "learning_rate": 4.085739669397362e-06,
      "loss": 1.57268314,
      "memory(GiB)": 117.38,
      "step": 45775,
      "train_speed(iter/s)": 1.638239
    },
    {
      "acc": 0.6571804,
      "epoch": 1.1613394216133943,
      "grad_norm": 4.5625,
      "learning_rate": 4.084708745167504e-06,
      "loss": 1.54842768,
      "memory(GiB)": 117.38,
      "step": 45780,
      "train_speed(iter/s)": 1.63826
    },
    {
      "acc": 0.64140844,
      "epoch": 1.1614662607813293,
      "grad_norm": 5.71875,
      "learning_rate": 4.08367786119624e-06,
      "loss": 1.71731091,
      "memory(GiB)": 117.38,
      "step": 45785,
      "train_speed(iter/s)": 1.638279
    },
    {
      "acc": 0.66937094,
      "epoch": 1.1615930999492643,
      "grad_norm": 6.4375,
      "learning_rate": 4.082647017528918e-06,
      "loss": 1.58800392,
      "memory(GiB)": 117.38,
      "step": 45790,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.65077233,
      "epoch": 1.1617199391171993,
      "grad_norm": 6.96875,
      "learning_rate": 4.081616214210874e-06,
      "loss": 1.62099571,
      "memory(GiB)": 117.38,
      "step": 45795,
      "train_speed(iter/s)": 1.638319
    },
    {
      "acc": 0.66162701,
      "epoch": 1.1618467782851345,
      "grad_norm": 5.0,
      "learning_rate": 4.0805854512874485e-06,
      "loss": 1.59572086,
      "memory(GiB)": 117.38,
      "step": 45800,
      "train_speed(iter/s)": 1.638338
    },
    {
      "acc": 0.66195498,
      "epoch": 1.1619736174530695,
      "grad_norm": 5.15625,
      "learning_rate": 4.079554728803981e-06,
      "loss": 1.57355719,
      "memory(GiB)": 117.38,
      "step": 45805,
      "train_speed(iter/s)": 1.638357
    },
    {
      "acc": 0.66594782,
      "epoch": 1.1621004566210045,
      "grad_norm": 6.9375,
      "learning_rate": 4.078524046805806e-06,
      "loss": 1.59419041,
      "memory(GiB)": 117.38,
      "step": 45810,
      "train_speed(iter/s)": 1.638377
    },
    {
      "acc": 0.6529388,
      "epoch": 1.1622272957889397,
      "grad_norm": 6.53125,
      "learning_rate": 4.0774934053382576e-06,
      "loss": 1.5821641,
      "memory(GiB)": 117.38,
      "step": 45815,
      "train_speed(iter/s)": 1.638397
    },
    {
      "acc": 0.67012205,
      "epoch": 1.1623541349568747,
      "grad_norm": 5.875,
      "learning_rate": 4.076462804446667e-06,
      "loss": 1.59089537,
      "memory(GiB)": 117.38,
      "step": 45820,
      "train_speed(iter/s)": 1.638415
    },
    {
      "acc": 0.65776105,
      "epoch": 1.1624809741248097,
      "grad_norm": 6.96875,
      "learning_rate": 4.0754322441763654e-06,
      "loss": 1.63938808,
      "memory(GiB)": 117.38,
      "step": 45825,
      "train_speed(iter/s)": 1.638435
    },
    {
      "acc": 0.64684472,
      "epoch": 1.1626078132927449,
      "grad_norm": 5.0625,
      "learning_rate": 4.0744017245726834e-06,
      "loss": 1.64120007,
      "memory(GiB)": 117.38,
      "step": 45830,
      "train_speed(iter/s)": 1.638454
    },
    {
      "acc": 0.6506197,
      "epoch": 1.1627346524606799,
      "grad_norm": 5.8125,
      "learning_rate": 4.073371245680944e-06,
      "loss": 1.63957024,
      "memory(GiB)": 117.38,
      "step": 45835,
      "train_speed(iter/s)": 1.638472
    },
    {
      "acc": 0.64056778,
      "epoch": 1.1628614916286149,
      "grad_norm": 5.34375,
      "learning_rate": 4.0723408075464754e-06,
      "loss": 1.69090843,
      "memory(GiB)": 117.38,
      "step": 45840,
      "train_speed(iter/s)": 1.638493
    },
    {
      "acc": 0.64457269,
      "epoch": 1.16298833079655,
      "grad_norm": 5.75,
      "learning_rate": 4.071310410214598e-06,
      "loss": 1.65721245,
      "memory(GiB)": 117.38,
      "step": 45845,
      "train_speed(iter/s)": 1.638511
    },
    {
      "acc": 0.65111094,
      "epoch": 1.163115169964485,
      "grad_norm": 4.8125,
      "learning_rate": 4.070280053730639e-06,
      "loss": 1.69260597,
      "memory(GiB)": 117.38,
      "step": 45850,
      "train_speed(iter/s)": 1.638531
    },
    {
      "acc": 0.65409937,
      "epoch": 1.16324200913242,
      "grad_norm": 5.8125,
      "learning_rate": 4.069249738139911e-06,
      "loss": 1.62338486,
      "memory(GiB)": 117.38,
      "step": 45855,
      "train_speed(iter/s)": 1.638551
    },
    {
      "acc": 0.65706215,
      "epoch": 1.163368848300355,
      "grad_norm": 6.96875,
      "learning_rate": 4.068219463487736e-06,
      "loss": 1.60820999,
      "memory(GiB)": 117.38,
      "step": 45860,
      "train_speed(iter/s)": 1.638568
    },
    {
      "acc": 0.66592746,
      "epoch": 1.1634956874682902,
      "grad_norm": 5.5,
      "learning_rate": 4.0671892298194286e-06,
      "loss": 1.5888483,
      "memory(GiB)": 117.38,
      "step": 45865,
      "train_speed(iter/s)": 1.638587
    },
    {
      "acc": 0.66092119,
      "epoch": 1.1636225266362252,
      "grad_norm": 5.625,
      "learning_rate": 4.066159037180304e-06,
      "loss": 1.57736731,
      "memory(GiB)": 117.38,
      "step": 45870,
      "train_speed(iter/s)": 1.638605
    },
    {
      "acc": 0.65945692,
      "epoch": 1.1637493658041604,
      "grad_norm": 5.625,
      "learning_rate": 4.065128885615674e-06,
      "loss": 1.61729317,
      "memory(GiB)": 117.38,
      "step": 45875,
      "train_speed(iter/s)": 1.638624
    },
    {
      "acc": 0.62857614,
      "epoch": 1.1638762049720954,
      "grad_norm": 5.46875,
      "learning_rate": 4.064098775170849e-06,
      "loss": 1.64748878,
      "memory(GiB)": 117.38,
      "step": 45880,
      "train_speed(iter/s)": 1.638643
    },
    {
      "acc": 0.65355468,
      "epoch": 1.1640030441400304,
      "grad_norm": 5.84375,
      "learning_rate": 4.063068705891139e-06,
      "loss": 1.63424339,
      "memory(GiB)": 117.38,
      "step": 45885,
      "train_speed(iter/s)": 1.638663
    },
    {
      "acc": 0.65396433,
      "epoch": 1.1641298833079654,
      "grad_norm": 4.90625,
      "learning_rate": 4.062038677821852e-06,
      "loss": 1.63054276,
      "memory(GiB)": 117.38,
      "step": 45890,
      "train_speed(iter/s)": 1.638681
    },
    {
      "acc": 0.66293154,
      "epoch": 1.1642567224759006,
      "grad_norm": 5.65625,
      "learning_rate": 4.061008691008289e-06,
      "loss": 1.59148769,
      "memory(GiB)": 117.38,
      "step": 45895,
      "train_speed(iter/s)": 1.6387
    },
    {
      "acc": 0.66535788,
      "epoch": 1.1643835616438356,
      "grad_norm": 5.40625,
      "learning_rate": 4.059978745495757e-06,
      "loss": 1.56888647,
      "memory(GiB)": 117.38,
      "step": 45900,
      "train_speed(iter/s)": 1.638719
    },
    {
      "acc": 0.66330585,
      "epoch": 1.1645104008117706,
      "grad_norm": 4.875,
      "learning_rate": 4.058948841329557e-06,
      "loss": 1.59199753,
      "memory(GiB)": 117.38,
      "step": 45905,
      "train_speed(iter/s)": 1.638739
    },
    {
      "acc": 0.64743738,
      "epoch": 1.1646372399797058,
      "grad_norm": 5.625,
      "learning_rate": 4.057918978554989e-06,
      "loss": 1.6908392,
      "memory(GiB)": 117.38,
      "step": 45910,
      "train_speed(iter/s)": 1.638758
    },
    {
      "acc": 0.66670647,
      "epoch": 1.1647640791476408,
      "grad_norm": 5.8125,
      "learning_rate": 4.056889157217348e-06,
      "loss": 1.55901222,
      "memory(GiB)": 117.38,
      "step": 45915,
      "train_speed(iter/s)": 1.638424
    },
    {
      "acc": 0.63864088,
      "epoch": 1.1648909183155758,
      "grad_norm": 6.125,
      "learning_rate": 4.0558593773619346e-06,
      "loss": 1.68672218,
      "memory(GiB)": 117.38,
      "step": 45920,
      "train_speed(iter/s)": 1.638443
    },
    {
      "acc": 0.65399733,
      "epoch": 1.165017757483511,
      "grad_norm": 6.03125,
      "learning_rate": 4.05482963903404e-06,
      "loss": 1.62069855,
      "memory(GiB)": 117.38,
      "step": 45925,
      "train_speed(iter/s)": 1.638463
    },
    {
      "acc": 0.64970732,
      "epoch": 1.165144596651446,
      "grad_norm": 5.1875,
      "learning_rate": 4.05379994227896e-06,
      "loss": 1.67670097,
      "memory(GiB)": 117.38,
      "step": 45930,
      "train_speed(iter/s)": 1.638482
    },
    {
      "acc": 0.63536215,
      "epoch": 1.165271435819381,
      "grad_norm": 7.125,
      "learning_rate": 4.052770287141981e-06,
      "loss": 1.65553684,
      "memory(GiB)": 117.38,
      "step": 45935,
      "train_speed(iter/s)": 1.6385
    },
    {
      "acc": 0.6613471,
      "epoch": 1.1653982749873162,
      "grad_norm": 6.34375,
      "learning_rate": 4.051740673668393e-06,
      "loss": 1.517132,
      "memory(GiB)": 117.38,
      "step": 45940,
      "train_speed(iter/s)": 1.638519
    },
    {
      "acc": 0.65200567,
      "epoch": 1.1655251141552512,
      "grad_norm": 6.1875,
      "learning_rate": 4.0507111019034855e-06,
      "loss": 1.6038599,
      "memory(GiB)": 117.38,
      "step": 45945,
      "train_speed(iter/s)": 1.638538
    },
    {
      "acc": 0.65460243,
      "epoch": 1.1656519533231862,
      "grad_norm": 6.09375,
      "learning_rate": 4.049681571892543e-06,
      "loss": 1.59833679,
      "memory(GiB)": 117.38,
      "step": 45950,
      "train_speed(iter/s)": 1.638557
    },
    {
      "acc": 0.6562912,
      "epoch": 1.1657787924911212,
      "grad_norm": 5.25,
      "learning_rate": 4.048652083680847e-06,
      "loss": 1.59626503,
      "memory(GiB)": 117.38,
      "step": 45955,
      "train_speed(iter/s)": 1.638576
    },
    {
      "acc": 0.64160509,
      "epoch": 1.1659056316590564,
      "grad_norm": 5.15625,
      "learning_rate": 4.047622637313678e-06,
      "loss": 1.72627773,
      "memory(GiB)": 117.38,
      "step": 45960,
      "train_speed(iter/s)": 1.638595
    },
    {
      "acc": 0.66135969,
      "epoch": 1.1660324708269914,
      "grad_norm": 5.25,
      "learning_rate": 4.046593232836319e-06,
      "loss": 1.54990978,
      "memory(GiB)": 117.38,
      "step": 45965,
      "train_speed(iter/s)": 1.638614
    },
    {
      "acc": 0.65821109,
      "epoch": 1.1661593099949263,
      "grad_norm": 5.6875,
      "learning_rate": 4.045563870294047e-06,
      "loss": 1.62085037,
      "memory(GiB)": 117.38,
      "step": 45970,
      "train_speed(iter/s)": 1.638636
    },
    {
      "acc": 0.67386074,
      "epoch": 1.1662861491628616,
      "grad_norm": 6.0625,
      "learning_rate": 4.044534549732135e-06,
      "loss": 1.5024826,
      "memory(GiB)": 117.38,
      "step": 45975,
      "train_speed(iter/s)": 1.638655
    },
    {
      "acc": 0.67889004,
      "epoch": 1.1664129883307965,
      "grad_norm": 5.59375,
      "learning_rate": 4.043505271195861e-06,
      "loss": 1.51033525,
      "memory(GiB)": 117.38,
      "step": 45980,
      "train_speed(iter/s)": 1.638675
    },
    {
      "acc": 0.65484538,
      "epoch": 1.1665398274987315,
      "grad_norm": 6.78125,
      "learning_rate": 4.042476034730494e-06,
      "loss": 1.66983604,
      "memory(GiB)": 117.38,
      "step": 45985,
      "train_speed(iter/s)": 1.638694
    },
    {
      "acc": 0.64533615,
      "epoch": 1.1666666666666667,
      "grad_norm": 5.9375,
      "learning_rate": 4.041446840381309e-06,
      "loss": 1.65745106,
      "memory(GiB)": 117.38,
      "step": 45990,
      "train_speed(iter/s)": 1.638715
    },
    {
      "acc": 0.66766167,
      "epoch": 1.1667935058346017,
      "grad_norm": 5.5625,
      "learning_rate": 4.040417688193569e-06,
      "loss": 1.57248611,
      "memory(GiB)": 117.38,
      "step": 45995,
      "train_speed(iter/s)": 1.638734
    },
    {
      "acc": 0.65643172,
      "epoch": 1.1669203450025367,
      "grad_norm": 5.5,
      "learning_rate": 4.039388578212545e-06,
      "loss": 1.56962347,
      "memory(GiB)": 117.38,
      "step": 46000,
      "train_speed(iter/s)": 1.638753
    },
    {
      "epoch": 1.1669203450025367,
      "eval_acc": 0.6461841426563671,
      "eval_loss": 1.5733494758605957,
      "eval_runtime": 58.3367,
      "eval_samples_per_second": 109.194,
      "eval_steps_per_second": 27.307,
      "step": 46000
    },
    {
      "acc": 0.65896254,
      "epoch": 1.167047184170472,
      "grad_norm": 6.125,
      "learning_rate": 4.0383595104834975e-06,
      "loss": 1.58292599,
      "memory(GiB)": 117.38,
      "step": 46005,
      "train_speed(iter/s)": 1.635124
    },
    {
      "acc": 0.65144,
      "epoch": 1.167174023338407,
      "grad_norm": 8.0625,
      "learning_rate": 4.037330485051695e-06,
      "loss": 1.62439899,
      "memory(GiB)": 117.38,
      "step": 46010,
      "train_speed(iter/s)": 1.635143
    },
    {
      "acc": 0.64499083,
      "epoch": 1.167300862506342,
      "grad_norm": 4.78125,
      "learning_rate": 4.0363015019623955e-06,
      "loss": 1.63023815,
      "memory(GiB)": 117.38,
      "step": 46015,
      "train_speed(iter/s)": 1.635163
    },
    {
      "acc": 0.66844106,
      "epoch": 1.167427701674277,
      "grad_norm": 5.0,
      "learning_rate": 4.0352725612608565e-06,
      "loss": 1.57476463,
      "memory(GiB)": 117.38,
      "step": 46020,
      "train_speed(iter/s)": 1.635182
    },
    {
      "acc": 0.65385628,
      "epoch": 1.167554540842212,
      "grad_norm": 6.53125,
      "learning_rate": 4.0342436629923385e-06,
      "loss": 1.64859009,
      "memory(GiB)": 117.38,
      "step": 46025,
      "train_speed(iter/s)": 1.635201
    },
    {
      "acc": 0.66201344,
      "epoch": 1.167681380010147,
      "grad_norm": 5.75,
      "learning_rate": 4.033214807202098e-06,
      "loss": 1.5683795,
      "memory(GiB)": 117.38,
      "step": 46030,
      "train_speed(iter/s)": 1.63522
    },
    {
      "acc": 0.66079459,
      "epoch": 1.1678082191780823,
      "grad_norm": 5.9375,
      "learning_rate": 4.032185993935385e-06,
      "loss": 1.59754753,
      "memory(GiB)": 117.38,
      "step": 46035,
      "train_speed(iter/s)": 1.63524
    },
    {
      "acc": 0.66224594,
      "epoch": 1.1679350583460173,
      "grad_norm": 6.1875,
      "learning_rate": 4.031157223237452e-06,
      "loss": 1.54156399,
      "memory(GiB)": 117.38,
      "step": 46040,
      "train_speed(iter/s)": 1.635259
    },
    {
      "acc": 0.6721055,
      "epoch": 1.1680618975139523,
      "grad_norm": 6.1875,
      "learning_rate": 4.0301284951535504e-06,
      "loss": 1.54112396,
      "memory(GiB)": 117.38,
      "step": 46045,
      "train_speed(iter/s)": 1.635278
    },
    {
      "acc": 0.6689188,
      "epoch": 1.1681887366818873,
      "grad_norm": 5.1875,
      "learning_rate": 4.029099809728929e-06,
      "loss": 1.60583134,
      "memory(GiB)": 117.38,
      "step": 46050,
      "train_speed(iter/s)": 1.635298
    },
    {
      "acc": 0.65159554,
      "epoch": 1.1683155758498225,
      "grad_norm": 5.84375,
      "learning_rate": 4.028071167008831e-06,
      "loss": 1.63261814,
      "memory(GiB)": 117.38,
      "step": 46055,
      "train_speed(iter/s)": 1.635317
    },
    {
      "acc": 0.66183271,
      "epoch": 1.1684424150177575,
      "grad_norm": 6.0625,
      "learning_rate": 4.027042567038503e-06,
      "loss": 1.55288048,
      "memory(GiB)": 117.38,
      "step": 46060,
      "train_speed(iter/s)": 1.635333
    },
    {
      "acc": 0.65917106,
      "epoch": 1.1685692541856925,
      "grad_norm": 4.8125,
      "learning_rate": 4.026014009863186e-06,
      "loss": 1.61998405,
      "memory(GiB)": 117.38,
      "step": 46065,
      "train_speed(iter/s)": 1.635353
    },
    {
      "acc": 0.65444598,
      "epoch": 1.1686960933536277,
      "grad_norm": 5.75,
      "learning_rate": 4.024985495528124e-06,
      "loss": 1.61353416,
      "memory(GiB)": 117.38,
      "step": 46070,
      "train_speed(iter/s)": 1.635371
    },
    {
      "acc": 0.65920663,
      "epoch": 1.1688229325215627,
      "grad_norm": 6.03125,
      "learning_rate": 4.023957024078552e-06,
      "loss": 1.60706062,
      "memory(GiB)": 117.38,
      "step": 46075,
      "train_speed(iter/s)": 1.63539
    },
    {
      "acc": 0.6608573,
      "epoch": 1.1689497716894977,
      "grad_norm": 6.625,
      "learning_rate": 4.022928595559707e-06,
      "loss": 1.56209316,
      "memory(GiB)": 117.38,
      "step": 46080,
      "train_speed(iter/s)": 1.635409
    },
    {
      "acc": 0.65752068,
      "epoch": 1.1690766108574329,
      "grad_norm": 5.5625,
      "learning_rate": 4.021900210016824e-06,
      "loss": 1.61006203,
      "memory(GiB)": 117.38,
      "step": 46085,
      "train_speed(iter/s)": 1.635427
    },
    {
      "acc": 0.66767907,
      "epoch": 1.1692034500253679,
      "grad_norm": 6.1875,
      "learning_rate": 4.020871867495139e-06,
      "loss": 1.54134426,
      "memory(GiB)": 117.38,
      "step": 46090,
      "train_speed(iter/s)": 1.635447
    },
    {
      "acc": 0.66648788,
      "epoch": 1.1693302891933028,
      "grad_norm": 5.90625,
      "learning_rate": 4.01984356803988e-06,
      "loss": 1.5776721,
      "memory(GiB)": 117.38,
      "step": 46095,
      "train_speed(iter/s)": 1.635466
    },
    {
      "acc": 0.66540952,
      "epoch": 1.169457128361238,
      "grad_norm": 5.03125,
      "learning_rate": 4.018815311696274e-06,
      "loss": 1.60082035,
      "memory(GiB)": 117.38,
      "step": 46100,
      "train_speed(iter/s)": 1.635485
    },
    {
      "acc": 0.6458425,
      "epoch": 1.169583967529173,
      "grad_norm": 6.4375,
      "learning_rate": 4.017787098509555e-06,
      "loss": 1.62922802,
      "memory(GiB)": 117.38,
      "step": 46105,
      "train_speed(iter/s)": 1.635504
    },
    {
      "acc": 0.65588112,
      "epoch": 1.169710806697108,
      "grad_norm": 6.84375,
      "learning_rate": 4.016758928524944e-06,
      "loss": 1.58211861,
      "memory(GiB)": 117.38,
      "step": 46110,
      "train_speed(iter/s)": 1.635523
    },
    {
      "acc": 0.66940403,
      "epoch": 1.169837645865043,
      "grad_norm": 5.78125,
      "learning_rate": 4.015730801787663e-06,
      "loss": 1.51750841,
      "memory(GiB)": 117.38,
      "step": 46115,
      "train_speed(iter/s)": 1.635543
    },
    {
      "acc": 0.64090481,
      "epoch": 1.1699644850329782,
      "grad_norm": 5.46875,
      "learning_rate": 4.014702718342938e-06,
      "loss": 1.66367912,
      "memory(GiB)": 117.38,
      "step": 46120,
      "train_speed(iter/s)": 1.635563
    },
    {
      "acc": 0.64416966,
      "epoch": 1.1700913242009132,
      "grad_norm": 5.28125,
      "learning_rate": 4.013674678235985e-06,
      "loss": 1.63240318,
      "memory(GiB)": 117.38,
      "step": 46125,
      "train_speed(iter/s)": 1.635581
    },
    {
      "acc": 0.65053287,
      "epoch": 1.1702181633688482,
      "grad_norm": 5.40625,
      "learning_rate": 4.012646681512026e-06,
      "loss": 1.58833714,
      "memory(GiB)": 117.38,
      "step": 46130,
      "train_speed(iter/s)": 1.6356
    },
    {
      "acc": 0.65251641,
      "epoch": 1.1703450025367834,
      "grad_norm": 5.84375,
      "learning_rate": 4.011618728216271e-06,
      "loss": 1.64492264,
      "memory(GiB)": 117.38,
      "step": 46135,
      "train_speed(iter/s)": 1.635621
    },
    {
      "acc": 0.65192852,
      "epoch": 1.1704718417047184,
      "grad_norm": 5.96875,
      "learning_rate": 4.010590818393938e-06,
      "loss": 1.63998642,
      "memory(GiB)": 117.38,
      "step": 46140,
      "train_speed(iter/s)": 1.635641
    },
    {
      "acc": 0.67863503,
      "epoch": 1.1705986808726534,
      "grad_norm": 6.4375,
      "learning_rate": 4.009562952090238e-06,
      "loss": 1.55675259,
      "memory(GiB)": 117.38,
      "step": 46145,
      "train_speed(iter/s)": 1.635661
    },
    {
      "acc": 0.65378156,
      "epoch": 1.1707255200405886,
      "grad_norm": 8.875,
      "learning_rate": 4.0085351293503825e-06,
      "loss": 1.57858629,
      "memory(GiB)": 117.38,
      "step": 46150,
      "train_speed(iter/s)": 1.635682
    },
    {
      "acc": 0.65080643,
      "epoch": 1.1708523592085236,
      "grad_norm": 4.40625,
      "learning_rate": 4.007507350219578e-06,
      "loss": 1.68782082,
      "memory(GiB)": 117.38,
      "step": 46155,
      "train_speed(iter/s)": 1.635701
    },
    {
      "acc": 0.66592493,
      "epoch": 1.1709791983764586,
      "grad_norm": 5.5625,
      "learning_rate": 4.0064796147430305e-06,
      "loss": 1.51166668,
      "memory(GiB)": 117.38,
      "step": 46160,
      "train_speed(iter/s)": 1.635721
    },
    {
      "acc": 0.66495256,
      "epoch": 1.1711060375443938,
      "grad_norm": 5.625,
      "learning_rate": 4.005451922965946e-06,
      "loss": 1.61604385,
      "memory(GiB)": 117.38,
      "step": 46165,
      "train_speed(iter/s)": 1.635741
    },
    {
      "acc": 0.66606326,
      "epoch": 1.1712328767123288,
      "grad_norm": 5.125,
      "learning_rate": 4.0044242749335285e-06,
      "loss": 1.55738316,
      "memory(GiB)": 117.38,
      "step": 46170,
      "train_speed(iter/s)": 1.63576
    },
    {
      "acc": 0.64758453,
      "epoch": 1.1713597158802638,
      "grad_norm": 5.96875,
      "learning_rate": 4.0033966706909735e-06,
      "loss": 1.69288578,
      "memory(GiB)": 117.38,
      "step": 46175,
      "train_speed(iter/s)": 1.635779
    },
    {
      "acc": 0.65085211,
      "epoch": 1.1714865550481988,
      "grad_norm": 5.46875,
      "learning_rate": 4.002369110283482e-06,
      "loss": 1.58347263,
      "memory(GiB)": 117.38,
      "step": 46180,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.63970423,
      "epoch": 1.171613394216134,
      "grad_norm": 5.4375,
      "learning_rate": 4.001341593756253e-06,
      "loss": 1.64356747,
      "memory(GiB)": 117.38,
      "step": 46185,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.66085186,
      "epoch": 1.171740233384069,
      "grad_norm": 7.4375,
      "learning_rate": 4.00031412115448e-06,
      "loss": 1.63768768,
      "memory(GiB)": 117.38,
      "step": 46190,
      "train_speed(iter/s)": 1.635837
    },
    {
      "acc": 0.65669694,
      "epoch": 1.1718670725520042,
      "grad_norm": 6.09375,
      "learning_rate": 3.999286692523352e-06,
      "loss": 1.57838173,
      "memory(GiB)": 117.38,
      "step": 46195,
      "train_speed(iter/s)": 1.635859
    },
    {
      "acc": 0.67000427,
      "epoch": 1.1719939117199392,
      "grad_norm": 5.21875,
      "learning_rate": 3.998259307908065e-06,
      "loss": 1.53306589,
      "memory(GiB)": 117.38,
      "step": 46200,
      "train_speed(iter/s)": 1.635878
    },
    {
      "acc": 0.64608984,
      "epoch": 1.1721207508878742,
      "grad_norm": 6.78125,
      "learning_rate": 3.997231967353806e-06,
      "loss": 1.62242565,
      "memory(GiB)": 117.38,
      "step": 46205,
      "train_speed(iter/s)": 1.635899
    },
    {
      "acc": 0.65822487,
      "epoch": 1.1722475900558091,
      "grad_norm": 5.21875,
      "learning_rate": 3.996204670905765e-06,
      "loss": 1.58023291,
      "memory(GiB)": 117.38,
      "step": 46210,
      "train_speed(iter/s)": 1.635918
    },
    {
      "acc": 0.66076365,
      "epoch": 1.1723744292237444,
      "grad_norm": 5.40625,
      "learning_rate": 3.9951774186091195e-06,
      "loss": 1.58831854,
      "memory(GiB)": 117.38,
      "step": 46215,
      "train_speed(iter/s)": 1.635938
    },
    {
      "acc": 0.67891111,
      "epoch": 1.1725012683916793,
      "grad_norm": 6.59375,
      "learning_rate": 3.9941502105090594e-06,
      "loss": 1.52629194,
      "memory(GiB)": 117.38,
      "step": 46220,
      "train_speed(iter/s)": 1.635957
    },
    {
      "acc": 0.67207718,
      "epoch": 1.1726281075596143,
      "grad_norm": 5.65625,
      "learning_rate": 3.9931230466507634e-06,
      "loss": 1.52454329,
      "memory(GiB)": 117.38,
      "step": 46225,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.65900068,
      "epoch": 1.1727549467275495,
      "grad_norm": 5.3125,
      "learning_rate": 3.992095927079412e-06,
      "loss": 1.63008919,
      "memory(GiB)": 117.38,
      "step": 46230,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.67170353,
      "epoch": 1.1728817858954845,
      "grad_norm": 6.96875,
      "learning_rate": 3.991068851840182e-06,
      "loss": 1.52715607,
      "memory(GiB)": 117.38,
      "step": 46235,
      "train_speed(iter/s)": 1.636015
    },
    {
      "acc": 0.6656848,
      "epoch": 1.1730086250634195,
      "grad_norm": 6.125,
      "learning_rate": 3.990041820978246e-06,
      "loss": 1.53950558,
      "memory(GiB)": 117.38,
      "step": 46240,
      "train_speed(iter/s)": 1.636034
    },
    {
      "acc": 0.66400881,
      "epoch": 1.1731354642313547,
      "grad_norm": 6.28125,
      "learning_rate": 3.989014834538782e-06,
      "loss": 1.54228439,
      "memory(GiB)": 117.38,
      "step": 46245,
      "train_speed(iter/s)": 1.636054
    },
    {
      "acc": 0.67212768,
      "epoch": 1.1732623033992897,
      "grad_norm": 6.9375,
      "learning_rate": 3.987987892566959e-06,
      "loss": 1.55905561,
      "memory(GiB)": 117.38,
      "step": 46250,
      "train_speed(iter/s)": 1.636074
    },
    {
      "acc": 0.67626667,
      "epoch": 1.1733891425672247,
      "grad_norm": 4.90625,
      "learning_rate": 3.986960995107948e-06,
      "loss": 1.54889126,
      "memory(GiB)": 117.38,
      "step": 46255,
      "train_speed(iter/s)": 1.636094
    },
    {
      "acc": 0.66708112,
      "epoch": 1.17351598173516,
      "grad_norm": 5.1875,
      "learning_rate": 3.985934142206912e-06,
      "loss": 1.58366413,
      "memory(GiB)": 117.38,
      "step": 46260,
      "train_speed(iter/s)": 1.636113
    },
    {
      "acc": 0.66411142,
      "epoch": 1.173642820903095,
      "grad_norm": 5.375,
      "learning_rate": 3.984907333909022e-06,
      "loss": 1.60409508,
      "memory(GiB)": 117.38,
      "step": 46265,
      "train_speed(iter/s)": 1.636132
    },
    {
      "acc": 0.64951048,
      "epoch": 1.17376966007103,
      "grad_norm": 6.28125,
      "learning_rate": 3.983880570259441e-06,
      "loss": 1.62533855,
      "memory(GiB)": 117.38,
      "step": 46270,
      "train_speed(iter/s)": 1.63615
    },
    {
      "acc": 0.65605001,
      "epoch": 1.1738964992389649,
      "grad_norm": 5.8125,
      "learning_rate": 3.982853851303327e-06,
      "loss": 1.57857924,
      "memory(GiB)": 117.38,
      "step": 46275,
      "train_speed(iter/s)": 1.63617
    },
    {
      "acc": 0.64148264,
      "epoch": 1.1740233384069,
      "grad_norm": 6.6875,
      "learning_rate": 3.981827177085842e-06,
      "loss": 1.67251167,
      "memory(GiB)": 117.38,
      "step": 46280,
      "train_speed(iter/s)": 1.63619
    },
    {
      "acc": 0.64799423,
      "epoch": 1.174150177574835,
      "grad_norm": 6.375,
      "learning_rate": 3.980800547652143e-06,
      "loss": 1.61811638,
      "memory(GiB)": 117.38,
      "step": 46285,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.63844075,
      "epoch": 1.17427701674277,
      "grad_norm": 5.21875,
      "learning_rate": 3.979773963047388e-06,
      "loss": 1.72852268,
      "memory(GiB)": 117.38,
      "step": 46290,
      "train_speed(iter/s)": 1.636231
    },
    {
      "acc": 0.66433301,
      "epoch": 1.1744038559107053,
      "grad_norm": 5.75,
      "learning_rate": 3.978747423316729e-06,
      "loss": 1.55439405,
      "memory(GiB)": 117.38,
      "step": 46295,
      "train_speed(iter/s)": 1.636249
    },
    {
      "acc": 0.64813147,
      "epoch": 1.1745306950786403,
      "grad_norm": 5.34375,
      "learning_rate": 3.977720928505317e-06,
      "loss": 1.66334076,
      "memory(GiB)": 117.38,
      "step": 46300,
      "train_speed(iter/s)": 1.636269
    },
    {
      "acc": 0.6505229,
      "epoch": 1.1746575342465753,
      "grad_norm": 7.0,
      "learning_rate": 3.976694478658301e-06,
      "loss": 1.66106415,
      "memory(GiB)": 117.38,
      "step": 46305,
      "train_speed(iter/s)": 1.636288
    },
    {
      "acc": 0.66378841,
      "epoch": 1.1747843734145105,
      "grad_norm": 6.15625,
      "learning_rate": 3.975668073820834e-06,
      "loss": 1.56796207,
      "memory(GiB)": 117.38,
      "step": 46310,
      "train_speed(iter/s)": 1.636307
    },
    {
      "acc": 0.68842468,
      "epoch": 1.1749112125824455,
      "grad_norm": 6.71875,
      "learning_rate": 3.9746417140380576e-06,
      "loss": 1.49618874,
      "memory(GiB)": 117.38,
      "step": 46315,
      "train_speed(iter/s)": 1.636329
    },
    {
      "acc": 0.65510254,
      "epoch": 1.1750380517503805,
      "grad_norm": 7.625,
      "learning_rate": 3.973615399355114e-06,
      "loss": 1.62154598,
      "memory(GiB)": 117.38,
      "step": 46320,
      "train_speed(iter/s)": 1.636347
    },
    {
      "acc": 0.65530357,
      "epoch": 1.1751648909183157,
      "grad_norm": 5.09375,
      "learning_rate": 3.97258912981715e-06,
      "loss": 1.63688202,
      "memory(GiB)": 117.38,
      "step": 46325,
      "train_speed(iter/s)": 1.636366
    },
    {
      "acc": 0.66143141,
      "epoch": 1.1752917300862507,
      "grad_norm": 6.8125,
      "learning_rate": 3.9715629054693035e-06,
      "loss": 1.55610132,
      "memory(GiB)": 117.38,
      "step": 46330,
      "train_speed(iter/s)": 1.636385
    },
    {
      "acc": 0.66292305,
      "epoch": 1.1754185692541856,
      "grad_norm": 5.6875,
      "learning_rate": 3.970536726356711e-06,
      "loss": 1.59044476,
      "memory(GiB)": 117.38,
      "step": 46335,
      "train_speed(iter/s)": 1.636404
    },
    {
      "acc": 0.64639225,
      "epoch": 1.1755454084221206,
      "grad_norm": 5.4375,
      "learning_rate": 3.969510592524509e-06,
      "loss": 1.62415352,
      "memory(GiB)": 117.38,
      "step": 46340,
      "train_speed(iter/s)": 1.636423
    },
    {
      "acc": 0.65371771,
      "epoch": 1.1756722475900558,
      "grad_norm": 5.03125,
      "learning_rate": 3.968484504017833e-06,
      "loss": 1.64485283,
      "memory(GiB)": 117.38,
      "step": 46345,
      "train_speed(iter/s)": 1.636441
    },
    {
      "acc": 0.67239647,
      "epoch": 1.1757990867579908,
      "grad_norm": 7.46875,
      "learning_rate": 3.967458460881815e-06,
      "loss": 1.52457695,
      "memory(GiB)": 117.38,
      "step": 46350,
      "train_speed(iter/s)": 1.636461
    },
    {
      "acc": 0.67113972,
      "epoch": 1.175925925925926,
      "grad_norm": 5.625,
      "learning_rate": 3.96643246316158e-06,
      "loss": 1.54952898,
      "memory(GiB)": 117.38,
      "step": 46355,
      "train_speed(iter/s)": 1.636479
    },
    {
      "acc": 0.66188698,
      "epoch": 1.176052765093861,
      "grad_norm": 4.84375,
      "learning_rate": 3.965406510902263e-06,
      "loss": 1.53429222,
      "memory(GiB)": 117.38,
      "step": 46360,
      "train_speed(iter/s)": 1.636497
    },
    {
      "acc": 0.67016797,
      "epoch": 1.176179604261796,
      "grad_norm": 6.71875,
      "learning_rate": 3.9643806041489855e-06,
      "loss": 1.55963669,
      "memory(GiB)": 117.38,
      "step": 46365,
      "train_speed(iter/s)": 1.636517
    },
    {
      "acc": 0.65091772,
      "epoch": 1.176306443429731,
      "grad_norm": 6.0,
      "learning_rate": 3.963354742946874e-06,
      "loss": 1.60377045,
      "memory(GiB)": 117.38,
      "step": 46370,
      "train_speed(iter/s)": 1.636535
    },
    {
      "acc": 0.65209856,
      "epoch": 1.1764332825976662,
      "grad_norm": 4.5,
      "learning_rate": 3.962328927341048e-06,
      "loss": 1.6182106,
      "memory(GiB)": 117.38,
      "step": 46375,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.67671337,
      "epoch": 1.1765601217656012,
      "grad_norm": 6.84375,
      "learning_rate": 3.961303157376628e-06,
      "loss": 1.534589,
      "memory(GiB)": 117.38,
      "step": 46380,
      "train_speed(iter/s)": 1.636574
    },
    {
      "acc": 0.64846845,
      "epoch": 1.1766869609335362,
      "grad_norm": 6.53125,
      "learning_rate": 3.960277433098734e-06,
      "loss": 1.57788858,
      "memory(GiB)": 117.38,
      "step": 46385,
      "train_speed(iter/s)": 1.636595
    },
    {
      "acc": 0.65173607,
      "epoch": 1.1768138001014714,
      "grad_norm": 6.5,
      "learning_rate": 3.959251754552481e-06,
      "loss": 1.65551491,
      "memory(GiB)": 117.38,
      "step": 46390,
      "train_speed(iter/s)": 1.636615
    },
    {
      "acc": 0.64662571,
      "epoch": 1.1769406392694064,
      "grad_norm": 5.0,
      "learning_rate": 3.958226121782982e-06,
      "loss": 1.61326485,
      "memory(GiB)": 117.38,
      "step": 46395,
      "train_speed(iter/s)": 1.636634
    },
    {
      "acc": 0.65729303,
      "epoch": 1.1770674784373414,
      "grad_norm": 5.78125,
      "learning_rate": 3.9572005348353486e-06,
      "loss": 1.58313999,
      "memory(GiB)": 117.38,
      "step": 46400,
      "train_speed(iter/s)": 1.636653
    },
    {
      "acc": 0.64366126,
      "epoch": 1.1771943176052766,
      "grad_norm": 7.5625,
      "learning_rate": 3.956174993754691e-06,
      "loss": 1.63947544,
      "memory(GiB)": 117.38,
      "step": 46405,
      "train_speed(iter/s)": 1.636673
    },
    {
      "acc": 0.66744175,
      "epoch": 1.1773211567732116,
      "grad_norm": 4.875,
      "learning_rate": 3.955149498586119e-06,
      "loss": 1.49518757,
      "memory(GiB)": 117.38,
      "step": 46410,
      "train_speed(iter/s)": 1.636693
    },
    {
      "acc": 0.64974432,
      "epoch": 1.1774479959411466,
      "grad_norm": 7.15625,
      "learning_rate": 3.954124049374736e-06,
      "loss": 1.62570076,
      "memory(GiB)": 117.38,
      "step": 46415,
      "train_speed(iter/s)": 1.636713
    },
    {
      "acc": 0.66807938,
      "epoch": 1.1775748351090818,
      "grad_norm": 4.90625,
      "learning_rate": 3.9530986461656465e-06,
      "loss": 1.53927374,
      "memory(GiB)": 117.38,
      "step": 46420,
      "train_speed(iter/s)": 1.636731
    },
    {
      "acc": 0.65424414,
      "epoch": 1.1777016742770168,
      "grad_norm": 6.125,
      "learning_rate": 3.952073289003953e-06,
      "loss": 1.57309361,
      "memory(GiB)": 117.38,
      "step": 46425,
      "train_speed(iter/s)": 1.63675
    },
    {
      "acc": 0.67555165,
      "epoch": 1.1778285134449518,
      "grad_norm": 6.0625,
      "learning_rate": 3.9510479779347566e-06,
      "loss": 1.49967442,
      "memory(GiB)": 117.38,
      "step": 46430,
      "train_speed(iter/s)": 1.636769
    },
    {
      "acc": 0.65240335,
      "epoch": 1.1779553526128868,
      "grad_norm": 5.875,
      "learning_rate": 3.950022713003151e-06,
      "loss": 1.59517345,
      "memory(GiB)": 117.38,
      "step": 46435,
      "train_speed(iter/s)": 1.636788
    },
    {
      "acc": 0.65068264,
      "epoch": 1.178082191780822,
      "grad_norm": 5.53125,
      "learning_rate": 3.9489974942542355e-06,
      "loss": 1.6094698,
      "memory(GiB)": 117.38,
      "step": 46440,
      "train_speed(iter/s)": 1.636807
    },
    {
      "acc": 0.65861225,
      "epoch": 1.178209030948757,
      "grad_norm": 5.84375,
      "learning_rate": 3.947972321733101e-06,
      "loss": 1.57066135,
      "memory(GiB)": 117.38,
      "step": 46445,
      "train_speed(iter/s)": 1.636827
    },
    {
      "acc": 0.65753689,
      "epoch": 1.178335870116692,
      "grad_norm": 5.28125,
      "learning_rate": 3.946947195484843e-06,
      "loss": 1.60609932,
      "memory(GiB)": 117.38,
      "step": 46450,
      "train_speed(iter/s)": 1.636847
    },
    {
      "acc": 0.65107908,
      "epoch": 1.1784627092846272,
      "grad_norm": 5.8125,
      "learning_rate": 3.945922115554548e-06,
      "loss": 1.63570251,
      "memory(GiB)": 117.38,
      "step": 46455,
      "train_speed(iter/s)": 1.636866
    },
    {
      "acc": 0.66598992,
      "epoch": 1.1785895484525621,
      "grad_norm": 4.9375,
      "learning_rate": 3.944897081987303e-06,
      "loss": 1.6481945,
      "memory(GiB)": 117.38,
      "step": 46460,
      "train_speed(iter/s)": 1.636886
    },
    {
      "acc": 0.65200796,
      "epoch": 1.1787163876204971,
      "grad_norm": 8.25,
      "learning_rate": 3.943872094828197e-06,
      "loss": 1.6242588,
      "memory(GiB)": 117.38,
      "step": 46465,
      "train_speed(iter/s)": 1.636907
    },
    {
      "acc": 0.6718935,
      "epoch": 1.1788432267884323,
      "grad_norm": 5.125,
      "learning_rate": 3.942847154122312e-06,
      "loss": 1.51998882,
      "memory(GiB)": 117.38,
      "step": 46470,
      "train_speed(iter/s)": 1.636927
    },
    {
      "acc": 0.6716465,
      "epoch": 1.1789700659563673,
      "grad_norm": 6.40625,
      "learning_rate": 3.941822259914728e-06,
      "loss": 1.51986094,
      "memory(GiB)": 117.38,
      "step": 46475,
      "train_speed(iter/s)": 1.636947
    },
    {
      "acc": 0.66168108,
      "epoch": 1.1790969051243023,
      "grad_norm": 4.71875,
      "learning_rate": 3.940797412250524e-06,
      "loss": 1.58920403,
      "memory(GiB)": 117.38,
      "step": 46480,
      "train_speed(iter/s)": 1.636966
    },
    {
      "acc": 0.65981693,
      "epoch": 1.1792237442922375,
      "grad_norm": 4.625,
      "learning_rate": 3.93977261117478e-06,
      "loss": 1.60927353,
      "memory(GiB)": 117.38,
      "step": 46485,
      "train_speed(iter/s)": 1.636985
    },
    {
      "acc": 0.65412836,
      "epoch": 1.1793505834601725,
      "grad_norm": 5.0625,
      "learning_rate": 3.938747856732572e-06,
      "loss": 1.59747219,
      "memory(GiB)": 117.38,
      "step": 46490,
      "train_speed(iter/s)": 1.637004
    },
    {
      "acc": 0.67093582,
      "epoch": 1.1794774226281075,
      "grad_norm": 5.84375,
      "learning_rate": 3.9377231489689685e-06,
      "loss": 1.62085819,
      "memory(GiB)": 117.38,
      "step": 46495,
      "train_speed(iter/s)": 1.637023
    },
    {
      "acc": 0.65070896,
      "epoch": 1.1796042617960425,
      "grad_norm": 5.46875,
      "learning_rate": 3.936698487929045e-06,
      "loss": 1.61765633,
      "memory(GiB)": 117.38,
      "step": 46500,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.66881008,
      "epoch": 1.1797311009639777,
      "grad_norm": 5.03125,
      "learning_rate": 3.935673873657868e-06,
      "loss": 1.54812698,
      "memory(GiB)": 117.38,
      "step": 46505,
      "train_speed(iter/s)": 1.637062
    },
    {
      "acc": 0.65189404,
      "epoch": 1.1798579401319127,
      "grad_norm": 6.78125,
      "learning_rate": 3.934649306200508e-06,
      "loss": 1.65766869,
      "memory(GiB)": 117.38,
      "step": 46510,
      "train_speed(iter/s)": 1.637081
    },
    {
      "acc": 0.65781188,
      "epoch": 1.179984779299848,
      "grad_norm": 5.5625,
      "learning_rate": 3.933624785602027e-06,
      "loss": 1.60242882,
      "memory(GiB)": 117.38,
      "step": 46515,
      "train_speed(iter/s)": 1.637099
    },
    {
      "acc": 0.64470778,
      "epoch": 1.180111618467783,
      "grad_norm": 5.15625,
      "learning_rate": 3.932600311907489e-06,
      "loss": 1.61195717,
      "memory(GiB)": 117.38,
      "step": 46520,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.6503222,
      "epoch": 1.1802384576357179,
      "grad_norm": 4.84375,
      "learning_rate": 3.931575885161955e-06,
      "loss": 1.6452364,
      "memory(GiB)": 117.38,
      "step": 46525,
      "train_speed(iter/s)": 1.637139
    },
    {
      "acc": 0.64573269,
      "epoch": 1.1803652968036529,
      "grad_norm": 5.4375,
      "learning_rate": 3.930551505410484e-06,
      "loss": 1.60496044,
      "memory(GiB)": 117.38,
      "step": 46530,
      "train_speed(iter/s)": 1.637157
    },
    {
      "acc": 0.66018629,
      "epoch": 1.180492135971588,
      "grad_norm": 5.59375,
      "learning_rate": 3.929527172698132e-06,
      "loss": 1.60194473,
      "memory(GiB)": 117.38,
      "step": 46535,
      "train_speed(iter/s)": 1.637176
    },
    {
      "acc": 0.64571295,
      "epoch": 1.180618975139523,
      "grad_norm": 4.9375,
      "learning_rate": 3.928502887069954e-06,
      "loss": 1.60859489,
      "memory(GiB)": 117.38,
      "step": 46540,
      "train_speed(iter/s)": 1.637194
    },
    {
      "acc": 0.67574348,
      "epoch": 1.180745814307458,
      "grad_norm": 6.46875,
      "learning_rate": 3.927478648571003e-06,
      "loss": 1.48915281,
      "memory(GiB)": 117.38,
      "step": 46545,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.65071945,
      "epoch": 1.1808726534753933,
      "grad_norm": 6.71875,
      "learning_rate": 3.926454457246331e-06,
      "loss": 1.58496666,
      "memory(GiB)": 117.38,
      "step": 46550,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.64780989,
      "epoch": 1.1809994926433283,
      "grad_norm": 6.5625,
      "learning_rate": 3.9254303131409834e-06,
      "loss": 1.63332024,
      "memory(GiB)": 117.38,
      "step": 46555,
      "train_speed(iter/s)": 1.637247
    },
    {
      "acc": 0.65483541,
      "epoch": 1.1811263318112633,
      "grad_norm": 6.65625,
      "learning_rate": 3.924406216300009e-06,
      "loss": 1.54679193,
      "memory(GiB)": 117.38,
      "step": 46560,
      "train_speed(iter/s)": 1.637266
    },
    {
      "acc": 0.66200409,
      "epoch": 1.1812531709791985,
      "grad_norm": 5.53125,
      "learning_rate": 3.92338216676845e-06,
      "loss": 1.59225531,
      "memory(GiB)": 117.38,
      "step": 46565,
      "train_speed(iter/s)": 1.637284
    },
    {
      "acc": 0.64841104,
      "epoch": 1.1813800101471335,
      "grad_norm": 7.0625,
      "learning_rate": 3.922358164591353e-06,
      "loss": 1.65743294,
      "memory(GiB)": 117.38,
      "step": 46570,
      "train_speed(iter/s)": 1.637303
    },
    {
      "acc": 0.65236654,
      "epoch": 1.1815068493150684,
      "grad_norm": 5.59375,
      "learning_rate": 3.921334209813752e-06,
      "loss": 1.64063339,
      "memory(GiB)": 117.38,
      "step": 46575,
      "train_speed(iter/s)": 1.637321
    },
    {
      "acc": 0.65048494,
      "epoch": 1.1816336884830037,
      "grad_norm": 5.40625,
      "learning_rate": 3.92031030248069e-06,
      "loss": 1.63058529,
      "memory(GiB)": 117.38,
      "step": 46580,
      "train_speed(iter/s)": 1.637339
    },
    {
      "acc": 0.63938637,
      "epoch": 1.1817605276509386,
      "grad_norm": 5.34375,
      "learning_rate": 3.9192864426372e-06,
      "loss": 1.66378632,
      "memory(GiB)": 117.38,
      "step": 46585,
      "train_speed(iter/s)": 1.637358
    },
    {
      "acc": 0.67071295,
      "epoch": 1.1818873668188736,
      "grad_norm": 5.53125,
      "learning_rate": 3.918262630328319e-06,
      "loss": 1.5374176,
      "memory(GiB)": 117.38,
      "step": 46590,
      "train_speed(iter/s)": 1.637376
    },
    {
      "acc": 0.66215916,
      "epoch": 1.1820142059868086,
      "grad_norm": 4.875,
      "learning_rate": 3.917238865599077e-06,
      "loss": 1.59837723,
      "memory(GiB)": 117.38,
      "step": 46595,
      "train_speed(iter/s)": 1.637394
    },
    {
      "acc": 0.68290043,
      "epoch": 1.1821410451547438,
      "grad_norm": 5.34375,
      "learning_rate": 3.916215148494502e-06,
      "loss": 1.49271126,
      "memory(GiB)": 117.38,
      "step": 46600,
      "train_speed(iter/s)": 1.637412
    },
    {
      "acc": 0.63864908,
      "epoch": 1.1822678843226788,
      "grad_norm": 6.625,
      "learning_rate": 3.9151914790596255e-06,
      "loss": 1.66548157,
      "memory(GiB)": 117.38,
      "step": 46605,
      "train_speed(iter/s)": 1.63743
    },
    {
      "acc": 0.66377449,
      "epoch": 1.1823947234906138,
      "grad_norm": 6.71875,
      "learning_rate": 3.914167857339472e-06,
      "loss": 1.57629261,
      "memory(GiB)": 117.38,
      "step": 46610,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.64592457,
      "epoch": 1.182521562658549,
      "grad_norm": 5.8125,
      "learning_rate": 3.913144283379061e-06,
      "loss": 1.63500137,
      "memory(GiB)": 117.38,
      "step": 46615,
      "train_speed(iter/s)": 1.637469
    },
    {
      "acc": 0.6622323,
      "epoch": 1.182648401826484,
      "grad_norm": 6.40625,
      "learning_rate": 3.912120757223418e-06,
      "loss": 1.67293892,
      "memory(GiB)": 117.38,
      "step": 46620,
      "train_speed(iter/s)": 1.637487
    },
    {
      "acc": 0.667243,
      "epoch": 1.182775240994419,
      "grad_norm": 6.625,
      "learning_rate": 3.911097278917561e-06,
      "loss": 1.59625711,
      "memory(GiB)": 117.38,
      "step": 46625,
      "train_speed(iter/s)": 1.637506
    },
    {
      "acc": 0.65368738,
      "epoch": 1.1829020801623542,
      "grad_norm": 6.1875,
      "learning_rate": 3.91007384850651e-06,
      "loss": 1.57545414,
      "memory(GiB)": 117.38,
      "step": 46630,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.66435733,
      "epoch": 1.1830289193302892,
      "grad_norm": 5.28125,
      "learning_rate": 3.909050466035274e-06,
      "loss": 1.52634087,
      "memory(GiB)": 117.38,
      "step": 46635,
      "train_speed(iter/s)": 1.637542
    },
    {
      "acc": 0.65399513,
      "epoch": 1.1831557584982242,
      "grad_norm": 6.375,
      "learning_rate": 3.90802713154887e-06,
      "loss": 1.61824646,
      "memory(GiB)": 117.38,
      "step": 46640,
      "train_speed(iter/s)": 1.63756
    },
    {
      "acc": 0.67156687,
      "epoch": 1.1832825976661594,
      "grad_norm": 5.59375,
      "learning_rate": 3.9070038450923074e-06,
      "loss": 1.50965052,
      "memory(GiB)": 117.38,
      "step": 46645,
      "train_speed(iter/s)": 1.637578
    },
    {
      "acc": 0.64963045,
      "epoch": 1.1834094368340944,
      "grad_norm": 6.46875,
      "learning_rate": 3.9059806067105985e-06,
      "loss": 1.6243679,
      "memory(GiB)": 117.38,
      "step": 46650,
      "train_speed(iter/s)": 1.637597
    },
    {
      "acc": 0.65542622,
      "epoch": 1.1835362760020294,
      "grad_norm": 6.5625,
      "learning_rate": 3.904957416448744e-06,
      "loss": 1.64733887,
      "memory(GiB)": 117.38,
      "step": 46655,
      "train_speed(iter/s)": 1.637615
    },
    {
      "acc": 0.66609335,
      "epoch": 1.1836631151699644,
      "grad_norm": 6.21875,
      "learning_rate": 3.903934274351753e-06,
      "loss": 1.58386116,
      "memory(GiB)": 117.38,
      "step": 46660,
      "train_speed(iter/s)": 1.637633
    },
    {
      "acc": 0.66952004,
      "epoch": 1.1837899543378996,
      "grad_norm": 6.09375,
      "learning_rate": 3.9029111804646245e-06,
      "loss": 1.5711689,
      "memory(GiB)": 117.38,
      "step": 46665,
      "train_speed(iter/s)": 1.637651
    },
    {
      "acc": 0.65237675,
      "epoch": 1.1839167935058346,
      "grad_norm": 6.0,
      "learning_rate": 3.9018881348323626e-06,
      "loss": 1.60015182,
      "memory(GiB)": 117.38,
      "step": 46670,
      "train_speed(iter/s)": 1.637671
    },
    {
      "acc": 0.66778841,
      "epoch": 1.1840436326737698,
      "grad_norm": 4.71875,
      "learning_rate": 3.9008651374999615e-06,
      "loss": 1.56850548,
      "memory(GiB)": 117.38,
      "step": 46675,
      "train_speed(iter/s)": 1.637689
    },
    {
      "acc": 0.65507216,
      "epoch": 1.1841704718417048,
      "grad_norm": 5.28125,
      "learning_rate": 3.899842188512419e-06,
      "loss": 1.59441872,
      "memory(GiB)": 117.38,
      "step": 46680,
      "train_speed(iter/s)": 1.637706
    },
    {
      "acc": 0.65663738,
      "epoch": 1.1842973110096398,
      "grad_norm": 4.75,
      "learning_rate": 3.898819287914729e-06,
      "loss": 1.56393642,
      "memory(GiB)": 117.38,
      "step": 46685,
      "train_speed(iter/s)": 1.637725
    },
    {
      "acc": 0.65431476,
      "epoch": 1.1844241501775747,
      "grad_norm": 6.0625,
      "learning_rate": 3.897796435751885e-06,
      "loss": 1.6180912,
      "memory(GiB)": 117.38,
      "step": 46690,
      "train_speed(iter/s)": 1.637743
    },
    {
      "acc": 0.6413754,
      "epoch": 1.18455098934551,
      "grad_norm": 6.375,
      "learning_rate": 3.896773632068873e-06,
      "loss": 1.65171318,
      "memory(GiB)": 117.38,
      "step": 46695,
      "train_speed(iter/s)": 1.637761
    },
    {
      "acc": 0.65373116,
      "epoch": 1.184677828513445,
      "grad_norm": 5.71875,
      "learning_rate": 3.8957508769106825e-06,
      "loss": 1.61337585,
      "memory(GiB)": 117.38,
      "step": 46700,
      "train_speed(iter/s)": 1.63778
    },
    {
      "acc": 0.66132746,
      "epoch": 1.18480466768138,
      "grad_norm": 4.8125,
      "learning_rate": 3.894728170322298e-06,
      "loss": 1.61730728,
      "memory(GiB)": 117.38,
      "step": 46705,
      "train_speed(iter/s)": 1.637798
    },
    {
      "acc": 0.67446327,
      "epoch": 1.1849315068493151,
      "grad_norm": 5.65625,
      "learning_rate": 3.893705512348705e-06,
      "loss": 1.5165102,
      "memory(GiB)": 117.38,
      "step": 46710,
      "train_speed(iter/s)": 1.637817
    },
    {
      "acc": 0.6756988,
      "epoch": 1.1850583460172501,
      "grad_norm": 4.6875,
      "learning_rate": 3.89268290303488e-06,
      "loss": 1.5932478,
      "memory(GiB)": 117.38,
      "step": 46715,
      "train_speed(iter/s)": 1.637833
    },
    {
      "acc": 0.66301041,
      "epoch": 1.1851851851851851,
      "grad_norm": 5.125,
      "learning_rate": 3.891660342425807e-06,
      "loss": 1.5150445,
      "memory(GiB)": 117.38,
      "step": 46720,
      "train_speed(iter/s)": 1.637851
    },
    {
      "acc": 0.67786813,
      "epoch": 1.1853120243531203,
      "grad_norm": 6.75,
      "learning_rate": 3.890637830566459e-06,
      "loss": 1.50318928,
      "memory(GiB)": 117.38,
      "step": 46725,
      "train_speed(iter/s)": 1.63787
    },
    {
      "acc": 0.65086222,
      "epoch": 1.1854388635210553,
      "grad_norm": 7.375,
      "learning_rate": 3.889615367501815e-06,
      "loss": 1.64343834,
      "memory(GiB)": 117.38,
      "step": 46730,
      "train_speed(iter/s)": 1.637887
    },
    {
      "acc": 0.66089849,
      "epoch": 1.1855657026889903,
      "grad_norm": 4.84375,
      "learning_rate": 3.888592953276842e-06,
      "loss": 1.59469709,
      "memory(GiB)": 117.38,
      "step": 46735,
      "train_speed(iter/s)": 1.637904
    },
    {
      "acc": 0.67004609,
      "epoch": 1.1856925418569255,
      "grad_norm": 5.0,
      "learning_rate": 3.8875705879365135e-06,
      "loss": 1.50744762,
      "memory(GiB)": 117.38,
      "step": 46740,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.65864906,
      "epoch": 1.1858193810248605,
      "grad_norm": 6.15625,
      "learning_rate": 3.886548271525797e-06,
      "loss": 1.55310087,
      "memory(GiB)": 117.38,
      "step": 46745,
      "train_speed(iter/s)": 1.637941
    },
    {
      "acc": 0.66627016,
      "epoch": 1.1859462201927955,
      "grad_norm": 6.78125,
      "learning_rate": 3.88552600408966e-06,
      "loss": 1.53203316,
      "memory(GiB)": 117.38,
      "step": 46750,
      "train_speed(iter/s)": 1.63796
    },
    {
      "acc": 0.65460253,
      "epoch": 1.1860730593607305,
      "grad_norm": 6.53125,
      "learning_rate": 3.8845037856730646e-06,
      "loss": 1.66312294,
      "memory(GiB)": 117.38,
      "step": 46755,
      "train_speed(iter/s)": 1.637979
    },
    {
      "acc": 0.6571754,
      "epoch": 1.1861998985286657,
      "grad_norm": 5.71875,
      "learning_rate": 3.883481616320972e-06,
      "loss": 1.65006809,
      "memory(GiB)": 117.38,
      "step": 46760,
      "train_speed(iter/s)": 1.637998
    },
    {
      "acc": 0.66273699,
      "epoch": 1.1863267376966007,
      "grad_norm": 6.0625,
      "learning_rate": 3.882459496078343e-06,
      "loss": 1.60198097,
      "memory(GiB)": 117.38,
      "step": 46765,
      "train_speed(iter/s)": 1.638015
    },
    {
      "acc": 0.66143765,
      "epoch": 1.1864535768645357,
      "grad_norm": 7.71875,
      "learning_rate": 3.881437424990137e-06,
      "loss": 1.5615799,
      "memory(GiB)": 117.38,
      "step": 46770,
      "train_speed(iter/s)": 1.638033
    },
    {
      "acc": 0.65304499,
      "epoch": 1.1865804160324709,
      "grad_norm": 5.96875,
      "learning_rate": 3.880415403101304e-06,
      "loss": 1.60328407,
      "memory(GiB)": 117.38,
      "step": 46775,
      "train_speed(iter/s)": 1.638052
    },
    {
      "acc": 0.67337389,
      "epoch": 1.1867072552004059,
      "grad_norm": 5.96875,
      "learning_rate": 3.879393430456801e-06,
      "loss": 1.50395489,
      "memory(GiB)": 117.38,
      "step": 46780,
      "train_speed(iter/s)": 1.638067
    },
    {
      "acc": 0.65592957,
      "epoch": 1.1868340943683409,
      "grad_norm": 5.28125,
      "learning_rate": 3.87837150710158e-06,
      "loss": 1.63608093,
      "memory(GiB)": 117.38,
      "step": 46785,
      "train_speed(iter/s)": 1.638085
    },
    {
      "acc": 0.66808338,
      "epoch": 1.186960933536276,
      "grad_norm": 6.46875,
      "learning_rate": 3.877349633080587e-06,
      "loss": 1.50783844,
      "memory(GiB)": 117.38,
      "step": 46790,
      "train_speed(iter/s)": 1.638104
    },
    {
      "acc": 0.66317434,
      "epoch": 1.187087772704211,
      "grad_norm": 5.84375,
      "learning_rate": 3.876327808438767e-06,
      "loss": 1.61880226,
      "memory(GiB)": 117.38,
      "step": 46795,
      "train_speed(iter/s)": 1.638121
    },
    {
      "acc": 0.66592922,
      "epoch": 1.187214611872146,
      "grad_norm": 5.625,
      "learning_rate": 3.875306033221069e-06,
      "loss": 1.53188953,
      "memory(GiB)": 117.38,
      "step": 46800,
      "train_speed(iter/s)": 1.638141
    },
    {
      "acc": 0.64452128,
      "epoch": 1.1873414510400813,
      "grad_norm": 6.0625,
      "learning_rate": 3.874284307472432e-06,
      "loss": 1.66028233,
      "memory(GiB)": 117.38,
      "step": 46805,
      "train_speed(iter/s)": 1.638158
    },
    {
      "acc": 0.6567986,
      "epoch": 1.1874682902080163,
      "grad_norm": 5.40625,
      "learning_rate": 3.873262631237799e-06,
      "loss": 1.63864193,
      "memory(GiB)": 117.38,
      "step": 46810,
      "train_speed(iter/s)": 1.638176
    },
    {
      "acc": 0.64220586,
      "epoch": 1.1875951293759512,
      "grad_norm": 5.6875,
      "learning_rate": 3.872241004562105e-06,
      "loss": 1.65563698,
      "memory(GiB)": 117.38,
      "step": 46815,
      "train_speed(iter/s)": 1.638195
    },
    {
      "acc": 0.67266512,
      "epoch": 1.1877219685438862,
      "grad_norm": 7.75,
      "learning_rate": 3.871219427490285e-06,
      "loss": 1.54418936,
      "memory(GiB)": 117.38,
      "step": 46820,
      "train_speed(iter/s)": 1.638215
    },
    {
      "acc": 0.64564209,
      "epoch": 1.1878488077118214,
      "grad_norm": 6.75,
      "learning_rate": 3.870197900067276e-06,
      "loss": 1.64141026,
      "memory(GiB)": 117.38,
      "step": 46825,
      "train_speed(iter/s)": 1.638234
    },
    {
      "acc": 0.65323009,
      "epoch": 1.1879756468797564,
      "grad_norm": 5.71875,
      "learning_rate": 3.869176422338009e-06,
      "loss": 1.5795701,
      "memory(GiB)": 117.38,
      "step": 46830,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.6664484,
      "epoch": 1.1881024860476916,
      "grad_norm": 7.84375,
      "learning_rate": 3.868154994347409e-06,
      "loss": 1.59463205,
      "memory(GiB)": 117.38,
      "step": 46835,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.67412996,
      "epoch": 1.1882293252156266,
      "grad_norm": 6.875,
      "learning_rate": 3.867133616140406e-06,
      "loss": 1.53573093,
      "memory(GiB)": 117.38,
      "step": 46840,
      "train_speed(iter/s)": 1.63829
    },
    {
      "acc": 0.66909733,
      "epoch": 1.1883561643835616,
      "grad_norm": 6.1875,
      "learning_rate": 3.866112287761926e-06,
      "loss": 1.63225441,
      "memory(GiB)": 117.38,
      "step": 46845,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.6454906,
      "epoch": 1.1884830035514966,
      "grad_norm": 6.65625,
      "learning_rate": 3.86509100925689e-06,
      "loss": 1.62626801,
      "memory(GiB)": 117.38,
      "step": 46850,
      "train_speed(iter/s)": 1.638326
    },
    {
      "acc": 0.65258489,
      "epoch": 1.1886098427194318,
      "grad_norm": 5.28125,
      "learning_rate": 3.8640697806702166e-06,
      "loss": 1.58213806,
      "memory(GiB)": 117.38,
      "step": 46855,
      "train_speed(iter/s)": 1.638344
    },
    {
      "acc": 0.65415382,
      "epoch": 1.1887366818873668,
      "grad_norm": 6.34375,
      "learning_rate": 3.8630486020468265e-06,
      "loss": 1.64464989,
      "memory(GiB)": 117.38,
      "step": 46860,
      "train_speed(iter/s)": 1.638362
    },
    {
      "acc": 0.65757532,
      "epoch": 1.1888635210553018,
      "grad_norm": 5.875,
      "learning_rate": 3.862027473431634e-06,
      "loss": 1.59540005,
      "memory(GiB)": 117.38,
      "step": 46865,
      "train_speed(iter/s)": 1.638381
    },
    {
      "acc": 0.66706028,
      "epoch": 1.188990360223237,
      "grad_norm": 7.4375,
      "learning_rate": 3.861006394869558e-06,
      "loss": 1.55304089,
      "memory(GiB)": 117.38,
      "step": 46870,
      "train_speed(iter/s)": 1.638401
    },
    {
      "acc": 0.64790449,
      "epoch": 1.189117199391172,
      "grad_norm": 5.21875,
      "learning_rate": 3.859985366405502e-06,
      "loss": 1.63214626,
      "memory(GiB)": 117.38,
      "step": 46875,
      "train_speed(iter/s)": 1.63842
    },
    {
      "acc": 0.65056038,
      "epoch": 1.189244038559107,
      "grad_norm": 4.6875,
      "learning_rate": 3.85896438808438e-06,
      "loss": 1.59073887,
      "memory(GiB)": 117.38,
      "step": 46880,
      "train_speed(iter/s)": 1.638439
    },
    {
      "acc": 0.64938898,
      "epoch": 1.1893708777270422,
      "grad_norm": 6.03125,
      "learning_rate": 3.857943459951099e-06,
      "loss": 1.60665169,
      "memory(GiB)": 117.38,
      "step": 46885,
      "train_speed(iter/s)": 1.638456
    },
    {
      "acc": 0.65199142,
      "epoch": 1.1894977168949772,
      "grad_norm": 6.28125,
      "learning_rate": 3.856922582050565e-06,
      "loss": 1.63446999,
      "memory(GiB)": 117.38,
      "step": 46890,
      "train_speed(iter/s)": 1.638475
    },
    {
      "acc": 0.67521443,
      "epoch": 1.1896245560629122,
      "grad_norm": 5.8125,
      "learning_rate": 3.855901754427678e-06,
      "loss": 1.48979893,
      "memory(GiB)": 117.38,
      "step": 46895,
      "train_speed(iter/s)": 1.638492
    },
    {
      "acc": 0.63665633,
      "epoch": 1.1897513952308474,
      "grad_norm": 6.09375,
      "learning_rate": 3.854880977127339e-06,
      "loss": 1.68439789,
      "memory(GiB)": 117.38,
      "step": 46900,
      "train_speed(iter/s)": 1.63851
    },
    {
      "acc": 0.65221219,
      "epoch": 1.1898782343987824,
      "grad_norm": 5.84375,
      "learning_rate": 3.8538602501944475e-06,
      "loss": 1.55919418,
      "memory(GiB)": 117.38,
      "step": 46905,
      "train_speed(iter/s)": 1.638528
    },
    {
      "acc": 0.65973887,
      "epoch": 1.1900050735667174,
      "grad_norm": 7.0625,
      "learning_rate": 3.852839573673902e-06,
      "loss": 1.58765039,
      "memory(GiB)": 117.38,
      "step": 46910,
      "train_speed(iter/s)": 1.638546
    },
    {
      "acc": 0.66354303,
      "epoch": 1.1901319127346524,
      "grad_norm": 5.46875,
      "learning_rate": 3.851818947610591e-06,
      "loss": 1.55327473,
      "memory(GiB)": 117.38,
      "step": 46915,
      "train_speed(iter/s)": 1.638565
    },
    {
      "acc": 0.65805817,
      "epoch": 1.1902587519025876,
      "grad_norm": 6.0625,
      "learning_rate": 3.850798372049409e-06,
      "loss": 1.57910995,
      "memory(GiB)": 117.38,
      "step": 46920,
      "train_speed(iter/s)": 1.638583
    },
    {
      "acc": 0.66380262,
      "epoch": 1.1903855910705226,
      "grad_norm": 5.21875,
      "learning_rate": 3.849777847035246e-06,
      "loss": 1.55484943,
      "memory(GiB)": 117.38,
      "step": 46925,
      "train_speed(iter/s)": 1.638602
    },
    {
      "acc": 0.67047691,
      "epoch": 1.1905124302384575,
      "grad_norm": 6.0,
      "learning_rate": 3.84875737261299e-06,
      "loss": 1.49261589,
      "memory(GiB)": 117.38,
      "step": 46930,
      "train_speed(iter/s)": 1.638621
    },
    {
      "acc": 0.65380487,
      "epoch": 1.1906392694063928,
      "grad_norm": 8.3125,
      "learning_rate": 3.847736948827523e-06,
      "loss": 1.64001846,
      "memory(GiB)": 117.38,
      "step": 46935,
      "train_speed(iter/s)": 1.638638
    },
    {
      "acc": 0.66836205,
      "epoch": 1.1907661085743277,
      "grad_norm": 6.375,
      "learning_rate": 3.846716575723729e-06,
      "loss": 1.54684496,
      "memory(GiB)": 117.38,
      "step": 46940,
      "train_speed(iter/s)": 1.638657
    },
    {
      "acc": 0.65215092,
      "epoch": 1.1908929477422627,
      "grad_norm": 4.78125,
      "learning_rate": 3.845696253346489e-06,
      "loss": 1.63874702,
      "memory(GiB)": 117.38,
      "step": 46945,
      "train_speed(iter/s)": 1.638676
    },
    {
      "acc": 0.66589842,
      "epoch": 1.191019786910198,
      "grad_norm": 5.59375,
      "learning_rate": 3.8446759817406835e-06,
      "loss": 1.58005333,
      "memory(GiB)": 117.38,
      "step": 46950,
      "train_speed(iter/s)": 1.638692
    },
    {
      "acc": 0.66307163,
      "epoch": 1.191146626078133,
      "grad_norm": 5.03125,
      "learning_rate": 3.8436557609511856e-06,
      "loss": 1.61401005,
      "memory(GiB)": 117.38,
      "step": 46955,
      "train_speed(iter/s)": 1.638711
    },
    {
      "acc": 0.67856674,
      "epoch": 1.191273465246068,
      "grad_norm": 5.625,
      "learning_rate": 3.842635591022869e-06,
      "loss": 1.45879087,
      "memory(GiB)": 117.38,
      "step": 46960,
      "train_speed(iter/s)": 1.638729
    },
    {
      "acc": 0.6582922,
      "epoch": 1.1914003044140031,
      "grad_norm": 7.0,
      "learning_rate": 3.8416154720006065e-06,
      "loss": 1.60393963,
      "memory(GiB)": 117.38,
      "step": 46965,
      "train_speed(iter/s)": 1.638747
    },
    {
      "acc": 0.66006708,
      "epoch": 1.1915271435819381,
      "grad_norm": 4.8125,
      "learning_rate": 3.840595403929269e-06,
      "loss": 1.58464184,
      "memory(GiB)": 117.38,
      "step": 46970,
      "train_speed(iter/s)": 1.638765
    },
    {
      "acc": 0.642693,
      "epoch": 1.191653982749873,
      "grad_norm": 5.9375,
      "learning_rate": 3.839575386853721e-06,
      "loss": 1.57218618,
      "memory(GiB)": 117.38,
      "step": 46975,
      "train_speed(iter/s)": 1.638784
    },
    {
      "acc": 0.65564809,
      "epoch": 1.191780821917808,
      "grad_norm": 5.21875,
      "learning_rate": 3.838555420818827e-06,
      "loss": 1.55778847,
      "memory(GiB)": 117.38,
      "step": 46980,
      "train_speed(iter/s)": 1.638802
    },
    {
      "acc": 0.66052308,
      "epoch": 1.1919076610857433,
      "grad_norm": 5.84375,
      "learning_rate": 3.837535505869453e-06,
      "loss": 1.58639746,
      "memory(GiB)": 117.38,
      "step": 46985,
      "train_speed(iter/s)": 1.638821
    },
    {
      "acc": 0.65801759,
      "epoch": 1.1920345002536783,
      "grad_norm": 5.9375,
      "learning_rate": 3.836515642050458e-06,
      "loss": 1.56650105,
      "memory(GiB)": 117.38,
      "step": 46990,
      "train_speed(iter/s)": 1.638838
    },
    {
      "acc": 0.63576598,
      "epoch": 1.1921613394216135,
      "grad_norm": 5.09375,
      "learning_rate": 3.835495829406698e-06,
      "loss": 1.61660118,
      "memory(GiB)": 117.38,
      "step": 46995,
      "train_speed(iter/s)": 1.638857
    },
    {
      "acc": 0.66022162,
      "epoch": 1.1922881785895485,
      "grad_norm": 6.125,
      "learning_rate": 3.834476067983031e-06,
      "loss": 1.60717754,
      "memory(GiB)": 117.38,
      "step": 47000,
      "train_speed(iter/s)": 1.638875
    },
    {
      "epoch": 1.1922881785895485,
      "eval_acc": 0.6462300853908369,
      "eval_loss": 1.5734953880310059,
      "eval_runtime": 58.9324,
      "eval_samples_per_second": 108.09,
      "eval_steps_per_second": 27.031,
      "step": 47000
    },
    {
      "acc": 0.65335493,
      "epoch": 1.1924150177574835,
      "grad_norm": 5.59375,
      "learning_rate": 3.83345635782431e-06,
      "loss": 1.66574478,
      "memory(GiB)": 117.38,
      "step": 47005,
      "train_speed(iter/s)": 1.635285
    },
    {
      "acc": 0.6601388,
      "epoch": 1.1925418569254185,
      "grad_norm": 6.71875,
      "learning_rate": 3.832436698975388e-06,
      "loss": 1.61326675,
      "memory(GiB)": 117.38,
      "step": 47010,
      "train_speed(iter/s)": 1.635303
    },
    {
      "acc": 0.67149887,
      "epoch": 1.1926686960933537,
      "grad_norm": 5.75,
      "learning_rate": 3.831417091481111e-06,
      "loss": 1.53228226,
      "memory(GiB)": 117.38,
      "step": 47015,
      "train_speed(iter/s)": 1.635322
    },
    {
      "acc": 0.67479167,
      "epoch": 1.1927955352612887,
      "grad_norm": 5.3125,
      "learning_rate": 3.830397535386328e-06,
      "loss": 1.56198921,
      "memory(GiB)": 117.38,
      "step": 47020,
      "train_speed(iter/s)": 1.635339
    },
    {
      "acc": 0.65695648,
      "epoch": 1.1929223744292237,
      "grad_norm": 6.34375,
      "learning_rate": 3.829378030735883e-06,
      "loss": 1.61739349,
      "memory(GiB)": 117.38,
      "step": 47025,
      "train_speed(iter/s)": 1.635357
    },
    {
      "acc": 0.66081076,
      "epoch": 1.1930492135971589,
      "grad_norm": 5.125,
      "learning_rate": 3.82835857757462e-06,
      "loss": 1.59440498,
      "memory(GiB)": 117.38,
      "step": 47030,
      "train_speed(iter/s)": 1.635375
    },
    {
      "acc": 0.65916214,
      "epoch": 1.1931760527650939,
      "grad_norm": 4.59375,
      "learning_rate": 3.827339175947378e-06,
      "loss": 1.620998,
      "memory(GiB)": 117.38,
      "step": 47035,
      "train_speed(iter/s)": 1.635393
    },
    {
      "acc": 0.65532503,
      "epoch": 1.1933028919330289,
      "grad_norm": 6.0625,
      "learning_rate": 3.826319825898992e-06,
      "loss": 1.55248051,
      "memory(GiB)": 117.38,
      "step": 47040,
      "train_speed(iter/s)": 1.63541
    },
    {
      "acc": 0.66407051,
      "epoch": 1.193429731100964,
      "grad_norm": 5.90625,
      "learning_rate": 3.825300527474302e-06,
      "loss": 1.58592749,
      "memory(GiB)": 117.38,
      "step": 47045,
      "train_speed(iter/s)": 1.635428
    },
    {
      "acc": 0.65257359,
      "epoch": 1.193556570268899,
      "grad_norm": 6.40625,
      "learning_rate": 3.824281280718141e-06,
      "loss": 1.65087433,
      "memory(GiB)": 117.38,
      "step": 47050,
      "train_speed(iter/s)": 1.635445
    },
    {
      "acc": 0.65596094,
      "epoch": 1.193683409436834,
      "grad_norm": 7.65625,
      "learning_rate": 3.823262085675337e-06,
      "loss": 1.59080009,
      "memory(GiB)": 117.38,
      "step": 47055,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.64660559,
      "epoch": 1.1938102486047693,
      "grad_norm": 6.0,
      "learning_rate": 3.822242942390718e-06,
      "loss": 1.64604568,
      "memory(GiB)": 117.38,
      "step": 47060,
      "train_speed(iter/s)": 1.635481
    },
    {
      "acc": 0.66204052,
      "epoch": 1.1939370877727042,
      "grad_norm": 5.65625,
      "learning_rate": 3.821223850909115e-06,
      "loss": 1.60249252,
      "memory(GiB)": 117.38,
      "step": 47065,
      "train_speed(iter/s)": 1.635498
    },
    {
      "acc": 0.65209904,
      "epoch": 1.1940639269406392,
      "grad_norm": 6.125,
      "learning_rate": 3.820204811275351e-06,
      "loss": 1.63516426,
      "memory(GiB)": 117.38,
      "step": 47070,
      "train_speed(iter/s)": 1.635517
    },
    {
      "acc": 0.66477499,
      "epoch": 1.1941907661085742,
      "grad_norm": 5.1875,
      "learning_rate": 3.8191858235342446e-06,
      "loss": 1.61611938,
      "memory(GiB)": 117.38,
      "step": 47075,
      "train_speed(iter/s)": 1.635535
    },
    {
      "acc": 0.64243298,
      "epoch": 1.1943176052765094,
      "grad_norm": 7.09375,
      "learning_rate": 3.818166887730618e-06,
      "loss": 1.63458061,
      "memory(GiB)": 117.38,
      "step": 47080,
      "train_speed(iter/s)": 1.635554
    },
    {
      "acc": 0.64643278,
      "epoch": 1.1944444444444444,
      "grad_norm": 5.28125,
      "learning_rate": 3.817148003909288e-06,
      "loss": 1.67639561,
      "memory(GiB)": 117.38,
      "step": 47085,
      "train_speed(iter/s)": 1.635573
    },
    {
      "acc": 0.65950713,
      "epoch": 1.1945712836123794,
      "grad_norm": 5.21875,
      "learning_rate": 3.816129172115073e-06,
      "loss": 1.58476839,
      "memory(GiB)": 117.38,
      "step": 47090,
      "train_speed(iter/s)": 1.635593
    },
    {
      "acc": 0.64445076,
      "epoch": 1.1946981227803146,
      "grad_norm": 6.3125,
      "learning_rate": 3.815110392392778e-06,
      "loss": 1.62359428,
      "memory(GiB)": 117.38,
      "step": 47095,
      "train_speed(iter/s)": 1.635611
    },
    {
      "acc": 0.6504828,
      "epoch": 1.1948249619482496,
      "grad_norm": 5.40625,
      "learning_rate": 3.8140916647872204e-06,
      "loss": 1.59726362,
      "memory(GiB)": 117.38,
      "step": 47100,
      "train_speed(iter/s)": 1.63563
    },
    {
      "acc": 0.64806466,
      "epoch": 1.1949518011161846,
      "grad_norm": 6.84375,
      "learning_rate": 3.813072989343205e-06,
      "loss": 1.63519783,
      "memory(GiB)": 117.38,
      "step": 47105,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.65763292,
      "epoch": 1.1950786402841198,
      "grad_norm": 5.28125,
      "learning_rate": 3.812054366105541e-06,
      "loss": 1.59549179,
      "memory(GiB)": 117.38,
      "step": 47110,
      "train_speed(iter/s)": 1.635667
    },
    {
      "acc": 0.66659985,
      "epoch": 1.1952054794520548,
      "grad_norm": 5.6875,
      "learning_rate": 3.8110357951190284e-06,
      "loss": 1.49302597,
      "memory(GiB)": 117.38,
      "step": 47115,
      "train_speed(iter/s)": 1.635686
    },
    {
      "acc": 0.62299747,
      "epoch": 1.1953323186199898,
      "grad_norm": 4.96875,
      "learning_rate": 3.8100172764284694e-06,
      "loss": 1.68569717,
      "memory(GiB)": 117.38,
      "step": 47120,
      "train_speed(iter/s)": 1.635704
    },
    {
      "acc": 0.67369828,
      "epoch": 1.195459157787925,
      "grad_norm": 5.5,
      "learning_rate": 3.8089988100786635e-06,
      "loss": 1.56372166,
      "memory(GiB)": 117.38,
      "step": 47125,
      "train_speed(iter/s)": 1.635722
    },
    {
      "acc": 0.65630674,
      "epoch": 1.19558599695586,
      "grad_norm": 6.875,
      "learning_rate": 3.807980396114409e-06,
      "loss": 1.55270224,
      "memory(GiB)": 117.38,
      "step": 47130,
      "train_speed(iter/s)": 1.635742
    },
    {
      "acc": 0.64401217,
      "epoch": 1.195712836123795,
      "grad_norm": 5.4375,
      "learning_rate": 3.8069620345804974e-06,
      "loss": 1.60807705,
      "memory(GiB)": 117.38,
      "step": 47135,
      "train_speed(iter/s)": 1.63576
    },
    {
      "acc": 0.65233974,
      "epoch": 1.19583967529173,
      "grad_norm": 5.46875,
      "learning_rate": 3.8059437255217214e-06,
      "loss": 1.65920963,
      "memory(GiB)": 117.38,
      "step": 47140,
      "train_speed(iter/s)": 1.635779
    },
    {
      "acc": 0.66710558,
      "epoch": 1.1959665144596652,
      "grad_norm": 5.5625,
      "learning_rate": 3.8049254689828723e-06,
      "loss": 1.55732212,
      "memory(GiB)": 117.38,
      "step": 47145,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.65564604,
      "epoch": 1.1960933536276002,
      "grad_norm": 6.4375,
      "learning_rate": 3.8039072650087377e-06,
      "loss": 1.58524265,
      "memory(GiB)": 117.38,
      "step": 47150,
      "train_speed(iter/s)": 1.635817
    },
    {
      "acc": 0.65488806,
      "epoch": 1.1962201927955354,
      "grad_norm": 4.96875,
      "learning_rate": 3.8028891136440994e-06,
      "loss": 1.55358696,
      "memory(GiB)": 117.38,
      "step": 47155,
      "train_speed(iter/s)": 1.635837
    },
    {
      "acc": 0.64907103,
      "epoch": 1.1963470319634704,
      "grad_norm": 4.90625,
      "learning_rate": 3.801871014933744e-06,
      "loss": 1.58658066,
      "memory(GiB)": 117.38,
      "step": 47160,
      "train_speed(iter/s)": 1.635855
    },
    {
      "acc": 0.64994564,
      "epoch": 1.1964738711314054,
      "grad_norm": 6.21875,
      "learning_rate": 3.8008529689224493e-06,
      "loss": 1.60585403,
      "memory(GiB)": 117.38,
      "step": 47165,
      "train_speed(iter/s)": 1.635873
    },
    {
      "acc": 0.64978976,
      "epoch": 1.1966007102993403,
      "grad_norm": 6.28125,
      "learning_rate": 3.7998349756549974e-06,
      "loss": 1.60660477,
      "memory(GiB)": 117.38,
      "step": 47170,
      "train_speed(iter/s)": 1.635892
    },
    {
      "acc": 0.63557205,
      "epoch": 1.1967275494672756,
      "grad_norm": 5.5625,
      "learning_rate": 3.79881703517616e-06,
      "loss": 1.74159088,
      "memory(GiB)": 117.38,
      "step": 47175,
      "train_speed(iter/s)": 1.63591
    },
    {
      "acc": 0.66288357,
      "epoch": 1.1968543886352105,
      "grad_norm": 6.65625,
      "learning_rate": 3.797799147530713e-06,
      "loss": 1.61801758,
      "memory(GiB)": 117.38,
      "step": 47180,
      "train_speed(iter/s)": 1.635929
    },
    {
      "acc": 0.65557961,
      "epoch": 1.1969812278031455,
      "grad_norm": 6.125,
      "learning_rate": 3.796781312763425e-06,
      "loss": 1.5821619,
      "memory(GiB)": 117.38,
      "step": 47185,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.65856256,
      "epoch": 1.1971080669710807,
      "grad_norm": 5.5,
      "learning_rate": 3.79576353091907e-06,
      "loss": 1.50847769,
      "memory(GiB)": 117.38,
      "step": 47190,
      "train_speed(iter/s)": 1.635966
    },
    {
      "acc": 0.67521152,
      "epoch": 1.1972349061390157,
      "grad_norm": 6.21875,
      "learning_rate": 3.7947458020424094e-06,
      "loss": 1.56782303,
      "memory(GiB)": 117.38,
      "step": 47195,
      "train_speed(iter/s)": 1.635986
    },
    {
      "acc": 0.65726056,
      "epoch": 1.1973617453069507,
      "grad_norm": 4.875,
      "learning_rate": 3.793728126178209e-06,
      "loss": 1.603899,
      "memory(GiB)": 117.38,
      "step": 47200,
      "train_speed(iter/s)": 1.636001
    },
    {
      "acc": 0.66021533,
      "epoch": 1.197488584474886,
      "grad_norm": 6.0,
      "learning_rate": 3.792710503371232e-06,
      "loss": 1.52472906,
      "memory(GiB)": 117.38,
      "step": 47205,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.65039697,
      "epoch": 1.197615423642821,
      "grad_norm": 6.03125,
      "learning_rate": 3.7916929336662386e-06,
      "loss": 1.72632256,
      "memory(GiB)": 117.38,
      "step": 47210,
      "train_speed(iter/s)": 1.636038
    },
    {
      "acc": 0.65964088,
      "epoch": 1.197742262810756,
      "grad_norm": 5.53125,
      "learning_rate": 3.790675417107982e-06,
      "loss": 1.57490911,
      "memory(GiB)": 117.38,
      "step": 47215,
      "train_speed(iter/s)": 1.636055
    },
    {
      "acc": 0.65741148,
      "epoch": 1.1978691019786911,
      "grad_norm": 6.125,
      "learning_rate": 3.7896579537412213e-06,
      "loss": 1.56715117,
      "memory(GiB)": 117.38,
      "step": 47220,
      "train_speed(iter/s)": 1.636075
    },
    {
      "acc": 0.63056674,
      "epoch": 1.197995941146626,
      "grad_norm": 5.375,
      "learning_rate": 3.7886405436107076e-06,
      "loss": 1.69390068,
      "memory(GiB)": 117.38,
      "step": 47225,
      "train_speed(iter/s)": 1.636095
    },
    {
      "acc": 0.64632359,
      "epoch": 1.198122780314561,
      "grad_norm": 5.875,
      "learning_rate": 3.7876231867611917e-06,
      "loss": 1.61246586,
      "memory(GiB)": 117.38,
      "step": 47230,
      "train_speed(iter/s)": 1.636112
    },
    {
      "acc": 0.64396238,
      "epoch": 1.198249619482496,
      "grad_norm": 6.46875,
      "learning_rate": 3.7866058832374197e-06,
      "loss": 1.65640335,
      "memory(GiB)": 117.38,
      "step": 47235,
      "train_speed(iter/s)": 1.63613
    },
    {
      "acc": 0.65456672,
      "epoch": 1.1983764586504313,
      "grad_norm": 5.65625,
      "learning_rate": 3.7855886330841383e-06,
      "loss": 1.64422455,
      "memory(GiB)": 117.38,
      "step": 47240,
      "train_speed(iter/s)": 1.636149
    },
    {
      "acc": 0.65773706,
      "epoch": 1.1985032978183663,
      "grad_norm": 6.40625,
      "learning_rate": 3.7845714363460908e-06,
      "loss": 1.60695267,
      "memory(GiB)": 117.38,
      "step": 47245,
      "train_speed(iter/s)": 1.636167
    },
    {
      "acc": 0.66238337,
      "epoch": 1.1986301369863013,
      "grad_norm": 5.125,
      "learning_rate": 3.78355429306802e-06,
      "loss": 1.60711021,
      "memory(GiB)": 117.38,
      "step": 47250,
      "train_speed(iter/s)": 1.636184
    },
    {
      "acc": 0.66299458,
      "epoch": 1.1987569761542365,
      "grad_norm": 5.875,
      "learning_rate": 3.7825372032946605e-06,
      "loss": 1.55074139,
      "memory(GiB)": 117.38,
      "step": 47255,
      "train_speed(iter/s)": 1.636202
    },
    {
      "acc": 0.6659749,
      "epoch": 1.1988838153221715,
      "grad_norm": 5.5625,
      "learning_rate": 3.7815201670707502e-06,
      "loss": 1.58210211,
      "memory(GiB)": 117.38,
      "step": 47260,
      "train_speed(iter/s)": 1.636221
    },
    {
      "acc": 0.64726868,
      "epoch": 1.1990106544901065,
      "grad_norm": 6.21875,
      "learning_rate": 3.7805031844410235e-06,
      "loss": 1.67704773,
      "memory(GiB)": 117.38,
      "step": 47265,
      "train_speed(iter/s)": 1.636239
    },
    {
      "acc": 0.66102352,
      "epoch": 1.1991374936580417,
      "grad_norm": 5.21875,
      "learning_rate": 3.7794862554502126e-06,
      "loss": 1.55312347,
      "memory(GiB)": 117.38,
      "step": 47270,
      "train_speed(iter/s)": 1.636256
    },
    {
      "acc": 0.65674496,
      "epoch": 1.1992643328259767,
      "grad_norm": 6.59375,
      "learning_rate": 3.778469380143045e-06,
      "loss": 1.56913891,
      "memory(GiB)": 117.38,
      "step": 47275,
      "train_speed(iter/s)": 1.636275
    },
    {
      "acc": 0.659342,
      "epoch": 1.1993911719939117,
      "grad_norm": 5.40625,
      "learning_rate": 3.777452558564246e-06,
      "loss": 1.56910448,
      "memory(GiB)": 117.38,
      "step": 47280,
      "train_speed(iter/s)": 1.636293
    },
    {
      "acc": 0.66164751,
      "epoch": 1.1995180111618469,
      "grad_norm": 6.125,
      "learning_rate": 3.776435790758543e-06,
      "loss": 1.5565259,
      "memory(GiB)": 117.38,
      "step": 47285,
      "train_speed(iter/s)": 1.636313
    },
    {
      "acc": 0.66742363,
      "epoch": 1.1996448503297819,
      "grad_norm": 8.125,
      "learning_rate": 3.7754190767706577e-06,
      "loss": 1.49027596,
      "memory(GiB)": 117.38,
      "step": 47290,
      "train_speed(iter/s)": 1.636332
    },
    {
      "acc": 0.65464725,
      "epoch": 1.1997716894977168,
      "grad_norm": 6.5625,
      "learning_rate": 3.774402416645307e-06,
      "loss": 1.62291069,
      "memory(GiB)": 117.38,
      "step": 47295,
      "train_speed(iter/s)": 1.63635
    },
    {
      "acc": 0.65059195,
      "epoch": 1.1998985286656518,
      "grad_norm": 5.53125,
      "learning_rate": 3.77338581042721e-06,
      "loss": 1.61271954,
      "memory(GiB)": 117.38,
      "step": 47300,
      "train_speed(iter/s)": 1.636368
    },
    {
      "acc": 0.66629171,
      "epoch": 1.200025367833587,
      "grad_norm": 6.625,
      "learning_rate": 3.7723692581610817e-06,
      "loss": 1.62539082,
      "memory(GiB)": 117.38,
      "step": 47305,
      "train_speed(iter/s)": 1.636388
    },
    {
      "acc": 0.66174989,
      "epoch": 1.200152207001522,
      "grad_norm": 6.40625,
      "learning_rate": 3.771352759891637e-06,
      "loss": 1.67625656,
      "memory(GiB)": 117.38,
      "step": 47310,
      "train_speed(iter/s)": 1.636406
    },
    {
      "acc": 0.6528501,
      "epoch": 1.2002790461694572,
      "grad_norm": 5.71875,
      "learning_rate": 3.7703363156635807e-06,
      "loss": 1.67193146,
      "memory(GiB)": 117.38,
      "step": 47315,
      "train_speed(iter/s)": 1.636425
    },
    {
      "acc": 0.66145048,
      "epoch": 1.2004058853373922,
      "grad_norm": 5.625,
      "learning_rate": 3.769319925521624e-06,
      "loss": 1.60974789,
      "memory(GiB)": 117.38,
      "step": 47320,
      "train_speed(iter/s)": 1.636445
    },
    {
      "acc": 0.66777153,
      "epoch": 1.2005327245053272,
      "grad_norm": 6.34375,
      "learning_rate": 3.76830358951047e-06,
      "loss": 1.59220238,
      "memory(GiB)": 117.38,
      "step": 47325,
      "train_speed(iter/s)": 1.636464
    },
    {
      "acc": 0.65019341,
      "epoch": 1.2006595636732622,
      "grad_norm": 6.59375,
      "learning_rate": 3.767287307674826e-06,
      "loss": 1.54215317,
      "memory(GiB)": 117.38,
      "step": 47330,
      "train_speed(iter/s)": 1.636483
    },
    {
      "acc": 0.65536499,
      "epoch": 1.2007864028411974,
      "grad_norm": 6.0625,
      "learning_rate": 3.766271080059389e-06,
      "loss": 1.59739056,
      "memory(GiB)": 117.38,
      "step": 47335,
      "train_speed(iter/s)": 1.636501
    },
    {
      "acc": 0.65930529,
      "epoch": 1.2009132420091324,
      "grad_norm": 6.875,
      "learning_rate": 3.7652549067088568e-06,
      "loss": 1.56609592,
      "memory(GiB)": 117.38,
      "step": 47340,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.66127963,
      "epoch": 1.2010400811770674,
      "grad_norm": 7.21875,
      "learning_rate": 3.7642387876679275e-06,
      "loss": 1.62591248,
      "memory(GiB)": 117.38,
      "step": 47345,
      "train_speed(iter/s)": 1.636538
    },
    {
      "acc": 0.65140915,
      "epoch": 1.2011669203450026,
      "grad_norm": 4.40625,
      "learning_rate": 3.7632227229812947e-06,
      "loss": 1.60852928,
      "memory(GiB)": 117.38,
      "step": 47350,
      "train_speed(iter/s)": 1.636555
    },
    {
      "acc": 0.65252433,
      "epoch": 1.2012937595129376,
      "grad_norm": 6.65625,
      "learning_rate": 3.7622067126936475e-06,
      "loss": 1.63069954,
      "memory(GiB)": 117.38,
      "step": 47355,
      "train_speed(iter/s)": 1.636573
    },
    {
      "acc": 0.65450258,
      "epoch": 1.2014205986808726,
      "grad_norm": 5.53125,
      "learning_rate": 3.761190756849674e-06,
      "loss": 1.57487478,
      "memory(GiB)": 117.38,
      "step": 47360,
      "train_speed(iter/s)": 1.636592
    },
    {
      "acc": 0.65323658,
      "epoch": 1.2015474378488078,
      "grad_norm": 5.96875,
      "learning_rate": 3.7601748554940633e-06,
      "loss": 1.60449963,
      "memory(GiB)": 117.38,
      "step": 47365,
      "train_speed(iter/s)": 1.636612
    },
    {
      "acc": 0.65836024,
      "epoch": 1.2016742770167428,
      "grad_norm": 5.25,
      "learning_rate": 3.7591590086714984e-06,
      "loss": 1.52106028,
      "memory(GiB)": 117.38,
      "step": 47370,
      "train_speed(iter/s)": 1.63663
    },
    {
      "acc": 0.64723535,
      "epoch": 1.2018011161846778,
      "grad_norm": 7.78125,
      "learning_rate": 3.7581432164266587e-06,
      "loss": 1.57000971,
      "memory(GiB)": 117.38,
      "step": 47375,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.65823426,
      "epoch": 1.201927955352613,
      "grad_norm": 5.6875,
      "learning_rate": 3.7571274788042255e-06,
      "loss": 1.58905058,
      "memory(GiB)": 117.38,
      "step": 47380,
      "train_speed(iter/s)": 1.636667
    },
    {
      "acc": 0.65346446,
      "epoch": 1.202054794520548,
      "grad_norm": 5.75,
      "learning_rate": 3.756111795848874e-06,
      "loss": 1.62489872,
      "memory(GiB)": 117.38,
      "step": 47385,
      "train_speed(iter/s)": 1.636687
    },
    {
      "acc": 0.65487256,
      "epoch": 1.202181633688483,
      "grad_norm": 5.78125,
      "learning_rate": 3.755096167605281e-06,
      "loss": 1.64551773,
      "memory(GiB)": 117.38,
      "step": 47390,
      "train_speed(iter/s)": 1.636706
    },
    {
      "acc": 0.67012811,
      "epoch": 1.202308472856418,
      "grad_norm": 5.40625,
      "learning_rate": 3.7540805941181165e-06,
      "loss": 1.53682079,
      "memory(GiB)": 117.38,
      "step": 47395,
      "train_speed(iter/s)": 1.636726
    },
    {
      "acc": 0.65593185,
      "epoch": 1.2024353120243532,
      "grad_norm": 5.28125,
      "learning_rate": 3.7530650754320492e-06,
      "loss": 1.66485558,
      "memory(GiB)": 117.38,
      "step": 47400,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.68244362,
      "epoch": 1.2025621511922882,
      "grad_norm": 5.0,
      "learning_rate": 3.752049611591746e-06,
      "loss": 1.49088202,
      "memory(GiB)": 117.38,
      "step": 47405,
      "train_speed(iter/s)": 1.636763
    },
    {
      "acc": 0.65459929,
      "epoch": 1.2026889903602231,
      "grad_norm": 5.90625,
      "learning_rate": 3.7510342026418756e-06,
      "loss": 1.63167744,
      "memory(GiB)": 117.38,
      "step": 47410,
      "train_speed(iter/s)": 1.636782
    },
    {
      "acc": 0.65682025,
      "epoch": 1.2028158295281584,
      "grad_norm": 6.28125,
      "learning_rate": 3.7500188486270948e-06,
      "loss": 1.58512192,
      "memory(GiB)": 117.38,
      "step": 47415,
      "train_speed(iter/s)": 1.636801
    },
    {
      "acc": 0.64577866,
      "epoch": 1.2029426686960933,
      "grad_norm": 6.15625,
      "learning_rate": 3.7490035495920664e-06,
      "loss": 1.66989307,
      "memory(GiB)": 117.38,
      "step": 47420,
      "train_speed(iter/s)": 1.63682
    },
    {
      "acc": 0.66750875,
      "epoch": 1.2030695078640283,
      "grad_norm": 5.09375,
      "learning_rate": 3.747988305581447e-06,
      "loss": 1.54458151,
      "memory(GiB)": 117.38,
      "step": 47425,
      "train_speed(iter/s)": 1.636839
    },
    {
      "acc": 0.65450706,
      "epoch": 1.2031963470319635,
      "grad_norm": 5.71875,
      "learning_rate": 3.7469731166398933e-06,
      "loss": 1.62785263,
      "memory(GiB)": 117.38,
      "step": 47430,
      "train_speed(iter/s)": 1.636857
    },
    {
      "acc": 0.64825401,
      "epoch": 1.2033231861998985,
      "grad_norm": 5.34375,
      "learning_rate": 3.745957982812054e-06,
      "loss": 1.63887672,
      "memory(GiB)": 117.38,
      "step": 47435,
      "train_speed(iter/s)": 1.636876
    },
    {
      "acc": 0.64772463,
      "epoch": 1.2034500253678335,
      "grad_norm": 5.40625,
      "learning_rate": 3.744942904142582e-06,
      "loss": 1.60876656,
      "memory(GiB)": 117.38,
      "step": 47440,
      "train_speed(iter/s)": 1.636894
    },
    {
      "acc": 0.67257252,
      "epoch": 1.2035768645357687,
      "grad_norm": 4.90625,
      "learning_rate": 3.743927880676125e-06,
      "loss": 1.53730631,
      "memory(GiB)": 117.38,
      "step": 47445,
      "train_speed(iter/s)": 1.636912
    },
    {
      "acc": 0.67098913,
      "epoch": 1.2037037037037037,
      "grad_norm": 4.75,
      "learning_rate": 3.742912912457329e-06,
      "loss": 1.5400835,
      "memory(GiB)": 117.38,
      "step": 47450,
      "train_speed(iter/s)": 1.63693
    },
    {
      "acc": 0.65312138,
      "epoch": 1.2038305428716387,
      "grad_norm": 5.6875,
      "learning_rate": 3.7418979995308336e-06,
      "loss": 1.59181395,
      "memory(GiB)": 117.38,
      "step": 47455,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.65536499,
      "epoch": 1.2039573820395737,
      "grad_norm": 5.21875,
      "learning_rate": 3.740883141941282e-06,
      "loss": 1.549055,
      "memory(GiB)": 117.38,
      "step": 47460,
      "train_speed(iter/s)": 1.636966
    },
    {
      "acc": 0.66491232,
      "epoch": 1.204084221207509,
      "grad_norm": 5.6875,
      "learning_rate": 3.7398683397333103e-06,
      "loss": 1.61014748,
      "memory(GiB)": 117.38,
      "step": 47465,
      "train_speed(iter/s)": 1.636985
    },
    {
      "acc": 0.6578342,
      "epoch": 1.204211060375444,
      "grad_norm": 4.78125,
      "learning_rate": 3.7388535929515573e-06,
      "loss": 1.55983791,
      "memory(GiB)": 117.38,
      "step": 47470,
      "train_speed(iter/s)": 1.637004
    },
    {
      "acc": 0.66928706,
      "epoch": 1.204337899543379,
      "grad_norm": 6.125,
      "learning_rate": 3.737838901640653e-06,
      "loss": 1.5119709,
      "memory(GiB)": 117.38,
      "step": 47475,
      "train_speed(iter/s)": 1.637022
    },
    {
      "acc": 0.66631021,
      "epoch": 1.204464738711314,
      "grad_norm": 7.96875,
      "learning_rate": 3.736824265845228e-06,
      "loss": 1.57373447,
      "memory(GiB)": 117.38,
      "step": 47480,
      "train_speed(iter/s)": 1.63704
    },
    {
      "acc": 0.66325426,
      "epoch": 1.204591577879249,
      "grad_norm": 5.40625,
      "learning_rate": 3.7358096856099118e-06,
      "loss": 1.57546215,
      "memory(GiB)": 117.38,
      "step": 47485,
      "train_speed(iter/s)": 1.63706
    },
    {
      "acc": 0.66052427,
      "epoch": 1.204718417047184,
      "grad_norm": 5.8125,
      "learning_rate": 3.7347951609793315e-06,
      "loss": 1.55634375,
      "memory(GiB)": 117.38,
      "step": 47490,
      "train_speed(iter/s)": 1.637079
    },
    {
      "acc": 0.64717827,
      "epoch": 1.2048452562151193,
      "grad_norm": 6.375,
      "learning_rate": 3.7337806919981077e-06,
      "loss": 1.599263,
      "memory(GiB)": 117.38,
      "step": 47495,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.66443281,
      "epoch": 1.2049720953830543,
      "grad_norm": 5.84375,
      "learning_rate": 3.732766278710861e-06,
      "loss": 1.55201912,
      "memory(GiB)": 117.38,
      "step": 47500,
      "train_speed(iter/s)": 1.637114
    },
    {
      "acc": 0.66117811,
      "epoch": 1.2050989345509893,
      "grad_norm": 5.28125,
      "learning_rate": 3.7317519211622123e-06,
      "loss": 1.58700237,
      "memory(GiB)": 117.38,
      "step": 47505,
      "train_speed(iter/s)": 1.637133
    },
    {
      "acc": 0.65153933,
      "epoch": 1.2052257737189245,
      "grad_norm": 5.8125,
      "learning_rate": 3.7307376193967772e-06,
      "loss": 1.63909302,
      "memory(GiB)": 117.38,
      "step": 47510,
      "train_speed(iter/s)": 1.637152
    },
    {
      "acc": 0.65395942,
      "epoch": 1.2053526128868595,
      "grad_norm": 5.90625,
      "learning_rate": 3.7297233734591664e-06,
      "loss": 1.59631739,
      "memory(GiB)": 117.38,
      "step": 47515,
      "train_speed(iter/s)": 1.637169
    },
    {
      "acc": 0.65194464,
      "epoch": 1.2054794520547945,
      "grad_norm": 6.71875,
      "learning_rate": 3.7287091833939948e-06,
      "loss": 1.65482063,
      "memory(GiB)": 117.38,
      "step": 47520,
      "train_speed(iter/s)": 1.637186
    },
    {
      "acc": 0.67703719,
      "epoch": 1.2056062912227297,
      "grad_norm": 5.28125,
      "learning_rate": 3.7276950492458675e-06,
      "loss": 1.51242867,
      "memory(GiB)": 117.38,
      "step": 47525,
      "train_speed(iter/s)": 1.637205
    },
    {
      "acc": 0.6592165,
      "epoch": 1.2057331303906647,
      "grad_norm": 5.34375,
      "learning_rate": 3.7266809710593956e-06,
      "loss": 1.58366737,
      "memory(GiB)": 117.38,
      "step": 47530,
      "train_speed(iter/s)": 1.637224
    },
    {
      "acc": 0.66053681,
      "epoch": 1.2058599695585996,
      "grad_norm": 6.78125,
      "learning_rate": 3.7256669488791763e-06,
      "loss": 1.57286053,
      "memory(GiB)": 117.38,
      "step": 47535,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.66028633,
      "epoch": 1.2059868087265349,
      "grad_norm": 5.84375,
      "learning_rate": 3.7246529827498156e-06,
      "loss": 1.51440296,
      "memory(GiB)": 117.38,
      "step": 47540,
      "train_speed(iter/s)": 1.637261
    },
    {
      "acc": 0.65863075,
      "epoch": 1.2061136478944698,
      "grad_norm": 5.4375,
      "learning_rate": 3.7236390727159094e-06,
      "loss": 1.58927555,
      "memory(GiB)": 117.38,
      "step": 47545,
      "train_speed(iter/s)": 1.63728
    },
    {
      "acc": 0.66223278,
      "epoch": 1.2062404870624048,
      "grad_norm": 5.0625,
      "learning_rate": 3.7226252188220573e-06,
      "loss": 1.53088074,
      "memory(GiB)": 117.38,
      "step": 47550,
      "train_speed(iter/s)": 1.637299
    },
    {
      "acc": 0.662673,
      "epoch": 1.2063673262303398,
      "grad_norm": 6.28125,
      "learning_rate": 3.7216114211128505e-06,
      "loss": 1.57977867,
      "memory(GiB)": 117.38,
      "step": 47555,
      "train_speed(iter/s)": 1.637318
    },
    {
      "acc": 0.6630456,
      "epoch": 1.206494165398275,
      "grad_norm": 6.375,
      "learning_rate": 3.720597679632879e-06,
      "loss": 1.56296463,
      "memory(GiB)": 117.38,
      "step": 47560,
      "train_speed(iter/s)": 1.637336
    },
    {
      "acc": 0.65326242,
      "epoch": 1.20662100456621,
      "grad_norm": 7.53125,
      "learning_rate": 3.7195839944267357e-06,
      "loss": 1.68292923,
      "memory(GiB)": 117.38,
      "step": 47565,
      "train_speed(iter/s)": 1.637355
    },
    {
      "acc": 0.66865773,
      "epoch": 1.206747843734145,
      "grad_norm": 4.9375,
      "learning_rate": 3.718570365539006e-06,
      "loss": 1.51249352,
      "memory(GiB)": 117.38,
      "step": 47570,
      "train_speed(iter/s)": 1.637373
    },
    {
      "acc": 0.66530428,
      "epoch": 1.2068746829020802,
      "grad_norm": 6.28125,
      "learning_rate": 3.717556793014271e-06,
      "loss": 1.60455418,
      "memory(GiB)": 117.38,
      "step": 47575,
      "train_speed(iter/s)": 1.637391
    },
    {
      "acc": 0.65385714,
      "epoch": 1.2070015220700152,
      "grad_norm": 5.3125,
      "learning_rate": 3.716543276897113e-06,
      "loss": 1.65254364,
      "memory(GiB)": 117.38,
      "step": 47580,
      "train_speed(iter/s)": 1.63741
    },
    {
      "acc": 0.6525938,
      "epoch": 1.2071283612379502,
      "grad_norm": 5.03125,
      "learning_rate": 3.715529817232114e-06,
      "loss": 1.58671665,
      "memory(GiB)": 117.38,
      "step": 47585,
      "train_speed(iter/s)": 1.637428
    },
    {
      "acc": 0.65385737,
      "epoch": 1.2072552004058854,
      "grad_norm": 6.5625,
      "learning_rate": 3.7145164140638483e-06,
      "loss": 1.61153679,
      "memory(GiB)": 117.38,
      "step": 47590,
      "train_speed(iter/s)": 1.637447
    },
    {
      "acc": 0.64803929,
      "epoch": 1.2073820395738204,
      "grad_norm": 5.5,
      "learning_rate": 3.713503067436889e-06,
      "loss": 1.62950382,
      "memory(GiB)": 117.38,
      "step": 47595,
      "train_speed(iter/s)": 1.637465
    },
    {
      "acc": 0.65665474,
      "epoch": 1.2075088787417554,
      "grad_norm": 6.8125,
      "learning_rate": 3.7124897773958084e-06,
      "loss": 1.57136297,
      "memory(GiB)": 117.38,
      "step": 47600,
      "train_speed(iter/s)": 1.637484
    },
    {
      "acc": 0.65311942,
      "epoch": 1.2076357179096906,
      "grad_norm": 6.15625,
      "learning_rate": 3.7114765439851752e-06,
      "loss": 1.69280968,
      "memory(GiB)": 117.38,
      "step": 47605,
      "train_speed(iter/s)": 1.637502
    },
    {
      "acc": 0.66585712,
      "epoch": 1.2077625570776256,
      "grad_norm": 5.03125,
      "learning_rate": 3.7104633672495584e-06,
      "loss": 1.54819841,
      "memory(GiB)": 117.38,
      "step": 47610,
      "train_speed(iter/s)": 1.637521
    },
    {
      "acc": 0.67353563,
      "epoch": 1.2078893962455606,
      "grad_norm": 7.875,
      "learning_rate": 3.709450247233519e-06,
      "loss": 1.48895893,
      "memory(GiB)": 117.38,
      "step": 47615,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.63848372,
      "epoch": 1.2080162354134956,
      "grad_norm": 10.4375,
      "learning_rate": 3.7084371839816204e-06,
      "loss": 1.6762125,
      "memory(GiB)": 117.38,
      "step": 47620,
      "train_speed(iter/s)": 1.637559
    },
    {
      "acc": 0.67311344,
      "epoch": 1.2081430745814308,
      "grad_norm": 4.5625,
      "learning_rate": 3.707424177538419e-06,
      "loss": 1.5390255,
      "memory(GiB)": 117.38,
      "step": 47625,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.66044602,
      "epoch": 1.2082699137493658,
      "grad_norm": 5.625,
      "learning_rate": 3.7064112279484753e-06,
      "loss": 1.55703392,
      "memory(GiB)": 117.38,
      "step": 47630,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.66800637,
      "epoch": 1.208396752917301,
      "grad_norm": 6.5625,
      "learning_rate": 3.7053983352563407e-06,
      "loss": 1.58711338,
      "memory(GiB)": 117.38,
      "step": 47635,
      "train_speed(iter/s)": 1.637614
    },
    {
      "acc": 0.66317296,
      "epoch": 1.208523592085236,
      "grad_norm": 4.84375,
      "learning_rate": 3.704385499506565e-06,
      "loss": 1.54421234,
      "memory(GiB)": 117.38,
      "step": 47640,
      "train_speed(iter/s)": 1.637632
    },
    {
      "acc": 0.65253234,
      "epoch": 1.208650431253171,
      "grad_norm": 5.28125,
      "learning_rate": 3.703372720743702e-06,
      "loss": 1.59565544,
      "memory(GiB)": 117.38,
      "step": 47645,
      "train_speed(iter/s)": 1.63765
    },
    {
      "acc": 0.6506361,
      "epoch": 1.208777270421106,
      "grad_norm": 6.3125,
      "learning_rate": 3.7023599990122966e-06,
      "loss": 1.59613762,
      "memory(GiB)": 117.38,
      "step": 47650,
      "train_speed(iter/s)": 1.637669
    },
    {
      "acc": 0.65268602,
      "epoch": 1.2089041095890412,
      "grad_norm": 5.8125,
      "learning_rate": 3.7013473343568897e-06,
      "loss": 1.5973506,
      "memory(GiB)": 117.38,
      "step": 47655,
      "train_speed(iter/s)": 1.637688
    },
    {
      "acc": 0.66887655,
      "epoch": 1.2090309487569761,
      "grad_norm": 6.15625,
      "learning_rate": 3.700334726822026e-06,
      "loss": 1.55827675,
      "memory(GiB)": 117.38,
      "step": 47660,
      "train_speed(iter/s)": 1.637707
    },
    {
      "acc": 0.67833157,
      "epoch": 1.2091577879249111,
      "grad_norm": 6.15625,
      "learning_rate": 3.6993221764522435e-06,
      "loss": 1.5140213,
      "memory(GiB)": 117.38,
      "step": 47665,
      "train_speed(iter/s)": 1.637725
    },
    {
      "acc": 0.66088901,
      "epoch": 1.2092846270928463,
      "grad_norm": 5.15625,
      "learning_rate": 3.6983096832920806e-06,
      "loss": 1.59403515,
      "memory(GiB)": 117.38,
      "step": 47670,
      "train_speed(iter/s)": 1.637743
    },
    {
      "acc": 0.67114058,
      "epoch": 1.2094114662607813,
      "grad_norm": 5.6875,
      "learning_rate": 3.697297247386066e-06,
      "loss": 1.60326481,
      "memory(GiB)": 117.38,
      "step": 47675,
      "train_speed(iter/s)": 1.637762
    },
    {
      "acc": 0.65843749,
      "epoch": 1.2095383054287163,
      "grad_norm": 6.46875,
      "learning_rate": 3.6962848687787365e-06,
      "loss": 1.68851223,
      "memory(GiB)": 117.38,
      "step": 47680,
      "train_speed(iter/s)": 1.63778
    },
    {
      "acc": 0.65391116,
      "epoch": 1.2096651445966515,
      "grad_norm": 5.1875,
      "learning_rate": 3.6952725475146183e-06,
      "loss": 1.57427654,
      "memory(GiB)": 117.38,
      "step": 47685,
      "train_speed(iter/s)": 1.637796
    },
    {
      "acc": 0.65936494,
      "epoch": 1.2097919837645865,
      "grad_norm": 7.90625,
      "learning_rate": 3.69426028363824e-06,
      "loss": 1.56655807,
      "memory(GiB)": 117.38,
      "step": 47690,
      "train_speed(iter/s)": 1.637813
    },
    {
      "acc": 0.66478758,
      "epoch": 1.2099188229325215,
      "grad_norm": 6.65625,
      "learning_rate": 3.6932480771941237e-06,
      "loss": 1.54837332,
      "memory(GiB)": 117.38,
      "step": 47695,
      "train_speed(iter/s)": 1.637829
    },
    {
      "acc": 0.66540852,
      "epoch": 1.2100456621004567,
      "grad_norm": 6.0625,
      "learning_rate": 3.6922359282267904e-06,
      "loss": 1.55895596,
      "memory(GiB)": 117.38,
      "step": 47700,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.66476021,
      "epoch": 1.2101725012683917,
      "grad_norm": 5.375,
      "learning_rate": 3.6912238367807606e-06,
      "loss": 1.59085398,
      "memory(GiB)": 117.38,
      "step": 47705,
      "train_speed(iter/s)": 1.637865
    },
    {
      "acc": 0.63971148,
      "epoch": 1.2102993404363267,
      "grad_norm": 7.84375,
      "learning_rate": 3.6902118029005507e-06,
      "loss": 1.64715214,
      "memory(GiB)": 117.38,
      "step": 47710,
      "train_speed(iter/s)": 1.637884
    },
    {
      "acc": 0.65537806,
      "epoch": 1.2104261796042617,
      "grad_norm": 5.9375,
      "learning_rate": 3.6891998266306717e-06,
      "loss": 1.62349224,
      "memory(GiB)": 117.38,
      "step": 47715,
      "train_speed(iter/s)": 1.637904
    },
    {
      "acc": 0.65911365,
      "epoch": 1.210553018772197,
      "grad_norm": 5.8125,
      "learning_rate": 3.688187908015636e-06,
      "loss": 1.58713236,
      "memory(GiB)": 117.38,
      "step": 47720,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.66470337,
      "epoch": 1.2106798579401319,
      "grad_norm": 7.15625,
      "learning_rate": 3.6871760470999546e-06,
      "loss": 1.62640915,
      "memory(GiB)": 117.38,
      "step": 47725,
      "train_speed(iter/s)": 1.63794
    },
    {
      "acc": 0.65899296,
      "epoch": 1.2108066971080669,
      "grad_norm": 5.4375,
      "learning_rate": 3.6861642439281325e-06,
      "loss": 1.56760435,
      "memory(GiB)": 117.38,
      "step": 47730,
      "train_speed(iter/s)": 1.637959
    },
    {
      "acc": 0.64711294,
      "epoch": 1.210933536276002,
      "grad_norm": 5.40625,
      "learning_rate": 3.6851524985446707e-06,
      "loss": 1.63981018,
      "memory(GiB)": 117.38,
      "step": 47735,
      "train_speed(iter/s)": 1.637978
    },
    {
      "acc": 0.65686073,
      "epoch": 1.211060375443937,
      "grad_norm": 6.21875,
      "learning_rate": 3.6841408109940737e-06,
      "loss": 1.57695665,
      "memory(GiB)": 117.38,
      "step": 47740,
      "train_speed(iter/s)": 1.637997
    },
    {
      "acc": 0.65039511,
      "epoch": 1.211187214611872,
      "grad_norm": 7.03125,
      "learning_rate": 3.6831291813208377e-06,
      "loss": 1.6109127,
      "memory(GiB)": 117.38,
      "step": 47745,
      "train_speed(iter/s)": 1.638015
    },
    {
      "acc": 0.64319525,
      "epoch": 1.2113140537798073,
      "grad_norm": 6.21875,
      "learning_rate": 3.682117609569462e-06,
      "loss": 1.66742916,
      "memory(GiB)": 117.38,
      "step": 47750,
      "train_speed(iter/s)": 1.638033
    },
    {
      "acc": 0.64441476,
      "epoch": 1.2114408929477423,
      "grad_norm": 5.03125,
      "learning_rate": 3.681106095784436e-06,
      "loss": 1.65064392,
      "memory(GiB)": 117.38,
      "step": 47755,
      "train_speed(iter/s)": 1.638051
    },
    {
      "acc": 0.68024392,
      "epoch": 1.2115677321156773,
      "grad_norm": 4.40625,
      "learning_rate": 3.6800946400102522e-06,
      "loss": 1.47348938,
      "memory(GiB)": 117.38,
      "step": 47760,
      "train_speed(iter/s)": 1.638069
    },
    {
      "acc": 0.65860376,
      "epoch": 1.2116945712836125,
      "grad_norm": 5.8125,
      "learning_rate": 3.6790832422913984e-06,
      "loss": 1.60526657,
      "memory(GiB)": 117.38,
      "step": 47765,
      "train_speed(iter/s)": 1.638087
    },
    {
      "acc": 0.66641121,
      "epoch": 1.2118214104515475,
      "grad_norm": 8.0625,
      "learning_rate": 3.6780719026723632e-06,
      "loss": 1.58406124,
      "memory(GiB)": 117.38,
      "step": 47770,
      "train_speed(iter/s)": 1.638107
    },
    {
      "acc": 0.64353762,
      "epoch": 1.2119482496194824,
      "grad_norm": 6.125,
      "learning_rate": 3.677060621197627e-06,
      "loss": 1.69302654,
      "memory(GiB)": 117.38,
      "step": 47775,
      "train_speed(iter/s)": 1.638126
    },
    {
      "acc": 0.65377769,
      "epoch": 1.2120750887874174,
      "grad_norm": 4.90625,
      "learning_rate": 3.6760493979116696e-06,
      "loss": 1.61885071,
      "memory(GiB)": 117.38,
      "step": 47780,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.66489196,
      "epoch": 1.2122019279553526,
      "grad_norm": 5.1875,
      "learning_rate": 3.6750382328589725e-06,
      "loss": 1.58644619,
      "memory(GiB)": 117.38,
      "step": 47785,
      "train_speed(iter/s)": 1.638163
    },
    {
      "acc": 0.64299464,
      "epoch": 1.2123287671232876,
      "grad_norm": 7.625,
      "learning_rate": 3.67402712608401e-06,
      "loss": 1.65487499,
      "memory(GiB)": 117.38,
      "step": 47790,
      "train_speed(iter/s)": 1.638183
    },
    {
      "acc": 0.66387773,
      "epoch": 1.2124556062912228,
      "grad_norm": 5.9375,
      "learning_rate": 3.673016077631253e-06,
      "loss": 1.62087059,
      "memory(GiB)": 117.38,
      "step": 47795,
      "train_speed(iter/s)": 1.638201
    },
    {
      "acc": 0.65175467,
      "epoch": 1.2125824454591578,
      "grad_norm": 6.1875,
      "learning_rate": 3.672005087545173e-06,
      "loss": 1.59164972,
      "memory(GiB)": 117.38,
      "step": 47800,
      "train_speed(iter/s)": 1.63822
    },
    {
      "acc": 0.66121831,
      "epoch": 1.2127092846270928,
      "grad_norm": 5.0625,
      "learning_rate": 3.6709941558702393e-06,
      "loss": 1.59850273,
      "memory(GiB)": 117.38,
      "step": 47805,
      "train_speed(iter/s)": 1.638239
    },
    {
      "acc": 0.65541821,
      "epoch": 1.2128361237950278,
      "grad_norm": 5.6875,
      "learning_rate": 3.6699832826509174e-06,
      "loss": 1.62301617,
      "memory(GiB)": 117.38,
      "step": 47810,
      "train_speed(iter/s)": 1.638257
    },
    {
      "acc": 0.64708948,
      "epoch": 1.212962962962963,
      "grad_norm": 5.65625,
      "learning_rate": 3.6689724679316665e-06,
      "loss": 1.67039871,
      "memory(GiB)": 117.38,
      "step": 47815,
      "train_speed(iter/s)": 1.638276
    },
    {
      "acc": 0.65100551,
      "epoch": 1.213089802130898,
      "grad_norm": 6.15625,
      "learning_rate": 3.66796171175695e-06,
      "loss": 1.62899761,
      "memory(GiB)": 117.38,
      "step": 47820,
      "train_speed(iter/s)": 1.638295
    },
    {
      "acc": 0.6604032,
      "epoch": 1.213216641298833,
      "grad_norm": 5.5625,
      "learning_rate": 3.666951014171224e-06,
      "loss": 1.58865461,
      "memory(GiB)": 117.38,
      "step": 47825,
      "train_speed(iter/s)": 1.638315
    },
    {
      "acc": 0.6468338,
      "epoch": 1.2133434804667682,
      "grad_norm": 4.90625,
      "learning_rate": 3.6659403752189453e-06,
      "loss": 1.58457661,
      "memory(GiB)": 117.38,
      "step": 47830,
      "train_speed(iter/s)": 1.638334
    },
    {
      "acc": 0.658566,
      "epoch": 1.2134703196347032,
      "grad_norm": 5.5,
      "learning_rate": 3.664929794944565e-06,
      "loss": 1.5639081,
      "memory(GiB)": 117.38,
      "step": 47835,
      "train_speed(iter/s)": 1.638352
    },
    {
      "acc": 0.67037497,
      "epoch": 1.2135971588026382,
      "grad_norm": 5.78125,
      "learning_rate": 3.663919273392532e-06,
      "loss": 1.53871708,
      "memory(GiB)": 117.38,
      "step": 47840,
      "train_speed(iter/s)": 1.638371
    },
    {
      "acc": 0.68368177,
      "epoch": 1.2137239979705734,
      "grad_norm": 4.96875,
      "learning_rate": 3.662908810607294e-06,
      "loss": 1.47151546,
      "memory(GiB)": 117.38,
      "step": 47845,
      "train_speed(iter/s)": 1.638389
    },
    {
      "acc": 0.66479325,
      "epoch": 1.2138508371385084,
      "grad_norm": 5.375,
      "learning_rate": 3.6618984066332986e-06,
      "loss": 1.63972626,
      "memory(GiB)": 117.38,
      "step": 47850,
      "train_speed(iter/s)": 1.638408
    },
    {
      "acc": 0.66394463,
      "epoch": 1.2139776763064434,
      "grad_norm": 6.6875,
      "learning_rate": 3.660888061514984e-06,
      "loss": 1.58739681,
      "memory(GiB)": 117.38,
      "step": 47855,
      "train_speed(iter/s)": 1.638426
    },
    {
      "acc": 0.66627588,
      "epoch": 1.2141045154743786,
      "grad_norm": 6.6875,
      "learning_rate": 3.6598777752967896e-06,
      "loss": 1.62743893,
      "memory(GiB)": 117.38,
      "step": 47860,
      "train_speed(iter/s)": 1.638444
    },
    {
      "acc": 0.66282902,
      "epoch": 1.2142313546423136,
      "grad_norm": 6.4375,
      "learning_rate": 3.658867548023156e-06,
      "loss": 1.61625137,
      "memory(GiB)": 117.38,
      "step": 47865,
      "train_speed(iter/s)": 1.638462
    },
    {
      "acc": 0.6609745,
      "epoch": 1.2143581938102486,
      "grad_norm": 6.9375,
      "learning_rate": 3.657857379738515e-06,
      "loss": 1.62713032,
      "memory(GiB)": 117.38,
      "step": 47870,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.65567636,
      "epoch": 1.2144850329781836,
      "grad_norm": 6.3125,
      "learning_rate": 3.656847270487298e-06,
      "loss": 1.6482605,
      "memory(GiB)": 117.38,
      "step": 47875,
      "train_speed(iter/s)": 1.638499
    },
    {
      "acc": 0.649895,
      "epoch": 1.2146118721461188,
      "grad_norm": 5.40625,
      "learning_rate": 3.655837220313936e-06,
      "loss": 1.58997946,
      "memory(GiB)": 117.38,
      "step": 47880,
      "train_speed(iter/s)": 1.638517
    },
    {
      "acc": 0.64477386,
      "epoch": 1.2147387113140538,
      "grad_norm": 4.3125,
      "learning_rate": 3.654827229262852e-06,
      "loss": 1.64937248,
      "memory(GiB)": 117.38,
      "step": 47885,
      "train_speed(iter/s)": 1.638536
    },
    {
      "acc": 0.66857071,
      "epoch": 1.2148655504819887,
      "grad_norm": 5.8125,
      "learning_rate": 3.653817297378476e-06,
      "loss": 1.53666782,
      "memory(GiB)": 117.38,
      "step": 47890,
      "train_speed(iter/s)": 1.638554
    },
    {
      "acc": 0.66990514,
      "epoch": 1.214992389649924,
      "grad_norm": 6.375,
      "learning_rate": 3.6528074247052225e-06,
      "loss": 1.52699547,
      "memory(GiB)": 117.38,
      "step": 47895,
      "train_speed(iter/s)": 1.638573
    },
    {
      "acc": 0.66492877,
      "epoch": 1.215119228817859,
      "grad_norm": 4.53125,
      "learning_rate": 3.651797611287514e-06,
      "loss": 1.52720156,
      "memory(GiB)": 117.38,
      "step": 47900,
      "train_speed(iter/s)": 1.638592
    },
    {
      "acc": 0.68067183,
      "epoch": 1.215246067985794,
      "grad_norm": 7.40625,
      "learning_rate": 3.6507878571697646e-06,
      "loss": 1.54174871,
      "memory(GiB)": 117.38,
      "step": 47905,
      "train_speed(iter/s)": 1.63861
    },
    {
      "acc": 0.65826068,
      "epoch": 1.2153729071537291,
      "grad_norm": 5.59375,
      "learning_rate": 3.6497781623963915e-06,
      "loss": 1.61141224,
      "memory(GiB)": 117.38,
      "step": 47910,
      "train_speed(iter/s)": 1.638628
    },
    {
      "acc": 0.65827856,
      "epoch": 1.2154997463216641,
      "grad_norm": 5.0625,
      "learning_rate": 3.648768527011802e-06,
      "loss": 1.59890509,
      "memory(GiB)": 117.38,
      "step": 47915,
      "train_speed(iter/s)": 1.638645
    },
    {
      "acc": 0.6614789,
      "epoch": 1.2156265854895991,
      "grad_norm": 5.21875,
      "learning_rate": 3.6477589510604044e-06,
      "loss": 1.6151392,
      "memory(GiB)": 117.38,
      "step": 47920,
      "train_speed(iter/s)": 1.638664
    },
    {
      "acc": 0.64883261,
      "epoch": 1.2157534246575343,
      "grad_norm": 4.9375,
      "learning_rate": 3.646749434586607e-06,
      "loss": 1.60067787,
      "memory(GiB)": 117.38,
      "step": 47925,
      "train_speed(iter/s)": 1.638682
    },
    {
      "acc": 0.66785889,
      "epoch": 1.2158802638254693,
      "grad_norm": 7.75,
      "learning_rate": 3.645739977634811e-06,
      "loss": 1.56778088,
      "memory(GiB)": 117.38,
      "step": 47930,
      "train_speed(iter/s)": 1.638701
    },
    {
      "acc": 0.661728,
      "epoch": 1.2160071029934043,
      "grad_norm": 5.53125,
      "learning_rate": 3.6447305802494177e-06,
      "loss": 1.57314167,
      "memory(GiB)": 117.38,
      "step": 47935,
      "train_speed(iter/s)": 1.638719
    },
    {
      "acc": 0.66300807,
      "epoch": 1.2161339421613393,
      "grad_norm": 6.4375,
      "learning_rate": 3.6437212424748227e-06,
      "loss": 1.64865913,
      "memory(GiB)": 117.38,
      "step": 47940,
      "train_speed(iter/s)": 1.638739
    },
    {
      "acc": 0.65332799,
      "epoch": 1.2162607813292745,
      "grad_norm": 6.9375,
      "learning_rate": 3.642711964355423e-06,
      "loss": 1.62769623,
      "memory(GiB)": 117.38,
      "step": 47945,
      "train_speed(iter/s)": 1.638757
    },
    {
      "acc": 0.64575434,
      "epoch": 1.2163876204972095,
      "grad_norm": 7.25,
      "learning_rate": 3.6417027459356134e-06,
      "loss": 1.63560429,
      "memory(GiB)": 117.38,
      "step": 47950,
      "train_speed(iter/s)": 1.638775
    },
    {
      "acc": 0.63751421,
      "epoch": 1.2165144596651447,
      "grad_norm": 5.71875,
      "learning_rate": 3.640693587259778e-06,
      "loss": 1.7098753,
      "memory(GiB)": 117.38,
      "step": 47955,
      "train_speed(iter/s)": 1.638794
    },
    {
      "acc": 0.64397469,
      "epoch": 1.2166412988330797,
      "grad_norm": 4.78125,
      "learning_rate": 3.6396844883723092e-06,
      "loss": 1.67164021,
      "memory(GiB)": 117.38,
      "step": 47960,
      "train_speed(iter/s)": 1.638812
    },
    {
      "acc": 0.66225581,
      "epoch": 1.2167681380010147,
      "grad_norm": 5.5,
      "learning_rate": 3.6386754493175893e-06,
      "loss": 1.62941971,
      "memory(GiB)": 117.38,
      "step": 47965,
      "train_speed(iter/s)": 1.638831
    },
    {
      "acc": 0.65536351,
      "epoch": 1.2168949771689497,
      "grad_norm": 4.53125,
      "learning_rate": 3.637666470140003e-06,
      "loss": 1.58650637,
      "memory(GiB)": 117.38,
      "step": 47970,
      "train_speed(iter/s)": 1.638849
    },
    {
      "acc": 0.66715546,
      "epoch": 1.2170218163368849,
      "grad_norm": 5.5625,
      "learning_rate": 3.6366575508839265e-06,
      "loss": 1.57593946,
      "memory(GiB)": 117.38,
      "step": 47975,
      "train_speed(iter/s)": 1.638867
    },
    {
      "acc": 0.65075779,
      "epoch": 1.2171486555048199,
      "grad_norm": 6.4375,
      "learning_rate": 3.635648691593737e-06,
      "loss": 1.67411995,
      "memory(GiB)": 117.38,
      "step": 47980,
      "train_speed(iter/s)": 1.638884
    },
    {
      "acc": 0.65510664,
      "epoch": 1.2172754946727549,
      "grad_norm": 5.34375,
      "learning_rate": 3.6346398923138094e-06,
      "loss": 1.61411781,
      "memory(GiB)": 117.38,
      "step": 47985,
      "train_speed(iter/s)": 1.638903
    },
    {
      "acc": 0.65895376,
      "epoch": 1.21740233384069,
      "grad_norm": 5.09375,
      "learning_rate": 3.633631153088517e-06,
      "loss": 1.56976662,
      "memory(GiB)": 117.38,
      "step": 47990,
      "train_speed(iter/s)": 1.638921
    },
    {
      "acc": 0.6507803,
      "epoch": 1.217529173008625,
      "grad_norm": 6.28125,
      "learning_rate": 3.6326224739622255e-06,
      "loss": 1.67968559,
      "memory(GiB)": 117.38,
      "step": 47995,
      "train_speed(iter/s)": 1.638938
    },
    {
      "acc": 0.65087376,
      "epoch": 1.21765601217656,
      "grad_norm": 6.96875,
      "learning_rate": 3.6316138549793024e-06,
      "loss": 1.6004715,
      "memory(GiB)": 117.38,
      "step": 48000,
      "train_speed(iter/s)": 1.638956
    },
    {
      "epoch": 1.21765601217656,
      "eval_acc": 0.6462547274029616,
      "eval_loss": 1.5733972787857056,
      "eval_runtime": 58.7028,
      "eval_samples_per_second": 108.513,
      "eval_steps_per_second": 27.137,
      "step": 48000
    },
    {
      "acc": 0.65838628,
      "epoch": 1.2177828513444953,
      "grad_norm": 5.9375,
      "learning_rate": 3.630605296184111e-06,
      "loss": 1.53696594,
      "memory(GiB)": 117.38,
      "step": 48005,
      "train_speed(iter/s)": 1.635455
    },
    {
      "acc": 0.63744974,
      "epoch": 1.2179096905124303,
      "grad_norm": 5.75,
      "learning_rate": 3.6295967976210146e-06,
      "loss": 1.6900425,
      "memory(GiB)": 117.38,
      "step": 48010,
      "train_speed(iter/s)": 1.635473
    },
    {
      "acc": 0.63860259,
      "epoch": 1.2180365296803652,
      "grad_norm": 5.96875,
      "learning_rate": 3.6285883593343685e-06,
      "loss": 1.67360191,
      "memory(GiB)": 117.38,
      "step": 48015,
      "train_speed(iter/s)": 1.635489
    },
    {
      "acc": 0.65640388,
      "epoch": 1.2181633688483005,
      "grad_norm": 6.28125,
      "learning_rate": 3.6275799813685274e-06,
      "loss": 1.60907784,
      "memory(GiB)": 117.38,
      "step": 48020,
      "train_speed(iter/s)": 1.635507
    },
    {
      "acc": 0.65977535,
      "epoch": 1.2182902080162354,
      "grad_norm": 5.5625,
      "learning_rate": 3.6265716637678484e-06,
      "loss": 1.6038435,
      "memory(GiB)": 117.38,
      "step": 48025,
      "train_speed(iter/s)": 1.635525
    },
    {
      "acc": 0.66015639,
      "epoch": 1.2184170471841704,
      "grad_norm": 6.84375,
      "learning_rate": 3.62556340657668e-06,
      "loss": 1.61447563,
      "memory(GiB)": 117.38,
      "step": 48030,
      "train_speed(iter/s)": 1.635542
    },
    {
      "acc": 0.66132317,
      "epoch": 1.2185438863521054,
      "grad_norm": 5.875,
      "learning_rate": 3.6245552098393665e-06,
      "loss": 1.60795059,
      "memory(GiB)": 117.38,
      "step": 48035,
      "train_speed(iter/s)": 1.635559
    },
    {
      "acc": 0.65779886,
      "epoch": 1.2186707255200406,
      "grad_norm": 5.34375,
      "learning_rate": 3.6235470736002576e-06,
      "loss": 1.58638983,
      "memory(GiB)": 117.38,
      "step": 48040,
      "train_speed(iter/s)": 1.635577
    },
    {
      "acc": 0.65934896,
      "epoch": 1.2187975646879756,
      "grad_norm": 5.09375,
      "learning_rate": 3.622538997903693e-06,
      "loss": 1.6043148,
      "memory(GiB)": 117.38,
      "step": 48045,
      "train_speed(iter/s)": 1.635594
    },
    {
      "acc": 0.64994345,
      "epoch": 1.2189244038559106,
      "grad_norm": 6.65625,
      "learning_rate": 3.621530982794015e-06,
      "loss": 1.63655586,
      "memory(GiB)": 117.38,
      "step": 48050,
      "train_speed(iter/s)": 1.635612
    },
    {
      "acc": 0.65441327,
      "epoch": 1.2190512430238458,
      "grad_norm": 6.125,
      "learning_rate": 3.620523028315558e-06,
      "loss": 1.64425621,
      "memory(GiB)": 117.38,
      "step": 48055,
      "train_speed(iter/s)": 1.635629
    },
    {
      "acc": 0.6488965,
      "epoch": 1.2191780821917808,
      "grad_norm": 6.125,
      "learning_rate": 3.6195151345126556e-06,
      "loss": 1.6972702,
      "memory(GiB)": 117.38,
      "step": 48060,
      "train_speed(iter/s)": 1.635646
    },
    {
      "acc": 0.66290216,
      "epoch": 1.2193049213597158,
      "grad_norm": 5.0625,
      "learning_rate": 3.6185073014296425e-06,
      "loss": 1.58812561,
      "memory(GiB)": 117.38,
      "step": 48065,
      "train_speed(iter/s)": 1.635664
    },
    {
      "acc": 0.65696516,
      "epoch": 1.219431760527651,
      "grad_norm": 6.25,
      "learning_rate": 3.6174995291108474e-06,
      "loss": 1.54748907,
      "memory(GiB)": 117.38,
      "step": 48070,
      "train_speed(iter/s)": 1.635683
    },
    {
      "acc": 0.64347086,
      "epoch": 1.219558599695586,
      "grad_norm": 6.15625,
      "learning_rate": 3.6164918176005937e-06,
      "loss": 1.65549469,
      "memory(GiB)": 117.38,
      "step": 48075,
      "train_speed(iter/s)": 1.635699
    },
    {
      "acc": 0.6439455,
      "epoch": 1.219685438863521,
      "grad_norm": 6.03125,
      "learning_rate": 3.6154841669432062e-06,
      "loss": 1.60577316,
      "memory(GiB)": 117.38,
      "step": 48080,
      "train_speed(iter/s)": 1.635715
    },
    {
      "acc": 0.66491609,
      "epoch": 1.2198122780314562,
      "grad_norm": 5.53125,
      "learning_rate": 3.614476577183007e-06,
      "loss": 1.56016846,
      "memory(GiB)": 117.38,
      "step": 48085,
      "train_speed(iter/s)": 1.635729
    },
    {
      "acc": 0.66486816,
      "epoch": 1.2199391171993912,
      "grad_norm": 4.9375,
      "learning_rate": 3.6134690483643154e-06,
      "loss": 1.55486898,
      "memory(GiB)": 117.38,
      "step": 48090,
      "train_speed(iter/s)": 1.635747
    },
    {
      "acc": 0.65270376,
      "epoch": 1.2200659563673262,
      "grad_norm": 6.09375,
      "learning_rate": 3.6124615805314434e-06,
      "loss": 1.62458038,
      "memory(GiB)": 117.38,
      "step": 48095,
      "train_speed(iter/s)": 1.635764
    },
    {
      "acc": 0.644172,
      "epoch": 1.2201927955352612,
      "grad_norm": 5.5,
      "learning_rate": 3.611454173728707e-06,
      "loss": 1.62348328,
      "memory(GiB)": 117.38,
      "step": 48100,
      "train_speed(iter/s)": 1.635782
    },
    {
      "acc": 0.66178865,
      "epoch": 1.2203196347031964,
      "grad_norm": 5.1875,
      "learning_rate": 3.610446828000414e-06,
      "loss": 1.57568903,
      "memory(GiB)": 117.38,
      "step": 48105,
      "train_speed(iter/s)": 1.635799
    },
    {
      "acc": 0.64918442,
      "epoch": 1.2204464738711314,
      "grad_norm": 5.375,
      "learning_rate": 3.609439543390877e-06,
      "loss": 1.60787201,
      "memory(GiB)": 117.38,
      "step": 48110,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.65388942,
      "epoch": 1.2205733130390666,
      "grad_norm": 6.59375,
      "learning_rate": 3.608432319944394e-06,
      "loss": 1.62270794,
      "memory(GiB)": 117.38,
      "step": 48115,
      "train_speed(iter/s)": 1.635835
    },
    {
      "acc": 0.65328693,
      "epoch": 1.2207001522070016,
      "grad_norm": 6.78125,
      "learning_rate": 3.607425157705271e-06,
      "loss": 1.56901016,
      "memory(GiB)": 117.38,
      "step": 48120,
      "train_speed(iter/s)": 1.635852
    },
    {
      "acc": 0.65149579,
      "epoch": 1.2208269913749366,
      "grad_norm": 5.03125,
      "learning_rate": 3.6064180567178064e-06,
      "loss": 1.61642399,
      "memory(GiB)": 117.38,
      "step": 48125,
      "train_speed(iter/s)": 1.635872
    },
    {
      "acc": 0.66643205,
      "epoch": 1.2209538305428715,
      "grad_norm": 5.28125,
      "learning_rate": 3.6054110170263002e-06,
      "loss": 1.57058754,
      "memory(GiB)": 117.38,
      "step": 48130,
      "train_speed(iter/s)": 1.635889
    },
    {
      "acc": 0.65270786,
      "epoch": 1.2210806697108068,
      "grad_norm": 7.875,
      "learning_rate": 3.6044040386750423e-06,
      "loss": 1.59366932,
      "memory(GiB)": 117.38,
      "step": 48135,
      "train_speed(iter/s)": 1.635909
    },
    {
      "acc": 0.67044859,
      "epoch": 1.2212075088787417,
      "grad_norm": 6.28125,
      "learning_rate": 3.6033971217083242e-06,
      "loss": 1.55101824,
      "memory(GiB)": 117.38,
      "step": 48140,
      "train_speed(iter/s)": 1.635926
    },
    {
      "acc": 0.66404285,
      "epoch": 1.2213343480466767,
      "grad_norm": 5.0625,
      "learning_rate": 3.602390266170438e-06,
      "loss": 1.55217323,
      "memory(GiB)": 117.38,
      "step": 48145,
      "train_speed(iter/s)": 1.635945
    },
    {
      "acc": 0.65991001,
      "epoch": 1.221461187214612,
      "grad_norm": 6.96875,
      "learning_rate": 3.6013834721056683e-06,
      "loss": 1.5990694,
      "memory(GiB)": 117.38,
      "step": 48150,
      "train_speed(iter/s)": 1.635963
    },
    {
      "acc": 0.66775904,
      "epoch": 1.221588026382547,
      "grad_norm": 6.21875,
      "learning_rate": 3.6003767395582967e-06,
      "loss": 1.45775194,
      "memory(GiB)": 117.38,
      "step": 48155,
      "train_speed(iter/s)": 1.635981
    },
    {
      "acc": 0.66359086,
      "epoch": 1.221714865550482,
      "grad_norm": 6.5,
      "learning_rate": 3.599370068572604e-06,
      "loss": 1.5262516,
      "memory(GiB)": 117.38,
      "step": 48160,
      "train_speed(iter/s)": 1.636
    },
    {
      "acc": 0.66000738,
      "epoch": 1.2218417047184171,
      "grad_norm": 5.9375,
      "learning_rate": 3.5983634591928705e-06,
      "loss": 1.58534451,
      "memory(GiB)": 117.38,
      "step": 48165,
      "train_speed(iter/s)": 1.636018
    },
    {
      "acc": 0.66850915,
      "epoch": 1.2219685438863521,
      "grad_norm": 5.625,
      "learning_rate": 3.5973569114633704e-06,
      "loss": 1.59331226,
      "memory(GiB)": 117.38,
      "step": 48170,
      "train_speed(iter/s)": 1.636036
    },
    {
      "acc": 0.66196666,
      "epoch": 1.222095383054287,
      "grad_norm": 6.15625,
      "learning_rate": 3.5963504254283743e-06,
      "loss": 1.63140011,
      "memory(GiB)": 117.38,
      "step": 48175,
      "train_speed(iter/s)": 1.636054
    },
    {
      "acc": 0.65586276,
      "epoch": 1.2222222222222223,
      "grad_norm": 5.125,
      "learning_rate": 3.595344001132154e-06,
      "loss": 1.54338589,
      "memory(GiB)": 117.38,
      "step": 48180,
      "train_speed(iter/s)": 1.636075
    },
    {
      "acc": 0.65847549,
      "epoch": 1.2223490613901573,
      "grad_norm": 5.0625,
      "learning_rate": 3.5943376386189744e-06,
      "loss": 1.60298615,
      "memory(GiB)": 117.38,
      "step": 48185,
      "train_speed(iter/s)": 1.636093
    },
    {
      "acc": 0.6617394,
      "epoch": 1.2224759005580923,
      "grad_norm": 6.25,
      "learning_rate": 3.5933313379331047e-06,
      "loss": 1.611586,
      "memory(GiB)": 117.38,
      "step": 48190,
      "train_speed(iter/s)": 1.63611
    },
    {
      "acc": 0.64570665,
      "epoch": 1.2226027397260273,
      "grad_norm": 5.28125,
      "learning_rate": 3.5923250991188e-06,
      "loss": 1.61911278,
      "memory(GiB)": 117.38,
      "step": 48195,
      "train_speed(iter/s)": 1.636128
    },
    {
      "acc": 0.64999189,
      "epoch": 1.2227295788939625,
      "grad_norm": 6.1875,
      "learning_rate": 3.591318922220324e-06,
      "loss": 1.6365818,
      "memory(GiB)": 117.38,
      "step": 48200,
      "train_speed(iter/s)": 1.636147
    },
    {
      "acc": 0.65993419,
      "epoch": 1.2228564180618975,
      "grad_norm": 6.6875,
      "learning_rate": 3.5903128072819287e-06,
      "loss": 1.61732864,
      "memory(GiB)": 117.38,
      "step": 48205,
      "train_speed(iter/s)": 1.636166
    },
    {
      "acc": 0.64813704,
      "epoch": 1.2229832572298325,
      "grad_norm": 6.09375,
      "learning_rate": 3.5893067543478733e-06,
      "loss": 1.64195747,
      "memory(GiB)": 117.38,
      "step": 48210,
      "train_speed(iter/s)": 1.636185
    },
    {
      "acc": 0.6413188,
      "epoch": 1.2231100963977677,
      "grad_norm": 5.75,
      "learning_rate": 3.5883007634624033e-06,
      "loss": 1.66897259,
      "memory(GiB)": 117.38,
      "step": 48215,
      "train_speed(iter/s)": 1.636203
    },
    {
      "acc": 0.64631791,
      "epoch": 1.2232369355657027,
      "grad_norm": 5.3125,
      "learning_rate": 3.5872948346697676e-06,
      "loss": 1.57792816,
      "memory(GiB)": 117.38,
      "step": 48220,
      "train_speed(iter/s)": 1.636223
    },
    {
      "acc": 0.68499188,
      "epoch": 1.2233637747336377,
      "grad_norm": 6.09375,
      "learning_rate": 3.5862889680142133e-06,
      "loss": 1.53002768,
      "memory(GiB)": 117.38,
      "step": 48225,
      "train_speed(iter/s)": 1.636242
    },
    {
      "acc": 0.65768089,
      "epoch": 1.2234906139015729,
      "grad_norm": 5.78125,
      "learning_rate": 3.5852831635399833e-06,
      "loss": 1.60810699,
      "memory(GiB)": 117.38,
      "step": 48230,
      "train_speed(iter/s)": 1.63626
    },
    {
      "acc": 0.65890818,
      "epoch": 1.2236174530695079,
      "grad_norm": 6.15625,
      "learning_rate": 3.5842774212913144e-06,
      "loss": 1.621632,
      "memory(GiB)": 117.38,
      "step": 48235,
      "train_speed(iter/s)": 1.636279
    },
    {
      "acc": 0.65425215,
      "epoch": 1.2237442922374429,
      "grad_norm": 5.75,
      "learning_rate": 3.583271741312445e-06,
      "loss": 1.61629219,
      "memory(GiB)": 117.38,
      "step": 48240,
      "train_speed(iter/s)": 1.636299
    },
    {
      "acc": 0.65213137,
      "epoch": 1.223871131405378,
      "grad_norm": 7.5625,
      "learning_rate": 3.58226612364761e-06,
      "loss": 1.59334278,
      "memory(GiB)": 117.38,
      "step": 48245,
      "train_speed(iter/s)": 1.63632
    },
    {
      "acc": 0.66750031,
      "epoch": 1.223997970573313,
      "grad_norm": 5.46875,
      "learning_rate": 3.581260568341042e-06,
      "loss": 1.56542473,
      "memory(GiB)": 117.38,
      "step": 48250,
      "train_speed(iter/s)": 1.636339
    },
    {
      "acc": 0.64325552,
      "epoch": 1.224124809741248,
      "grad_norm": 5.9375,
      "learning_rate": 3.580255075436967e-06,
      "loss": 1.67417564,
      "memory(GiB)": 117.38,
      "step": 48255,
      "train_speed(iter/s)": 1.636358
    },
    {
      "acc": 0.65055532,
      "epoch": 1.224251648909183,
      "grad_norm": 5.09375,
      "learning_rate": 3.5792496449796127e-06,
      "loss": 1.59499588,
      "memory(GiB)": 117.38,
      "step": 48260,
      "train_speed(iter/s)": 1.636376
    },
    {
      "acc": 0.67153797,
      "epoch": 1.2243784880771182,
      "grad_norm": 5.125,
      "learning_rate": 3.578244277013201e-06,
      "loss": 1.55264473,
      "memory(GiB)": 117.38,
      "step": 48265,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.66648693,
      "epoch": 1.2245053272450532,
      "grad_norm": 4.65625,
      "learning_rate": 3.5772389715819568e-06,
      "loss": 1.60946083,
      "memory(GiB)": 117.38,
      "step": 48270,
      "train_speed(iter/s)": 1.636414
    },
    {
      "acc": 0.66712675,
      "epoch": 1.2246321664129884,
      "grad_norm": 5.875,
      "learning_rate": 3.5762337287300925e-06,
      "loss": 1.55492907,
      "memory(GiB)": 117.38,
      "step": 48275,
      "train_speed(iter/s)": 1.636433
    },
    {
      "acc": 0.65120153,
      "epoch": 1.2247590055809234,
      "grad_norm": 5.21875,
      "learning_rate": 3.575228548501825e-06,
      "loss": 1.61043358,
      "memory(GiB)": 117.38,
      "step": 48280,
      "train_speed(iter/s)": 1.636452
    },
    {
      "acc": 0.64543929,
      "epoch": 1.2248858447488584,
      "grad_norm": 5.90625,
      "learning_rate": 3.574223430941368e-06,
      "loss": 1.62257843,
      "memory(GiB)": 117.38,
      "step": 48285,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.64128962,
      "epoch": 1.2250126839167934,
      "grad_norm": 6.875,
      "learning_rate": 3.573218376092932e-06,
      "loss": 1.70697937,
      "memory(GiB)": 117.38,
      "step": 48290,
      "train_speed(iter/s)": 1.636489
    },
    {
      "acc": 0.65640936,
      "epoch": 1.2251395230847286,
      "grad_norm": 6.5625,
      "learning_rate": 3.5722133840007197e-06,
      "loss": 1.61730499,
      "memory(GiB)": 117.38,
      "step": 48295,
      "train_speed(iter/s)": 1.636508
    },
    {
      "acc": 0.6418808,
      "epoch": 1.2252663622526636,
      "grad_norm": 7.1875,
      "learning_rate": 3.5712084547089367e-06,
      "loss": 1.6134119,
      "memory(GiB)": 117.38,
      "step": 48300,
      "train_speed(iter/s)": 1.636527
    },
    {
      "acc": 0.65859146,
      "epoch": 1.2253932014205986,
      "grad_norm": 6.5,
      "learning_rate": 3.5702035882617857e-06,
      "loss": 1.60098457,
      "memory(GiB)": 117.38,
      "step": 48305,
      "train_speed(iter/s)": 1.636544
    },
    {
      "acc": 0.64533696,
      "epoch": 1.2255200405885338,
      "grad_norm": 6.0,
      "learning_rate": 3.5691987847034667e-06,
      "loss": 1.65100422,
      "memory(GiB)": 117.38,
      "step": 48310,
      "train_speed(iter/s)": 1.636562
    },
    {
      "acc": 0.65092273,
      "epoch": 1.2256468797564688,
      "grad_norm": 6.34375,
      "learning_rate": 3.5681940440781705e-06,
      "loss": 1.64381409,
      "memory(GiB)": 117.38,
      "step": 48315,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.67316704,
      "epoch": 1.2257737189244038,
      "grad_norm": 5.65625,
      "learning_rate": 3.5671893664300934e-06,
      "loss": 1.55052252,
      "memory(GiB)": 117.38,
      "step": 48320,
      "train_speed(iter/s)": 1.636597
    },
    {
      "acc": 0.66980958,
      "epoch": 1.225900558092339,
      "grad_norm": 5.1875,
      "learning_rate": 3.5661847518034244e-06,
      "loss": 1.49008713,
      "memory(GiB)": 117.38,
      "step": 48325,
      "train_speed(iter/s)": 1.636615
    },
    {
      "acc": 0.63822718,
      "epoch": 1.226027397260274,
      "grad_norm": 5.6875,
      "learning_rate": 3.5651802002423543e-06,
      "loss": 1.62146816,
      "memory(GiB)": 117.38,
      "step": 48330,
      "train_speed(iter/s)": 1.636632
    },
    {
      "acc": 0.66818008,
      "epoch": 1.226154236428209,
      "grad_norm": 6.34375,
      "learning_rate": 3.5641757117910625e-06,
      "loss": 1.57961998,
      "memory(GiB)": 117.38,
      "step": 48335,
      "train_speed(iter/s)": 1.636649
    },
    {
      "acc": 0.64077725,
      "epoch": 1.2262810755961442,
      "grad_norm": 6.125,
      "learning_rate": 3.563171286493734e-06,
      "loss": 1.67039032,
      "memory(GiB)": 117.38,
      "step": 48340,
      "train_speed(iter/s)": 1.636667
    },
    {
      "acc": 0.63302851,
      "epoch": 1.2264079147640792,
      "grad_norm": 6.0,
      "learning_rate": 3.5621669243945457e-06,
      "loss": 1.63067474,
      "memory(GiB)": 117.38,
      "step": 48345,
      "train_speed(iter/s)": 1.636684
    },
    {
      "acc": 0.65148969,
      "epoch": 1.2265347539320142,
      "grad_norm": 5.15625,
      "learning_rate": 3.5611626255376785e-06,
      "loss": 1.60756302,
      "memory(GiB)": 117.38,
      "step": 48350,
      "train_speed(iter/s)": 1.636703
    },
    {
      "acc": 0.67346864,
      "epoch": 1.2266615930999492,
      "grad_norm": 5.65625,
      "learning_rate": 3.560158389967302e-06,
      "loss": 1.59825792,
      "memory(GiB)": 117.38,
      "step": 48355,
      "train_speed(iter/s)": 1.636722
    },
    {
      "acc": 0.65535822,
      "epoch": 1.2267884322678844,
      "grad_norm": 7.96875,
      "learning_rate": 3.559154217727586e-06,
      "loss": 1.67674046,
      "memory(GiB)": 117.38,
      "step": 48360,
      "train_speed(iter/s)": 1.63674
    },
    {
      "acc": 0.65066137,
      "epoch": 1.2269152714358194,
      "grad_norm": 6.5625,
      "learning_rate": 3.5581501088627026e-06,
      "loss": 1.69373207,
      "memory(GiB)": 117.38,
      "step": 48365,
      "train_speed(iter/s)": 1.636759
    },
    {
      "acc": 0.66161833,
      "epoch": 1.2270421106037543,
      "grad_norm": 5.84375,
      "learning_rate": 3.557146063416815e-06,
      "loss": 1.57575531,
      "memory(GiB)": 117.38,
      "step": 48370,
      "train_speed(iter/s)": 1.636777
    },
    {
      "acc": 0.65113878,
      "epoch": 1.2271689497716896,
      "grad_norm": 5.625,
      "learning_rate": 3.5561420814340843e-06,
      "loss": 1.63597908,
      "memory(GiB)": 117.38,
      "step": 48375,
      "train_speed(iter/s)": 1.636796
    },
    {
      "acc": 0.64098854,
      "epoch": 1.2272957889396245,
      "grad_norm": 6.03125,
      "learning_rate": 3.555138162958671e-06,
      "loss": 1.66200447,
      "memory(GiB)": 117.38,
      "step": 48380,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.66435156,
      "epoch": 1.2274226281075595,
      "grad_norm": 6.34375,
      "learning_rate": 3.5541343080347325e-06,
      "loss": 1.54340277,
      "memory(GiB)": 117.38,
      "step": 48385,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.65611553,
      "epoch": 1.2275494672754947,
      "grad_norm": 5.4375,
      "learning_rate": 3.5531305167064234e-06,
      "loss": 1.60675163,
      "memory(GiB)": 117.38,
      "step": 48390,
      "train_speed(iter/s)": 1.636848
    },
    {
      "acc": 0.66836038,
      "epoch": 1.2276763064434297,
      "grad_norm": 5.6875,
      "learning_rate": 3.5521267890178922e-06,
      "loss": 1.54638166,
      "memory(GiB)": 117.38,
      "step": 48395,
      "train_speed(iter/s)": 1.636866
    },
    {
      "acc": 0.65401192,
      "epoch": 1.2278031456113647,
      "grad_norm": 5.0625,
      "learning_rate": 3.5511231250132905e-06,
      "loss": 1.58119812,
      "memory(GiB)": 117.38,
      "step": 48400,
      "train_speed(iter/s)": 1.636883
    },
    {
      "acc": 0.65706849,
      "epoch": 1.2279299847793,
      "grad_norm": 4.84375,
      "learning_rate": 3.550119524736761e-06,
      "loss": 1.657444,
      "memory(GiB)": 117.38,
      "step": 48405,
      "train_speed(iter/s)": 1.636901
    },
    {
      "acc": 0.64591465,
      "epoch": 1.228056823947235,
      "grad_norm": 6.84375,
      "learning_rate": 3.5491159882324513e-06,
      "loss": 1.6000946,
      "memory(GiB)": 117.38,
      "step": 48410,
      "train_speed(iter/s)": 1.636918
    },
    {
      "acc": 0.65541887,
      "epoch": 1.22818366311517,
      "grad_norm": 5.875,
      "learning_rate": 3.548112515544495e-06,
      "loss": 1.59203568,
      "memory(GiB)": 117.38,
      "step": 48415,
      "train_speed(iter/s)": 1.636935
    },
    {
      "acc": 0.65563235,
      "epoch": 1.228310502283105,
      "grad_norm": 7.28125,
      "learning_rate": 3.547109106717034e-06,
      "loss": 1.58837109,
      "memory(GiB)": 117.38,
      "step": 48420,
      "train_speed(iter/s)": 1.636952
    },
    {
      "acc": 0.65881538,
      "epoch": 1.22843734145104,
      "grad_norm": 5.78125,
      "learning_rate": 3.546105761794199e-06,
      "loss": 1.62506142,
      "memory(GiB)": 117.38,
      "step": 48425,
      "train_speed(iter/s)": 1.63697
    },
    {
      "acc": 0.65422049,
      "epoch": 1.228564180618975,
      "grad_norm": 5.6875,
      "learning_rate": 3.5451024808201268e-06,
      "loss": 1.60268974,
      "memory(GiB)": 117.38,
      "step": 48430,
      "train_speed(iter/s)": 1.636989
    },
    {
      "acc": 0.63837881,
      "epoch": 1.2286910197869103,
      "grad_norm": 5.46875,
      "learning_rate": 3.5440992638389417e-06,
      "loss": 1.66930008,
      "memory(GiB)": 117.38,
      "step": 48435,
      "train_speed(iter/s)": 1.637006
    },
    {
      "acc": 0.66433296,
      "epoch": 1.2288178589548453,
      "grad_norm": 6.625,
      "learning_rate": 3.5430961108947705e-06,
      "loss": 1.59345856,
      "memory(GiB)": 117.38,
      "step": 48440,
      "train_speed(iter/s)": 1.637023
    },
    {
      "acc": 0.64219632,
      "epoch": 1.2289446981227803,
      "grad_norm": 5.15625,
      "learning_rate": 3.5420930220317373e-06,
      "loss": 1.60974751,
      "memory(GiB)": 117.38,
      "step": 48445,
      "train_speed(iter/s)": 1.63704
    },
    {
      "acc": 0.66603765,
      "epoch": 1.2290715372907153,
      "grad_norm": 5.65625,
      "learning_rate": 3.541089997293964e-06,
      "loss": 1.5459074,
      "memory(GiB)": 117.38,
      "step": 48450,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.6596693,
      "epoch": 1.2291983764586505,
      "grad_norm": 6.75,
      "learning_rate": 3.5400870367255635e-06,
      "loss": 1.54205732,
      "memory(GiB)": 117.38,
      "step": 48455,
      "train_speed(iter/s)": 1.637076
    },
    {
      "acc": 0.64692497,
      "epoch": 1.2293252156265855,
      "grad_norm": 6.96875,
      "learning_rate": 3.539084140370654e-06,
      "loss": 1.58313284,
      "memory(GiB)": 117.38,
      "step": 48460,
      "train_speed(iter/s)": 1.637095
    },
    {
      "acc": 0.65525618,
      "epoch": 1.2294520547945205,
      "grad_norm": 5.5625,
      "learning_rate": 3.538081308273347e-06,
      "loss": 1.62422295,
      "memory(GiB)": 117.38,
      "step": 48465,
      "train_speed(iter/s)": 1.637112
    },
    {
      "acc": 0.65754433,
      "epoch": 1.2295788939624557,
      "grad_norm": 6.46875,
      "learning_rate": 3.537078540477752e-06,
      "loss": 1.58634987,
      "memory(GiB)": 117.38,
      "step": 48470,
      "train_speed(iter/s)": 1.63713
    },
    {
      "acc": 0.64122953,
      "epoch": 1.2297057331303907,
      "grad_norm": 5.75,
      "learning_rate": 3.5360758370279722e-06,
      "loss": 1.64821491,
      "memory(GiB)": 117.38,
      "step": 48475,
      "train_speed(iter/s)": 1.637147
    },
    {
      "acc": 0.66488471,
      "epoch": 1.2298325722983257,
      "grad_norm": 5.65625,
      "learning_rate": 3.535073197968114e-06,
      "loss": 1.54251547,
      "memory(GiB)": 117.38,
      "step": 48480,
      "train_speed(iter/s)": 1.637165
    },
    {
      "acc": 0.66227217,
      "epoch": 1.2299594114662609,
      "grad_norm": 5.9375,
      "learning_rate": 3.5340706233422763e-06,
      "loss": 1.6419178,
      "memory(GiB)": 117.38,
      "step": 48485,
      "train_speed(iter/s)": 1.637183
    },
    {
      "acc": 0.65090356,
      "epoch": 1.2300862506341959,
      "grad_norm": 6.1875,
      "learning_rate": 3.5330681131945588e-06,
      "loss": 1.60829468,
      "memory(GiB)": 117.38,
      "step": 48490,
      "train_speed(iter/s)": 1.637201
    },
    {
      "acc": 0.65389252,
      "epoch": 1.2302130898021308,
      "grad_norm": 5.21875,
      "learning_rate": 3.5320656675690546e-06,
      "loss": 1.55919609,
      "memory(GiB)": 117.38,
      "step": 48495,
      "train_speed(iter/s)": 1.637218
    },
    {
      "acc": 0.65006638,
      "epoch": 1.230339928970066,
      "grad_norm": 5.875,
      "learning_rate": 3.531063286509855e-06,
      "loss": 1.56216888,
      "memory(GiB)": 117.38,
      "step": 48500,
      "train_speed(iter/s)": 1.637235
    },
    {
      "acc": 0.66443434,
      "epoch": 1.230466768138001,
      "grad_norm": 6.40625,
      "learning_rate": 3.530060970061051e-06,
      "loss": 1.60653458,
      "memory(GiB)": 117.38,
      "step": 48505,
      "train_speed(iter/s)": 1.637254
    },
    {
      "acc": 0.65085635,
      "epoch": 1.230593607305936,
      "grad_norm": 5.625,
      "learning_rate": 3.52905871826673e-06,
      "loss": 1.65162544,
      "memory(GiB)": 117.38,
      "step": 48510,
      "train_speed(iter/s)": 1.637271
    },
    {
      "acc": 0.65777621,
      "epoch": 1.230720446473871,
      "grad_norm": 5.5625,
      "learning_rate": 3.5280565311709725e-06,
      "loss": 1.56496115,
      "memory(GiB)": 117.38,
      "step": 48515,
      "train_speed(iter/s)": 1.637289
    },
    {
      "acc": 0.64512844,
      "epoch": 1.2308472856418062,
      "grad_norm": 6.53125,
      "learning_rate": 3.5270544088178597e-06,
      "loss": 1.64786587,
      "memory(GiB)": 117.38,
      "step": 48520,
      "train_speed(iter/s)": 1.637307
    },
    {
      "acc": 0.65158119,
      "epoch": 1.2309741248097412,
      "grad_norm": 5.3125,
      "learning_rate": 3.526052351251471e-06,
      "loss": 1.62798271,
      "memory(GiB)": 117.38,
      "step": 48525,
      "train_speed(iter/s)": 1.637326
    },
    {
      "acc": 0.65848885,
      "epoch": 1.2311009639776762,
      "grad_norm": 6.375,
      "learning_rate": 3.5250503585158825e-06,
      "loss": 1.61647625,
      "memory(GiB)": 117.38,
      "step": 48530,
      "train_speed(iter/s)": 1.637343
    },
    {
      "acc": 0.65481181,
      "epoch": 1.2312278031456114,
      "grad_norm": 5.15625,
      "learning_rate": 3.5240484306551615e-06,
      "loss": 1.5835804,
      "memory(GiB)": 117.38,
      "step": 48535,
      "train_speed(iter/s)": 1.637361
    },
    {
      "acc": 0.65276566,
      "epoch": 1.2313546423135464,
      "grad_norm": 6.71875,
      "learning_rate": 3.5230465677133813e-06,
      "loss": 1.65755291,
      "memory(GiB)": 117.38,
      "step": 48540,
      "train_speed(iter/s)": 1.637379
    },
    {
      "acc": 0.64752941,
      "epoch": 1.2314814814814814,
      "grad_norm": 5.84375,
      "learning_rate": 3.5220447697346063e-06,
      "loss": 1.65055313,
      "memory(GiB)": 117.38,
      "step": 48545,
      "train_speed(iter/s)": 1.637397
    },
    {
      "acc": 0.65033035,
      "epoch": 1.2316083206494166,
      "grad_norm": 8.25,
      "learning_rate": 3.521043036762903e-06,
      "loss": 1.67698288,
      "memory(GiB)": 117.38,
      "step": 48550,
      "train_speed(iter/s)": 1.637414
    },
    {
      "acc": 0.65268202,
      "epoch": 1.2317351598173516,
      "grad_norm": 5.125,
      "learning_rate": 3.5200413688423284e-06,
      "loss": 1.59862814,
      "memory(GiB)": 117.38,
      "step": 48555,
      "train_speed(iter/s)": 1.637432
    },
    {
      "acc": 0.6540966,
      "epoch": 1.2318619989852866,
      "grad_norm": 6.8125,
      "learning_rate": 3.519039766016943e-06,
      "loss": 1.6391943,
      "memory(GiB)": 117.38,
      "step": 48560,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.64522371,
      "epoch": 1.2319888381532218,
      "grad_norm": 5.78125,
      "learning_rate": 3.5180382283307983e-06,
      "loss": 1.70728416,
      "memory(GiB)": 117.38,
      "step": 48565,
      "train_speed(iter/s)": 1.637468
    },
    {
      "acc": 0.66855621,
      "epoch": 1.2321156773211568,
      "grad_norm": 4.90625,
      "learning_rate": 3.517036755827952e-06,
      "loss": 1.56782322,
      "memory(GiB)": 117.38,
      "step": 48570,
      "train_speed(iter/s)": 1.637486
    },
    {
      "acc": 0.66778879,
      "epoch": 1.2322425164890918,
      "grad_norm": 5.1875,
      "learning_rate": 3.516035348552449e-06,
      "loss": 1.58797293,
      "memory(GiB)": 117.38,
      "step": 48575,
      "train_speed(iter/s)": 1.637504
    },
    {
      "acc": 0.65921392,
      "epoch": 1.2323693556570268,
      "grad_norm": 5.84375,
      "learning_rate": 3.515034006548335e-06,
      "loss": 1.5883091,
      "memory(GiB)": 117.38,
      "step": 48580,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.66598668,
      "epoch": 1.232496194824962,
      "grad_norm": 6.46875,
      "learning_rate": 3.5140327298596565e-06,
      "loss": 1.5754281,
      "memory(GiB)": 117.38,
      "step": 48585,
      "train_speed(iter/s)": 1.637541
    },
    {
      "acc": 0.65871267,
      "epoch": 1.232623033992897,
      "grad_norm": 4.875,
      "learning_rate": 3.5130315185304547e-06,
      "loss": 1.56855316,
      "memory(GiB)": 117.38,
      "step": 48590,
      "train_speed(iter/s)": 1.637559
    },
    {
      "acc": 0.65208883,
      "epoch": 1.2327498731608322,
      "grad_norm": 5.53125,
      "learning_rate": 3.5120303726047642e-06,
      "loss": 1.61041756,
      "memory(GiB)": 117.38,
      "step": 48595,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.64740219,
      "epoch": 1.2328767123287672,
      "grad_norm": 5.28125,
      "learning_rate": 3.51102929212662e-06,
      "loss": 1.69913807,
      "memory(GiB)": 117.38,
      "step": 48600,
      "train_speed(iter/s)": 1.637595
    },
    {
      "acc": 0.64866199,
      "epoch": 1.2330035514967022,
      "grad_norm": 5.375,
      "learning_rate": 3.5100282771400563e-06,
      "loss": 1.66856041,
      "memory(GiB)": 117.38,
      "step": 48605,
      "train_speed(iter/s)": 1.637614
    },
    {
      "acc": 0.64932766,
      "epoch": 1.2331303906646371,
      "grad_norm": 4.96875,
      "learning_rate": 3.5090273276891023e-06,
      "loss": 1.60012932,
      "memory(GiB)": 117.38,
      "step": 48610,
      "train_speed(iter/s)": 1.637632
    },
    {
      "acc": 0.65619082,
      "epoch": 1.2332572298325724,
      "grad_norm": 5.5625,
      "learning_rate": 3.5080264438177815e-06,
      "loss": 1.58124876,
      "memory(GiB)": 117.38,
      "step": 48615,
      "train_speed(iter/s)": 1.637649
    },
    {
      "acc": 0.64793406,
      "epoch": 1.2333840690005073,
      "grad_norm": 5.0625,
      "learning_rate": 3.50702562557012e-06,
      "loss": 1.67176228,
      "memory(GiB)": 117.38,
      "step": 48620,
      "train_speed(iter/s)": 1.637667
    },
    {
      "acc": 0.65264406,
      "epoch": 1.2335109081684423,
      "grad_norm": 4.6875,
      "learning_rate": 3.506024872990135e-06,
      "loss": 1.61403351,
      "memory(GiB)": 117.38,
      "step": 48625,
      "train_speed(iter/s)": 1.637685
    },
    {
      "acc": 0.65634069,
      "epoch": 1.2336377473363775,
      "grad_norm": 5.875,
      "learning_rate": 3.5050241861218493e-06,
      "loss": 1.56550303,
      "memory(GiB)": 117.38,
      "step": 48630,
      "train_speed(iter/s)": 1.637703
    },
    {
      "acc": 0.65947065,
      "epoch": 1.2337645865043125,
      "grad_norm": 5.65625,
      "learning_rate": 3.5040235650092725e-06,
      "loss": 1.57389193,
      "memory(GiB)": 117.38,
      "step": 48635,
      "train_speed(iter/s)": 1.637721
    },
    {
      "acc": 0.68081012,
      "epoch": 1.2338914256722475,
      "grad_norm": 6.6875,
      "learning_rate": 3.503023009696419e-06,
      "loss": 1.46642466,
      "memory(GiB)": 117.38,
      "step": 48640,
      "train_speed(iter/s)": 1.637739
    },
    {
      "acc": 0.65444117,
      "epoch": 1.2340182648401827,
      "grad_norm": 5.75,
      "learning_rate": 3.5020225202272963e-06,
      "loss": 1.63811436,
      "memory(GiB)": 117.38,
      "step": 48645,
      "train_speed(iter/s)": 1.637757
    },
    {
      "acc": 0.63551197,
      "epoch": 1.2341451040081177,
      "grad_norm": 5.65625,
      "learning_rate": 3.501022096645913e-06,
      "loss": 1.67545319,
      "memory(GiB)": 117.38,
      "step": 48650,
      "train_speed(iter/s)": 1.637774
    },
    {
      "acc": 0.64973135,
      "epoch": 1.2342719431760527,
      "grad_norm": 6.21875,
      "learning_rate": 3.5000217389962685e-06,
      "loss": 1.64544411,
      "memory(GiB)": 117.38,
      "step": 48655,
      "train_speed(iter/s)": 1.637793
    },
    {
      "acc": 0.65190687,
      "epoch": 1.234398782343988,
      "grad_norm": 5.0,
      "learning_rate": 3.499021447322365e-06,
      "loss": 1.66008663,
      "memory(GiB)": 117.38,
      "step": 48660,
      "train_speed(iter/s)": 1.637812
    },
    {
      "acc": 0.65976696,
      "epoch": 1.234525621511923,
      "grad_norm": 5.0,
      "learning_rate": 3.4980212216681997e-06,
      "loss": 1.61526642,
      "memory(GiB)": 117.38,
      "step": 48665,
      "train_speed(iter/s)": 1.637829
    },
    {
      "acc": 0.65877628,
      "epoch": 1.234652460679858,
      "grad_norm": 4.84375,
      "learning_rate": 3.4970210620777687e-06,
      "loss": 1.59925518,
      "memory(GiB)": 117.38,
      "step": 48670,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.67364249,
      "epoch": 1.2347792998477929,
      "grad_norm": 4.84375,
      "learning_rate": 3.496020968595059e-06,
      "loss": 1.60163021,
      "memory(GiB)": 117.38,
      "step": 48675,
      "train_speed(iter/s)": 1.637866
    },
    {
      "acc": 0.67047882,
      "epoch": 1.234906139015728,
      "grad_norm": 4.8125,
      "learning_rate": 3.4950209412640634e-06,
      "loss": 1.54944868,
      "memory(GiB)": 117.38,
      "step": 48680,
      "train_speed(iter/s)": 1.637885
    },
    {
      "acc": 0.66833687,
      "epoch": 1.235032978183663,
      "grad_norm": 5.5625,
      "learning_rate": 3.494020980128766e-06,
      "loss": 1.5363266,
      "memory(GiB)": 117.38,
      "step": 48685,
      "train_speed(iter/s)": 1.637903
    },
    {
      "acc": 0.66244659,
      "epoch": 1.235159817351598,
      "grad_norm": 6.34375,
      "learning_rate": 3.4930210852331505e-06,
      "loss": 1.52262459,
      "memory(GiB)": 117.38,
      "step": 48690,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.66017919,
      "epoch": 1.2352866565195333,
      "grad_norm": 6.90625,
      "learning_rate": 3.4920212566211943e-06,
      "loss": 1.49845982,
      "memory(GiB)": 117.38,
      "step": 48695,
      "train_speed(iter/s)": 1.637942
    },
    {
      "acc": 0.6517664,
      "epoch": 1.2354134956874683,
      "grad_norm": 5.9375,
      "learning_rate": 3.491021494336876e-06,
      "loss": 1.60274467,
      "memory(GiB)": 117.38,
      "step": 48700,
      "train_speed(iter/s)": 1.637961
    },
    {
      "acc": 0.64981036,
      "epoch": 1.2355403348554033,
      "grad_norm": 5.6875,
      "learning_rate": 3.4900217984241692e-06,
      "loss": 1.60644913,
      "memory(GiB)": 117.38,
      "step": 48705,
      "train_speed(iter/s)": 1.637979
    },
    {
      "acc": 0.65412197,
      "epoch": 1.2356671740233385,
      "grad_norm": 7.625,
      "learning_rate": 3.4890221689270466e-06,
      "loss": 1.58664608,
      "memory(GiB)": 117.38,
      "step": 48710,
      "train_speed(iter/s)": 1.637997
    },
    {
      "acc": 0.66361661,
      "epoch": 1.2357940131912735,
      "grad_norm": 5.96875,
      "learning_rate": 3.488022605889475e-06,
      "loss": 1.57928696,
      "memory(GiB)": 117.38,
      "step": 48715,
      "train_speed(iter/s)": 1.638015
    },
    {
      "acc": 0.65932579,
      "epoch": 1.2359208523592085,
      "grad_norm": 7.15625,
      "learning_rate": 3.4870231093554172e-06,
      "loss": 1.54066334,
      "memory(GiB)": 117.38,
      "step": 48720,
      "train_speed(iter/s)": 1.638034
    },
    {
      "acc": 0.6463203,
      "epoch": 1.2360476915271437,
      "grad_norm": 5.46875,
      "learning_rate": 3.4860236793688407e-06,
      "loss": 1.64001732,
      "memory(GiB)": 117.38,
      "step": 48725,
      "train_speed(iter/s)": 1.638052
    },
    {
      "acc": 0.65029235,
      "epoch": 1.2361745306950787,
      "grad_norm": 5.96875,
      "learning_rate": 3.4850243159737024e-06,
      "loss": 1.59087286,
      "memory(GiB)": 117.38,
      "step": 48730,
      "train_speed(iter/s)": 1.638071
    },
    {
      "acc": 0.6542088,
      "epoch": 1.2363013698630136,
      "grad_norm": 5.84375,
      "learning_rate": 3.4840250192139574e-06,
      "loss": 1.64720364,
      "memory(GiB)": 117.38,
      "step": 48735,
      "train_speed(iter/s)": 1.638089
    },
    {
      "acc": 0.66338096,
      "epoch": 1.2364282090309486,
      "grad_norm": 6.34375,
      "learning_rate": 3.4830257891335595e-06,
      "loss": 1.52913914,
      "memory(GiB)": 117.38,
      "step": 48740,
      "train_speed(iter/s)": 1.638108
    },
    {
      "acc": 0.65706387,
      "epoch": 1.2365550481988838,
      "grad_norm": 4.9375,
      "learning_rate": 3.4820266257764613e-06,
      "loss": 1.51138716,
      "memory(GiB)": 117.38,
      "step": 48745,
      "train_speed(iter/s)": 1.638128
    },
    {
      "acc": 0.66781425,
      "epoch": 1.2366818873668188,
      "grad_norm": 4.8125,
      "learning_rate": 3.4810275291866103e-06,
      "loss": 1.55542421,
      "memory(GiB)": 117.38,
      "step": 48750,
      "train_speed(iter/s)": 1.638146
    },
    {
      "acc": 0.65449901,
      "epoch": 1.236808726534754,
      "grad_norm": 5.71875,
      "learning_rate": 3.4800284994079487e-06,
      "loss": 1.5806776,
      "memory(GiB)": 117.38,
      "step": 48755,
      "train_speed(iter/s)": 1.638165
    },
    {
      "acc": 0.64442616,
      "epoch": 1.236935565702689,
      "grad_norm": 5.25,
      "learning_rate": 3.4790295364844207e-06,
      "loss": 1.63501282,
      "memory(GiB)": 117.38,
      "step": 48760,
      "train_speed(iter/s)": 1.638184
    },
    {
      "acc": 0.64733019,
      "epoch": 1.237062404870624,
      "grad_norm": 6.03125,
      "learning_rate": 3.4780306404599628e-06,
      "loss": 1.60623474,
      "memory(GiB)": 117.38,
      "step": 48765,
      "train_speed(iter/s)": 1.638202
    },
    {
      "acc": 0.64679103,
      "epoch": 1.237189244038559,
      "grad_norm": 7.125,
      "learning_rate": 3.4770318113785164e-06,
      "loss": 1.67006912,
      "memory(GiB)": 117.38,
      "step": 48770,
      "train_speed(iter/s)": 1.638221
    },
    {
      "acc": 0.65661383,
      "epoch": 1.2373160832064942,
      "grad_norm": 6.625,
      "learning_rate": 3.4760330492840065e-06,
      "loss": 1.64981728,
      "memory(GiB)": 117.38,
      "step": 48775,
      "train_speed(iter/s)": 1.638239
    },
    {
      "acc": 0.65511599,
      "epoch": 1.2374429223744292,
      "grad_norm": 6.25,
      "learning_rate": 3.4750343542203684e-06,
      "loss": 1.62428207,
      "memory(GiB)": 117.38,
      "step": 48780,
      "train_speed(iter/s)": 1.638258
    },
    {
      "acc": 0.649823,
      "epoch": 1.2375697615423642,
      "grad_norm": 5.65625,
      "learning_rate": 3.474035726231527e-06,
      "loss": 1.64878502,
      "memory(GiB)": 117.38,
      "step": 48785,
      "train_speed(iter/s)": 1.638276
    },
    {
      "acc": 0.66283646,
      "epoch": 1.2376966007102994,
      "grad_norm": 5.625,
      "learning_rate": 3.473037165361409e-06,
      "loss": 1.63088703,
      "memory(GiB)": 117.38,
      "step": 48790,
      "train_speed(iter/s)": 1.638294
    },
    {
      "acc": 0.66233521,
      "epoch": 1.2378234398782344,
      "grad_norm": 6.1875,
      "learning_rate": 3.4720386716539333e-06,
      "loss": 1.5894309,
      "memory(GiB)": 117.38,
      "step": 48795,
      "train_speed(iter/s)": 1.638313
    },
    {
      "acc": 0.65756836,
      "epoch": 1.2379502790461694,
      "grad_norm": 5.21875,
      "learning_rate": 3.471040245153018e-06,
      "loss": 1.59217596,
      "memory(GiB)": 117.38,
      "step": 48800,
      "train_speed(iter/s)": 1.638332
    },
    {
      "acc": 0.65424194,
      "epoch": 1.2380771182141046,
      "grad_norm": 5.21875,
      "learning_rate": 3.4700418859025793e-06,
      "loss": 1.61271763,
      "memory(GiB)": 117.38,
      "step": 48805,
      "train_speed(iter/s)": 1.63835
    },
    {
      "acc": 0.66022105,
      "epoch": 1.2382039573820396,
      "grad_norm": 6.84375,
      "learning_rate": 3.4690435939465307e-06,
      "loss": 1.57620773,
      "memory(GiB)": 117.38,
      "step": 48810,
      "train_speed(iter/s)": 1.638368
    },
    {
      "acc": 0.65015993,
      "epoch": 1.2383307965499746,
      "grad_norm": 4.90625,
      "learning_rate": 3.4680453693287786e-06,
      "loss": 1.56861115,
      "memory(GiB)": 117.38,
      "step": 48815,
      "train_speed(iter/s)": 1.638386
    },
    {
      "acc": 0.6541832,
      "epoch": 1.2384576357179098,
      "grad_norm": 6.6875,
      "learning_rate": 3.4670472120932297e-06,
      "loss": 1.62443256,
      "memory(GiB)": 117.38,
      "step": 48820,
      "train_speed(iter/s)": 1.638402
    },
    {
      "acc": 0.65462232,
      "epoch": 1.2385844748858448,
      "grad_norm": 6.84375,
      "learning_rate": 3.46604912228379e-06,
      "loss": 1.63511925,
      "memory(GiB)": 117.38,
      "step": 48825,
      "train_speed(iter/s)": 1.63842
    },
    {
      "acc": 0.66376519,
      "epoch": 1.2387113140537798,
      "grad_norm": 5.84375,
      "learning_rate": 3.46505109994436e-06,
      "loss": 1.58396301,
      "memory(GiB)": 117.38,
      "step": 48830,
      "train_speed(iter/s)": 1.638438
    },
    {
      "acc": 0.6551754,
      "epoch": 1.2388381532217148,
      "grad_norm": 5.40625,
      "learning_rate": 3.464053145118833e-06,
      "loss": 1.53278046,
      "memory(GiB)": 117.38,
      "step": 48835,
      "train_speed(iter/s)": 1.638454
    },
    {
      "acc": 0.66460633,
      "epoch": 1.23896499238965,
      "grad_norm": 7.59375,
      "learning_rate": 3.4630552578511073e-06,
      "loss": 1.62619667,
      "memory(GiB)": 117.38,
      "step": 48840,
      "train_speed(iter/s)": 1.638472
    },
    {
      "acc": 0.64749804,
      "epoch": 1.239091831557585,
      "grad_norm": 6.125,
      "learning_rate": 3.4620574381850723e-06,
      "loss": 1.61395226,
      "memory(GiB)": 117.38,
      "step": 48845,
      "train_speed(iter/s)": 1.638491
    },
    {
      "acc": 0.65421524,
      "epoch": 1.23921867072552,
      "grad_norm": 5.8125,
      "learning_rate": 3.4610596861646194e-06,
      "loss": 1.63729897,
      "memory(GiB)": 117.38,
      "step": 48850,
      "train_speed(iter/s)": 1.638508
    },
    {
      "acc": 0.66746111,
      "epoch": 1.2393455098934552,
      "grad_norm": 6.125,
      "learning_rate": 3.460062001833632e-06,
      "loss": 1.5572773,
      "memory(GiB)": 117.38,
      "step": 48855,
      "train_speed(iter/s)": 1.638526
    },
    {
      "acc": 0.68085127,
      "epoch": 1.2394723490613901,
      "grad_norm": 6.5,
      "learning_rate": 3.459064385235993e-06,
      "loss": 1.56359329,
      "memory(GiB)": 117.38,
      "step": 48860,
      "train_speed(iter/s)": 1.638542
    },
    {
      "acc": 0.64866028,
      "epoch": 1.2395991882293251,
      "grad_norm": 6.25,
      "learning_rate": 3.45806683641558e-06,
      "loss": 1.61035519,
      "memory(GiB)": 117.38,
      "step": 48865,
      "train_speed(iter/s)": 1.63856
    },
    {
      "acc": 0.66534638,
      "epoch": 1.2397260273972603,
      "grad_norm": 5.21875,
      "learning_rate": 3.457069355416275e-06,
      "loss": 1.50436878,
      "memory(GiB)": 117.38,
      "step": 48870,
      "train_speed(iter/s)": 1.638578
    },
    {
      "acc": 0.65916786,
      "epoch": 1.2398528665651953,
      "grad_norm": 6.03125,
      "learning_rate": 3.456071942281947e-06,
      "loss": 1.62215271,
      "memory(GiB)": 117.38,
      "step": 48875,
      "train_speed(iter/s)": 1.638598
    },
    {
      "acc": 0.64199252,
      "epoch": 1.2399797057331303,
      "grad_norm": 6.09375,
      "learning_rate": 3.455074597056467e-06,
      "loss": 1.66741104,
      "memory(GiB)": 117.38,
      "step": 48880,
      "train_speed(iter/s)": 1.638615
    },
    {
      "acc": 0.66439462,
      "epoch": 1.2401065449010655,
      "grad_norm": 5.90625,
      "learning_rate": 3.454077319783705e-06,
      "loss": 1.50258551,
      "memory(GiB)": 117.38,
      "step": 48885,
      "train_speed(iter/s)": 1.638633
    },
    {
      "acc": 0.64252748,
      "epoch": 1.2402333840690005,
      "grad_norm": 5.34375,
      "learning_rate": 3.4530801105075257e-06,
      "loss": 1.6697279,
      "memory(GiB)": 117.38,
      "step": 48890,
      "train_speed(iter/s)": 1.638651
    },
    {
      "acc": 0.64962044,
      "epoch": 1.2403602232369355,
      "grad_norm": 5.84375,
      "learning_rate": 3.4520829692717874e-06,
      "loss": 1.62925987,
      "memory(GiB)": 117.38,
      "step": 48895,
      "train_speed(iter/s)": 1.638669
    },
    {
      "acc": 0.64629426,
      "epoch": 1.2404870624048705,
      "grad_norm": 5.0625,
      "learning_rate": 3.451085896120352e-06,
      "loss": 1.63187428,
      "memory(GiB)": 117.38,
      "step": 48900,
      "train_speed(iter/s)": 1.638686
    },
    {
      "acc": 0.65156045,
      "epoch": 1.2406139015728057,
      "grad_norm": 5.21875,
      "learning_rate": 3.450088891097074e-06,
      "loss": 1.60220413,
      "memory(GiB)": 117.38,
      "step": 48905,
      "train_speed(iter/s)": 1.638704
    },
    {
      "acc": 0.65528502,
      "epoch": 1.2407407407407407,
      "grad_norm": 5.78125,
      "learning_rate": 3.4490919542458085e-06,
      "loss": 1.59906931,
      "memory(GiB)": 117.38,
      "step": 48910,
      "train_speed(iter/s)": 1.638721
    },
    {
      "acc": 0.63408594,
      "epoch": 1.240867579908676,
      "grad_norm": 5.25,
      "learning_rate": 3.4480950856104002e-06,
      "loss": 1.68478489,
      "memory(GiB)": 117.38,
      "step": 48915,
      "train_speed(iter/s)": 1.63874
    },
    {
      "acc": 0.64956036,
      "epoch": 1.240994419076611,
      "grad_norm": 6.0625,
      "learning_rate": 3.4470982852347e-06,
      "loss": 1.59546766,
      "memory(GiB)": 117.38,
      "step": 48920,
      "train_speed(iter/s)": 1.638758
    },
    {
      "acc": 0.63675089,
      "epoch": 1.2411212582445459,
      "grad_norm": 5.6875,
      "learning_rate": 3.44610155316255e-06,
      "loss": 1.65878162,
      "memory(GiB)": 117.38,
      "step": 48925,
      "train_speed(iter/s)": 1.638775
    },
    {
      "acc": 0.6600543,
      "epoch": 1.2412480974124809,
      "grad_norm": 5.5625,
      "learning_rate": 3.4451048894377925e-06,
      "loss": 1.58657312,
      "memory(GiB)": 117.38,
      "step": 48930,
      "train_speed(iter/s)": 1.638793
    },
    {
      "acc": 0.65964618,
      "epoch": 1.241374936580416,
      "grad_norm": 5.6875,
      "learning_rate": 3.444108294104264e-06,
      "loss": 1.59323673,
      "memory(GiB)": 117.38,
      "step": 48935,
      "train_speed(iter/s)": 1.638811
    },
    {
      "acc": 0.66680927,
      "epoch": 1.241501775748351,
      "grad_norm": 5.0625,
      "learning_rate": 3.443111767205797e-06,
      "loss": 1.60281677,
      "memory(GiB)": 117.38,
      "step": 48940,
      "train_speed(iter/s)": 1.638828
    },
    {
      "acc": 0.64805794,
      "epoch": 1.241628614916286,
      "grad_norm": 6.4375,
      "learning_rate": 3.442115308786227e-06,
      "loss": 1.61931705,
      "memory(GiB)": 117.38,
      "step": 48945,
      "train_speed(iter/s)": 1.638847
    },
    {
      "acc": 0.65316944,
      "epoch": 1.2417554540842213,
      "grad_norm": 6.3125,
      "learning_rate": 3.4411189188893822e-06,
      "loss": 1.61966019,
      "memory(GiB)": 117.38,
      "step": 48950,
      "train_speed(iter/s)": 1.638865
    },
    {
      "acc": 0.65052328,
      "epoch": 1.2418822932521563,
      "grad_norm": 5.15625,
      "learning_rate": 3.4401225975590867e-06,
      "loss": 1.65285664,
      "memory(GiB)": 117.38,
      "step": 48955,
      "train_speed(iter/s)": 1.638883
    },
    {
      "acc": 0.63330669,
      "epoch": 1.2420091324200913,
      "grad_norm": 4.5,
      "learning_rate": 3.439126344839163e-06,
      "loss": 1.61611805,
      "memory(GiB)": 117.38,
      "step": 48960,
      "train_speed(iter/s)": 1.638901
    },
    {
      "acc": 0.65963011,
      "epoch": 1.2421359715880265,
      "grad_norm": 5.96875,
      "learning_rate": 3.438130160773431e-06,
      "loss": 1.59650936,
      "memory(GiB)": 117.38,
      "step": 48965,
      "train_speed(iter/s)": 1.638918
    },
    {
      "acc": 0.65164604,
      "epoch": 1.2422628107559615,
      "grad_norm": 4.78125,
      "learning_rate": 3.43713404540571e-06,
      "loss": 1.65793076,
      "memory(GiB)": 117.38,
      "step": 48970,
      "train_speed(iter/s)": 1.638937
    },
    {
      "acc": 0.65964561,
      "epoch": 1.2423896499238964,
      "grad_norm": 5.125,
      "learning_rate": 3.4361379987798094e-06,
      "loss": 1.64129562,
      "memory(GiB)": 117.38,
      "step": 48975,
      "train_speed(iter/s)": 1.638954
    },
    {
      "acc": 0.65840273,
      "epoch": 1.2425164890918317,
      "grad_norm": 6.3125,
      "learning_rate": 3.435142020939542e-06,
      "loss": 1.568015,
      "memory(GiB)": 117.38,
      "step": 48980,
      "train_speed(iter/s)": 1.638971
    },
    {
      "acc": 0.66496153,
      "epoch": 1.2426433282597666,
      "grad_norm": 6.0,
      "learning_rate": 3.4341461119287144e-06,
      "loss": 1.55102015,
      "memory(GiB)": 117.38,
      "step": 48985,
      "train_speed(iter/s)": 1.638989
    },
    {
      "acc": 0.66440287,
      "epoch": 1.2427701674277016,
      "grad_norm": 7.28125,
      "learning_rate": 3.433150271791135e-06,
      "loss": 1.5986311,
      "memory(GiB)": 117.38,
      "step": 48990,
      "train_speed(iter/s)": 1.639007
    },
    {
      "acc": 0.66295815,
      "epoch": 1.2428970065956366,
      "grad_norm": 6.4375,
      "learning_rate": 3.432154500570599e-06,
      "loss": 1.54974756,
      "memory(GiB)": 117.38,
      "step": 48995,
      "train_speed(iter/s)": 1.639025
    },
    {
      "acc": 0.66154995,
      "epoch": 1.2430238457635718,
      "grad_norm": 4.78125,
      "learning_rate": 3.431158798310909e-06,
      "loss": 1.56376553,
      "memory(GiB)": 117.38,
      "step": 49000,
      "train_speed(iter/s)": 1.639043
    },
    {
      "epoch": 1.2430238457635718,
      "eval_acc": 0.6462434505499555,
      "eval_loss": 1.5734843015670776,
      "eval_runtime": 58.7744,
      "eval_samples_per_second": 108.381,
      "eval_steps_per_second": 27.104,
      "step": 49000
    },
    {
      "acc": 0.64433823,
      "epoch": 1.2431506849315068,
      "grad_norm": 5.6875,
      "learning_rate": 3.4301631650558588e-06,
      "loss": 1.63979473,
      "memory(GiB)": 117.38,
      "step": 49005,
      "train_speed(iter/s)": 1.635609
    },
    {
      "acc": 0.6495863,
      "epoch": 1.2432775240994418,
      "grad_norm": 6.03125,
      "learning_rate": 3.4291676008492424e-06,
      "loss": 1.57677193,
      "memory(GiB)": 117.38,
      "step": 49010,
      "train_speed(iter/s)": 1.635627
    },
    {
      "acc": 0.6675385,
      "epoch": 1.243404363267377,
      "grad_norm": 5.34375,
      "learning_rate": 3.428172105734848e-06,
      "loss": 1.57157698,
      "memory(GiB)": 117.38,
      "step": 49015,
      "train_speed(iter/s)": 1.635644
    },
    {
      "acc": 0.65708847,
      "epoch": 1.243531202435312,
      "grad_norm": 5.5,
      "learning_rate": 3.4271766797564608e-06,
      "loss": 1.62360497,
      "memory(GiB)": 117.38,
      "step": 49020,
      "train_speed(iter/s)": 1.635662
    },
    {
      "acc": 0.65542431,
      "epoch": 1.243658041603247,
      "grad_norm": 7.1875,
      "learning_rate": 3.4261813229578665e-06,
      "loss": 1.61267757,
      "memory(GiB)": 117.38,
      "step": 49025,
      "train_speed(iter/s)": 1.63568
    },
    {
      "acc": 0.66790047,
      "epoch": 1.2437848807711822,
      "grad_norm": 6.15625,
      "learning_rate": 3.425186035382846e-06,
      "loss": 1.55474529,
      "memory(GiB)": 117.38,
      "step": 49030,
      "train_speed(iter/s)": 1.635696
    },
    {
      "acc": 0.65289712,
      "epoch": 1.2439117199391172,
      "grad_norm": 6.25,
      "learning_rate": 3.4241908170751727e-06,
      "loss": 1.5711834,
      "memory(GiB)": 117.38,
      "step": 49035,
      "train_speed(iter/s)": 1.635713
    },
    {
      "acc": 0.65860806,
      "epoch": 1.2440385591070522,
      "grad_norm": 5.65625,
      "learning_rate": 3.4231956680786217e-06,
      "loss": 1.54433584,
      "memory(GiB)": 117.38,
      "step": 49040,
      "train_speed(iter/s)": 1.635731
    },
    {
      "acc": 0.65533757,
      "epoch": 1.2441653982749874,
      "grad_norm": 5.375,
      "learning_rate": 3.422200588436967e-06,
      "loss": 1.62314548,
      "memory(GiB)": 117.38,
      "step": 49045,
      "train_speed(iter/s)": 1.635748
    },
    {
      "acc": 0.64220061,
      "epoch": 1.2442922374429224,
      "grad_norm": 4.59375,
      "learning_rate": 3.4212055781939744e-06,
      "loss": 1.65563011,
      "memory(GiB)": 117.38,
      "step": 49050,
      "train_speed(iter/s)": 1.635765
    },
    {
      "acc": 0.65179672,
      "epoch": 1.2444190766108574,
      "grad_norm": 6.4375,
      "learning_rate": 3.4202106373934085e-06,
      "loss": 1.67295418,
      "memory(GiB)": 117.38,
      "step": 49055,
      "train_speed(iter/s)": 1.635783
    },
    {
      "acc": 0.65387635,
      "epoch": 1.2445459157787924,
      "grad_norm": 6.21875,
      "learning_rate": 3.4192157660790324e-06,
      "loss": 1.64216576,
      "memory(GiB)": 117.38,
      "step": 49060,
      "train_speed(iter/s)": 1.635796
    },
    {
      "acc": 0.64455962,
      "epoch": 1.2446727549467276,
      "grad_norm": 5.875,
      "learning_rate": 3.418220964294604e-06,
      "loss": 1.64652824,
      "memory(GiB)": 117.38,
      "step": 49065,
      "train_speed(iter/s)": 1.635813
    },
    {
      "acc": 0.64922309,
      "epoch": 1.2447995941146626,
      "grad_norm": 5.71875,
      "learning_rate": 3.417226232083881e-06,
      "loss": 1.63553753,
      "memory(GiB)": 117.38,
      "step": 49070,
      "train_speed(iter/s)": 1.635831
    },
    {
      "acc": 0.65225453,
      "epoch": 1.2449264332825978,
      "grad_norm": 6.4375,
      "learning_rate": 3.416231569490615e-06,
      "loss": 1.64579239,
      "memory(GiB)": 117.38,
      "step": 49075,
      "train_speed(iter/s)": 1.635847
    },
    {
      "acc": 0.69131317,
      "epoch": 1.2450532724505328,
      "grad_norm": 6.5625,
      "learning_rate": 3.4152369765585545e-06,
      "loss": 1.42371483,
      "memory(GiB)": 117.38,
      "step": 49080,
      "train_speed(iter/s)": 1.635865
    },
    {
      "acc": 0.6623085,
      "epoch": 1.2451801116184678,
      "grad_norm": 6.03125,
      "learning_rate": 3.4142424533314474e-06,
      "loss": 1.57119064,
      "memory(GiB)": 117.38,
      "step": 49085,
      "train_speed(iter/s)": 1.635883
    },
    {
      "acc": 0.63752794,
      "epoch": 1.2453069507864027,
      "grad_norm": 5.90625,
      "learning_rate": 3.4132479998530383e-06,
      "loss": 1.66836777,
      "memory(GiB)": 117.38,
      "step": 49090,
      "train_speed(iter/s)": 1.635901
    },
    {
      "acc": 0.65183396,
      "epoch": 1.245433789954338,
      "grad_norm": 5.28125,
      "learning_rate": 3.4122536161670656e-06,
      "loss": 1.59478483,
      "memory(GiB)": 117.38,
      "step": 49095,
      "train_speed(iter/s)": 1.635919
    },
    {
      "acc": 0.66704235,
      "epoch": 1.245560629122273,
      "grad_norm": 7.75,
      "learning_rate": 3.411259302317267e-06,
      "loss": 1.54212542,
      "memory(GiB)": 117.38,
      "step": 49100,
      "train_speed(iter/s)": 1.635936
    },
    {
      "acc": 0.65002127,
      "epoch": 1.245687468290208,
      "grad_norm": 5.4375,
      "learning_rate": 3.410265058347378e-06,
      "loss": 1.55483828,
      "memory(GiB)": 117.38,
      "step": 49105,
      "train_speed(iter/s)": 1.635954
    },
    {
      "acc": 0.66530871,
      "epoch": 1.2458143074581431,
      "grad_norm": 7.3125,
      "learning_rate": 3.4092708843011303e-06,
      "loss": 1.57090197,
      "memory(GiB)": 117.38,
      "step": 49110,
      "train_speed(iter/s)": 1.635971
    },
    {
      "acc": 0.64639974,
      "epoch": 1.2459411466260781,
      "grad_norm": 5.71875,
      "learning_rate": 3.4082767802222493e-06,
      "loss": 1.63317451,
      "memory(GiB)": 117.38,
      "step": 49115,
      "train_speed(iter/s)": 1.635989
    },
    {
      "acc": 0.65806675,
      "epoch": 1.2460679857940131,
      "grad_norm": 6.21875,
      "learning_rate": 3.4072827461544635e-06,
      "loss": 1.61141415,
      "memory(GiB)": 117.38,
      "step": 49120,
      "train_speed(iter/s)": 1.636005
    },
    {
      "acc": 0.65344653,
      "epoch": 1.2461948249619483,
      "grad_norm": 5.8125,
      "learning_rate": 3.4062887821414935e-06,
      "loss": 1.57752876,
      "memory(GiB)": 117.38,
      "step": 49125,
      "train_speed(iter/s)": 1.636023
    },
    {
      "acc": 0.65416827,
      "epoch": 1.2463216641298833,
      "grad_norm": 5.40625,
      "learning_rate": 3.4052948882270585e-06,
      "loss": 1.66028385,
      "memory(GiB)": 117.38,
      "step": 49130,
      "train_speed(iter/s)": 1.63604
    },
    {
      "acc": 0.65115395,
      "epoch": 1.2464485032978183,
      "grad_norm": 6.96875,
      "learning_rate": 3.404301064454873e-06,
      "loss": 1.69198875,
      "memory(GiB)": 117.38,
      "step": 49135,
      "train_speed(iter/s)": 1.636057
    },
    {
      "acc": 0.65345488,
      "epoch": 1.2465753424657535,
      "grad_norm": 5.96875,
      "learning_rate": 3.4033073108686515e-06,
      "loss": 1.61722412,
      "memory(GiB)": 117.38,
      "step": 49140,
      "train_speed(iter/s)": 1.636074
    },
    {
      "acc": 0.64142108,
      "epoch": 1.2467021816336885,
      "grad_norm": 5.5,
      "learning_rate": 3.4023136275121026e-06,
      "loss": 1.62568779,
      "memory(GiB)": 117.38,
      "step": 49145,
      "train_speed(iter/s)": 1.636092
    },
    {
      "acc": 0.63762274,
      "epoch": 1.2468290208016235,
      "grad_norm": 6.59375,
      "learning_rate": 3.401320014428935e-06,
      "loss": 1.70274925,
      "memory(GiB)": 117.38,
      "step": 49150,
      "train_speed(iter/s)": 1.636111
    },
    {
      "acc": 0.65560827,
      "epoch": 1.2469558599695585,
      "grad_norm": 5.34375,
      "learning_rate": 3.40032647166285e-06,
      "loss": 1.56986227,
      "memory(GiB)": 117.38,
      "step": 49155,
      "train_speed(iter/s)": 1.636129
    },
    {
      "acc": 0.6660059,
      "epoch": 1.2470826991374937,
      "grad_norm": 5.78125,
      "learning_rate": 3.3993329992575473e-06,
      "loss": 1.56397486,
      "memory(GiB)": 117.38,
      "step": 49160,
      "train_speed(iter/s)": 1.636146
    },
    {
      "acc": 0.64460168,
      "epoch": 1.2472095383054287,
      "grad_norm": 5.6875,
      "learning_rate": 3.3983395972567277e-06,
      "loss": 1.6256506,
      "memory(GiB)": 117.38,
      "step": 49165,
      "train_speed(iter/s)": 1.636165
    },
    {
      "acc": 0.65856991,
      "epoch": 1.2473363774733637,
      "grad_norm": 5.71875,
      "learning_rate": 3.397346265704084e-06,
      "loss": 1.56199818,
      "memory(GiB)": 117.38,
      "step": 49170,
      "train_speed(iter/s)": 1.636182
    },
    {
      "acc": 0.66867533,
      "epoch": 1.2474632166412989,
      "grad_norm": 6.125,
      "learning_rate": 3.396353004643306e-06,
      "loss": 1.54220743,
      "memory(GiB)": 117.38,
      "step": 49175,
      "train_speed(iter/s)": 1.636201
    },
    {
      "acc": 0.64999056,
      "epoch": 1.2475900558092339,
      "grad_norm": 5.15625,
      "learning_rate": 3.3953598141180817e-06,
      "loss": 1.62617893,
      "memory(GiB)": 117.38,
      "step": 49180,
      "train_speed(iter/s)": 1.636219
    },
    {
      "acc": 0.66229234,
      "epoch": 1.2477168949771689,
      "grad_norm": 9.3125,
      "learning_rate": 3.3943666941720978e-06,
      "loss": 1.57770357,
      "memory(GiB)": 117.38,
      "step": 49185,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.65444107,
      "epoch": 1.247843734145104,
      "grad_norm": 5.4375,
      "learning_rate": 3.3933736448490363e-06,
      "loss": 1.58703175,
      "memory(GiB)": 117.38,
      "step": 49190,
      "train_speed(iter/s)": 1.636255
    },
    {
      "acc": 0.65784883,
      "epoch": 1.247970573313039,
      "grad_norm": 7.59375,
      "learning_rate": 3.392380666192573e-06,
      "loss": 1.59426956,
      "memory(GiB)": 117.38,
      "step": 49195,
      "train_speed(iter/s)": 1.636273
    },
    {
      "acc": 0.64774876,
      "epoch": 1.248097412480974,
      "grad_norm": 5.28125,
      "learning_rate": 3.391387758246386e-06,
      "loss": 1.66628914,
      "memory(GiB)": 117.38,
      "step": 49200,
      "train_speed(iter/s)": 1.636291
    },
    {
      "acc": 0.67271008,
      "epoch": 1.2482242516489093,
      "grad_norm": 5.96875,
      "learning_rate": 3.3903949210541477e-06,
      "loss": 1.5133131,
      "memory(GiB)": 117.38,
      "step": 49205,
      "train_speed(iter/s)": 1.636309
    },
    {
      "acc": 0.65915227,
      "epoch": 1.2483510908168443,
      "grad_norm": 4.78125,
      "learning_rate": 3.389402154659529e-06,
      "loss": 1.61975937,
      "memory(GiB)": 117.38,
      "step": 49210,
      "train_speed(iter/s)": 1.636326
    },
    {
      "acc": 0.64514656,
      "epoch": 1.2484779299847792,
      "grad_norm": 6.3125,
      "learning_rate": 3.388409459106192e-06,
      "loss": 1.65869923,
      "memory(GiB)": 117.38,
      "step": 49215,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.6475831,
      "epoch": 1.2486047691527142,
      "grad_norm": 5.34375,
      "learning_rate": 3.3874168344378024e-06,
      "loss": 1.57521687,
      "memory(GiB)": 117.38,
      "step": 49220,
      "train_speed(iter/s)": 1.636358
    },
    {
      "acc": 0.66036663,
      "epoch": 1.2487316083206494,
      "grad_norm": 6.3125,
      "learning_rate": 3.38642428069802e-06,
      "loss": 1.58773098,
      "memory(GiB)": 117.38,
      "step": 49225,
      "train_speed(iter/s)": 1.636376
    },
    {
      "acc": 0.66770535,
      "epoch": 1.2488584474885844,
      "grad_norm": 7.15625,
      "learning_rate": 3.385431797930503e-06,
      "loss": 1.58640594,
      "memory(GiB)": 117.38,
      "step": 49230,
      "train_speed(iter/s)": 1.636394
    },
    {
      "acc": 0.64684696,
      "epoch": 1.2489852866565196,
      "grad_norm": 5.0625,
      "learning_rate": 3.3844393861789036e-06,
      "loss": 1.6708519,
      "memory(GiB)": 117.38,
      "step": 49235,
      "train_speed(iter/s)": 1.636411
    },
    {
      "acc": 0.64961667,
      "epoch": 1.2491121258244546,
      "grad_norm": 6.875,
      "learning_rate": 3.383447045486872e-06,
      "loss": 1.66736317,
      "memory(GiB)": 117.38,
      "step": 49240,
      "train_speed(iter/s)": 1.636428
    },
    {
      "acc": 0.6387187,
      "epoch": 1.2492389649923896,
      "grad_norm": 5.28125,
      "learning_rate": 3.382454775898057e-06,
      "loss": 1.6750761,
      "memory(GiB)": 117.38,
      "step": 49245,
      "train_speed(iter/s)": 1.636447
    },
    {
      "acc": 0.65937762,
      "epoch": 1.2493658041603246,
      "grad_norm": 6.875,
      "learning_rate": 3.381462577456104e-06,
      "loss": 1.60116062,
      "memory(GiB)": 117.38,
      "step": 49250,
      "train_speed(iter/s)": 1.636465
    },
    {
      "acc": 0.66407385,
      "epoch": 1.2494926433282598,
      "grad_norm": 5.96875,
      "learning_rate": 3.3804704502046527e-06,
      "loss": 1.5733098,
      "memory(GiB)": 117.38,
      "step": 49255,
      "train_speed(iter/s)": 1.636482
    },
    {
      "acc": 0.66543508,
      "epoch": 1.2496194824961948,
      "grad_norm": 5.46875,
      "learning_rate": 3.3794783941873406e-06,
      "loss": 1.54474564,
      "memory(GiB)": 117.38,
      "step": 49260,
      "train_speed(iter/s)": 1.6365
    },
    {
      "acc": 0.65033073,
      "epoch": 1.2497463216641298,
      "grad_norm": 7.9375,
      "learning_rate": 3.3784864094478044e-06,
      "loss": 1.64596825,
      "memory(GiB)": 117.38,
      "step": 49265,
      "train_speed(iter/s)": 1.636518
    },
    {
      "acc": 0.65129375,
      "epoch": 1.249873160832065,
      "grad_norm": 5.21875,
      "learning_rate": 3.377494496029677e-06,
      "loss": 1.59719515,
      "memory(GiB)": 117.38,
      "step": 49270,
      "train_speed(iter/s)": 1.636534
    },
    {
      "acc": 0.66075315,
      "epoch": 1.25,
      "grad_norm": 6.0,
      "learning_rate": 3.3765026539765832e-06,
      "loss": 1.59290981,
      "memory(GiB)": 117.38,
      "step": 49275,
      "train_speed(iter/s)": 1.636553
    },
    {
      "acc": 0.65610685,
      "epoch": 1.250126839167935,
      "grad_norm": 6.3125,
      "learning_rate": 3.375510883332152e-06,
      "loss": 1.6449049,
      "memory(GiB)": 117.38,
      "step": 49280,
      "train_speed(iter/s)": 1.636572
    },
    {
      "acc": 0.63608141,
      "epoch": 1.2502536783358702,
      "grad_norm": 5.1875,
      "learning_rate": 3.3745191841400037e-06,
      "loss": 1.6381237,
      "memory(GiB)": 117.38,
      "step": 49285,
      "train_speed(iter/s)": 1.636588
    },
    {
      "acc": 0.64765563,
      "epoch": 1.2503805175038052,
      "grad_norm": 4.8125,
      "learning_rate": 3.373527556443762e-06,
      "loss": 1.65049858,
      "memory(GiB)": 117.38,
      "step": 49290,
      "train_speed(iter/s)": 1.636605
    },
    {
      "acc": 0.65141196,
      "epoch": 1.2505073566717402,
      "grad_norm": 5.34375,
      "learning_rate": 3.372536000287038e-06,
      "loss": 1.60003185,
      "memory(GiB)": 117.38,
      "step": 49295,
      "train_speed(iter/s)": 1.636623
    },
    {
      "acc": 0.66425533,
      "epoch": 1.2506341958396754,
      "grad_norm": 7.53125,
      "learning_rate": 3.3715445157134474e-06,
      "loss": 1.57765913,
      "memory(GiB)": 117.38,
      "step": 49300,
      "train_speed(iter/s)": 1.636641
    },
    {
      "acc": 0.65659471,
      "epoch": 1.2507610350076104,
      "grad_norm": 5.8125,
      "learning_rate": 3.370553102766598e-06,
      "loss": 1.58344316,
      "memory(GiB)": 117.38,
      "step": 49305,
      "train_speed(iter/s)": 1.636658
    },
    {
      "acc": 0.64507413,
      "epoch": 1.2508878741755454,
      "grad_norm": 5.3125,
      "learning_rate": 3.369561761490101e-06,
      "loss": 1.6362751,
      "memory(GiB)": 117.38,
      "step": 49310,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.65808887,
      "epoch": 1.2510147133434804,
      "grad_norm": 7.0,
      "learning_rate": 3.3685704919275553e-06,
      "loss": 1.64250736,
      "memory(GiB)": 117.38,
      "step": 49315,
      "train_speed(iter/s)": 1.636692
    },
    {
      "acc": 0.67082272,
      "epoch": 1.2511415525114156,
      "grad_norm": 6.0625,
      "learning_rate": 3.3675792941225625e-06,
      "loss": 1.50596876,
      "memory(GiB)": 117.38,
      "step": 49320,
      "train_speed(iter/s)": 1.63671
    },
    {
      "acc": 0.66129255,
      "epoch": 1.2512683916793506,
      "grad_norm": 6.25,
      "learning_rate": 3.3665881681187214e-06,
      "loss": 1.62942715,
      "memory(GiB)": 117.38,
      "step": 49325,
      "train_speed(iter/s)": 1.636727
    },
    {
      "acc": 0.66666231,
      "epoch": 1.2513952308472858,
      "grad_norm": 6.5625,
      "learning_rate": 3.3655971139596265e-06,
      "loss": 1.6081974,
      "memory(GiB)": 117.38,
      "step": 49330,
      "train_speed(iter/s)": 1.636744
    },
    {
      "acc": 0.65374498,
      "epoch": 1.2515220700152208,
      "grad_norm": 5.4375,
      "learning_rate": 3.3646061316888655e-06,
      "loss": 1.63329544,
      "memory(GiB)": 117.38,
      "step": 49335,
      "train_speed(iter/s)": 1.636761
    },
    {
      "acc": 0.65066166,
      "epoch": 1.2516489091831557,
      "grad_norm": 6.96875,
      "learning_rate": 3.3636152213500295e-06,
      "loss": 1.66578007,
      "memory(GiB)": 117.38,
      "step": 49340,
      "train_speed(iter/s)": 1.636779
    },
    {
      "acc": 0.64874635,
      "epoch": 1.2517757483510907,
      "grad_norm": 6.21875,
      "learning_rate": 3.362624382986702e-06,
      "loss": 1.63890839,
      "memory(GiB)": 117.38,
      "step": 49345,
      "train_speed(iter/s)": 1.636796
    },
    {
      "acc": 0.66319718,
      "epoch": 1.251902587519026,
      "grad_norm": 5.03125,
      "learning_rate": 3.3616336166424653e-06,
      "loss": 1.60451412,
      "memory(GiB)": 117.38,
      "step": 49350,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.67067146,
      "epoch": 1.252029426686961,
      "grad_norm": 6.03125,
      "learning_rate": 3.360642922360895e-06,
      "loss": 1.50495701,
      "memory(GiB)": 117.38,
      "step": 49355,
      "train_speed(iter/s)": 1.636831
    },
    {
      "acc": 0.64086423,
      "epoch": 1.252156265854896,
      "grad_norm": 5.84375,
      "learning_rate": 3.3596523001855684e-06,
      "loss": 1.71857986,
      "memory(GiB)": 117.38,
      "step": 49360,
      "train_speed(iter/s)": 1.636848
    },
    {
      "acc": 0.67533321,
      "epoch": 1.2522831050228311,
      "grad_norm": 5.40625,
      "learning_rate": 3.358661750160057e-06,
      "loss": 1.50119133,
      "memory(GiB)": 117.38,
      "step": 49365,
      "train_speed(iter/s)": 1.636865
    },
    {
      "acc": 0.65372791,
      "epoch": 1.2524099441907661,
      "grad_norm": 4.8125,
      "learning_rate": 3.3576712723279326e-06,
      "loss": 1.57447681,
      "memory(GiB)": 117.38,
      "step": 49370,
      "train_speed(iter/s)": 1.636882
    },
    {
      "acc": 0.64516697,
      "epoch": 1.2525367833587011,
      "grad_norm": 5.25,
      "learning_rate": 3.3566808667327566e-06,
      "loss": 1.69332848,
      "memory(GiB)": 117.38,
      "step": 49375,
      "train_speed(iter/s)": 1.636901
    },
    {
      "acc": 0.67300053,
      "epoch": 1.252663622526636,
      "grad_norm": 6.0625,
      "learning_rate": 3.355690533418091e-06,
      "loss": 1.58054142,
      "memory(GiB)": 117.38,
      "step": 49380,
      "train_speed(iter/s)": 1.636918
    },
    {
      "acc": 0.64516811,
      "epoch": 1.2527904616945713,
      "grad_norm": 5.125,
      "learning_rate": 3.354700272427499e-06,
      "loss": 1.65030861,
      "memory(GiB)": 117.38,
      "step": 49385,
      "train_speed(iter/s)": 1.636935
    },
    {
      "acc": 0.65779667,
      "epoch": 1.2529173008625063,
      "grad_norm": 6.75,
      "learning_rate": 3.3537100838045356e-06,
      "loss": 1.66063805,
      "memory(GiB)": 117.38,
      "step": 49390,
      "train_speed(iter/s)": 1.636952
    },
    {
      "acc": 0.66360779,
      "epoch": 1.2530441400304415,
      "grad_norm": 6.53125,
      "learning_rate": 3.3527199675927526e-06,
      "loss": 1.63214455,
      "memory(GiB)": 117.38,
      "step": 49395,
      "train_speed(iter/s)": 1.63697
    },
    {
      "acc": 0.64223914,
      "epoch": 1.2531709791983765,
      "grad_norm": 6.21875,
      "learning_rate": 3.3517299238356982e-06,
      "loss": 1.67133522,
      "memory(GiB)": 117.38,
      "step": 49400,
      "train_speed(iter/s)": 1.636989
    },
    {
      "acc": 0.65898533,
      "epoch": 1.2532978183663115,
      "grad_norm": 5.0,
      "learning_rate": 3.3507399525769214e-06,
      "loss": 1.57467442,
      "memory(GiB)": 117.38,
      "step": 49405,
      "train_speed(iter/s)": 1.637006
    },
    {
      "acc": 0.64663792,
      "epoch": 1.2534246575342465,
      "grad_norm": 6.71875,
      "learning_rate": 3.3497500538599664e-06,
      "loss": 1.65558052,
      "memory(GiB)": 117.38,
      "step": 49410,
      "train_speed(iter/s)": 1.637024
    },
    {
      "acc": 0.6522294,
      "epoch": 1.2535514967021817,
      "grad_norm": 5.34375,
      "learning_rate": 3.34876022772837e-06,
      "loss": 1.64799156,
      "memory(GiB)": 117.38,
      "step": 49415,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.65776939,
      "epoch": 1.2536783358701167,
      "grad_norm": 5.53125,
      "learning_rate": 3.347770474225672e-06,
      "loss": 1.67325172,
      "memory(GiB)": 117.38,
      "step": 49420,
      "train_speed(iter/s)": 1.637061
    },
    {
      "acc": 0.65369911,
      "epoch": 1.2538051750380519,
      "grad_norm": 5.90625,
      "learning_rate": 3.3467807933954034e-06,
      "loss": 1.59110832,
      "memory(GiB)": 117.38,
      "step": 49425,
      "train_speed(iter/s)": 1.637079
    },
    {
      "acc": 0.64510088,
      "epoch": 1.2539320142059869,
      "grad_norm": 4.90625,
      "learning_rate": 3.345791185281101e-06,
      "loss": 1.60621529,
      "memory(GiB)": 117.38,
      "step": 49430,
      "train_speed(iter/s)": 1.637098
    },
    {
      "acc": 0.66607342,
      "epoch": 1.2540588533739219,
      "grad_norm": 5.875,
      "learning_rate": 3.3448016499262836e-06,
      "loss": 1.57269688,
      "memory(GiB)": 117.38,
      "step": 49435,
      "train_speed(iter/s)": 1.637116
    },
    {
      "acc": 0.64941864,
      "epoch": 1.2541856925418569,
      "grad_norm": 5.71875,
      "learning_rate": 3.3438121873744812e-06,
      "loss": 1.63756599,
      "memory(GiB)": 117.38,
      "step": 49440,
      "train_speed(iter/s)": 1.637134
    },
    {
      "acc": 0.64895248,
      "epoch": 1.254312531709792,
      "grad_norm": 5.40625,
      "learning_rate": 3.342822797669212e-06,
      "loss": 1.60085907,
      "memory(GiB)": 117.38,
      "step": 49445,
      "train_speed(iter/s)": 1.637152
    },
    {
      "acc": 0.65342312,
      "epoch": 1.254439370877727,
      "grad_norm": 5.90625,
      "learning_rate": 3.3418334808539966e-06,
      "loss": 1.66911201,
      "memory(GiB)": 117.38,
      "step": 49450,
      "train_speed(iter/s)": 1.63717
    },
    {
      "acc": 0.63991094,
      "epoch": 1.254566210045662,
      "grad_norm": 5.125,
      "learning_rate": 3.340844236972347e-06,
      "loss": 1.67086163,
      "memory(GiB)": 117.38,
      "step": 49455,
      "train_speed(iter/s)": 1.637188
    },
    {
      "acc": 0.6574192,
      "epoch": 1.2546930492135973,
      "grad_norm": 6.5,
      "learning_rate": 3.3398550660677748e-06,
      "loss": 1.6120245,
      "memory(GiB)": 117.38,
      "step": 49460,
      "train_speed(iter/s)": 1.637207
    },
    {
      "acc": 0.6440146,
      "epoch": 1.2548198883815322,
      "grad_norm": 7.375,
      "learning_rate": 3.3388659681837898e-06,
      "loss": 1.62507801,
      "memory(GiB)": 117.38,
      "step": 49465,
      "train_speed(iter/s)": 1.637225
    },
    {
      "acc": 0.64015174,
      "epoch": 1.2549467275494672,
      "grad_norm": 7.5625,
      "learning_rate": 3.3378769433638965e-06,
      "loss": 1.6876976,
      "memory(GiB)": 117.38,
      "step": 49470,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.6564045,
      "epoch": 1.2550735667174022,
      "grad_norm": 6.25,
      "learning_rate": 3.336887991651595e-06,
      "loss": 1.6462719,
      "memory(GiB)": 117.38,
      "step": 49475,
      "train_speed(iter/s)": 1.637261
    },
    {
      "acc": 0.65404158,
      "epoch": 1.2552004058853374,
      "grad_norm": 5.1875,
      "learning_rate": 3.3358991130903845e-06,
      "loss": 1.61365509,
      "memory(GiB)": 117.38,
      "step": 49480,
      "train_speed(iter/s)": 1.637279
    },
    {
      "acc": 0.68134108,
      "epoch": 1.2553272450532724,
      "grad_norm": 6.25,
      "learning_rate": 3.334910307723761e-06,
      "loss": 1.51830635,
      "memory(GiB)": 117.38,
      "step": 49485,
      "train_speed(iter/s)": 1.637298
    },
    {
      "acc": 0.66393123,
      "epoch": 1.2554540842212076,
      "grad_norm": 6.75,
      "learning_rate": 3.333921575595218e-06,
      "loss": 1.57889557,
      "memory(GiB)": 117.38,
      "step": 49490,
      "train_speed(iter/s)": 1.637316
    },
    {
      "acc": 0.63359385,
      "epoch": 1.2555809233891426,
      "grad_norm": 5.53125,
      "learning_rate": 3.3329329167482404e-06,
      "loss": 1.65368576,
      "memory(GiB)": 117.38,
      "step": 49495,
      "train_speed(iter/s)": 1.637335
    },
    {
      "acc": 0.67127409,
      "epoch": 1.2557077625570776,
      "grad_norm": 6.46875,
      "learning_rate": 3.331944331226317e-06,
      "loss": 1.58295994,
      "memory(GiB)": 117.38,
      "step": 49500,
      "train_speed(iter/s)": 1.637353
    },
    {
      "acc": 0.67291746,
      "epoch": 1.2558346017250126,
      "grad_norm": 5.03125,
      "learning_rate": 3.330955819072928e-06,
      "loss": 1.58522968,
      "memory(GiB)": 117.38,
      "step": 49505,
      "train_speed(iter/s)": 1.637371
    },
    {
      "acc": 0.68650851,
      "epoch": 1.2559614408929478,
      "grad_norm": 6.15625,
      "learning_rate": 3.329967380331556e-06,
      "loss": 1.51698895,
      "memory(GiB)": 117.38,
      "step": 49510,
      "train_speed(iter/s)": 1.63739
    },
    {
      "acc": 0.64403601,
      "epoch": 1.2560882800608828,
      "grad_norm": 9.25,
      "learning_rate": 3.3289790150456737e-06,
      "loss": 1.64234772,
      "memory(GiB)": 117.38,
      "step": 49515,
      "train_speed(iter/s)": 1.637408
    },
    {
      "acc": 0.64918604,
      "epoch": 1.2562151192288178,
      "grad_norm": 5.34375,
      "learning_rate": 3.327990723258755e-06,
      "loss": 1.61009407,
      "memory(GiB)": 117.38,
      "step": 49520,
      "train_speed(iter/s)": 1.637427
    },
    {
      "acc": 0.65023804,
      "epoch": 1.256341958396753,
      "grad_norm": 5.15625,
      "learning_rate": 3.3270025050142684e-06,
      "loss": 1.59021721,
      "memory(GiB)": 117.38,
      "step": 49525,
      "train_speed(iter/s)": 1.637444
    },
    {
      "acc": 0.65463924,
      "epoch": 1.256468797564688,
      "grad_norm": 4.90625,
      "learning_rate": 3.3260143603556827e-06,
      "loss": 1.58479223,
      "memory(GiB)": 117.38,
      "step": 49530,
      "train_speed(iter/s)": 1.637461
    },
    {
      "acc": 0.64296741,
      "epoch": 1.256595636732623,
      "grad_norm": 5.75,
      "learning_rate": 3.3250262893264583e-06,
      "loss": 1.62325058,
      "memory(GiB)": 117.38,
      "step": 49535,
      "train_speed(iter/s)": 1.637479
    },
    {
      "acc": 0.64920883,
      "epoch": 1.256722475900558,
      "grad_norm": 6.03125,
      "learning_rate": 3.3240382919700555e-06,
      "loss": 1.61260529,
      "memory(GiB)": 117.38,
      "step": 49540,
      "train_speed(iter/s)": 1.637497
    },
    {
      "acc": 0.64705167,
      "epoch": 1.2568493150684932,
      "grad_norm": 5.84375,
      "learning_rate": 3.3230503683299316e-06,
      "loss": 1.59736347,
      "memory(GiB)": 117.38,
      "step": 49545,
      "train_speed(iter/s)": 1.637515
    },
    {
      "acc": 0.65897007,
      "epoch": 1.2569761542364282,
      "grad_norm": 6.40625,
      "learning_rate": 3.3220625184495404e-06,
      "loss": 1.60819931,
      "memory(GiB)": 117.38,
      "step": 49550,
      "train_speed(iter/s)": 1.637534
    },
    {
      "acc": 0.6525991,
      "epoch": 1.2571029934043634,
      "grad_norm": 6.75,
      "learning_rate": 3.3210747423723293e-06,
      "loss": 1.5896122,
      "memory(GiB)": 117.38,
      "step": 49555,
      "train_speed(iter/s)": 1.637553
    },
    {
      "acc": 0.6589148,
      "epoch": 1.2572298325722984,
      "grad_norm": 5.21875,
      "learning_rate": 3.3200870401417486e-06,
      "loss": 1.61852245,
      "memory(GiB)": 117.38,
      "step": 49560,
      "train_speed(iter/s)": 1.637569
    },
    {
      "acc": 0.66295719,
      "epoch": 1.2573566717402334,
      "grad_norm": 5.28125,
      "learning_rate": 3.3190994118012387e-06,
      "loss": 1.63465843,
      "memory(GiB)": 117.38,
      "step": 49565,
      "train_speed(iter/s)": 1.637587
    },
    {
      "acc": 0.65516825,
      "epoch": 1.2574835109081683,
      "grad_norm": 7.3125,
      "learning_rate": 3.318111857394244e-06,
      "loss": 1.64129333,
      "memory(GiB)": 117.38,
      "step": 49570,
      "train_speed(iter/s)": 1.637604
    },
    {
      "acc": 0.66525354,
      "epoch": 1.2576103500761036,
      "grad_norm": 5.8125,
      "learning_rate": 3.3171243769641957e-06,
      "loss": 1.56072397,
      "memory(GiB)": 117.38,
      "step": 49575,
      "train_speed(iter/s)": 1.637622
    },
    {
      "acc": 0.65193467,
      "epoch": 1.2577371892440385,
      "grad_norm": 6.34375,
      "learning_rate": 3.316136970554532e-06,
      "loss": 1.6028141,
      "memory(GiB)": 117.38,
      "step": 49580,
      "train_speed(iter/s)": 1.63764
    },
    {
      "acc": 0.66604323,
      "epoch": 1.2578640284119738,
      "grad_norm": 6.03125,
      "learning_rate": 3.315149638208681e-06,
      "loss": 1.62820702,
      "memory(GiB)": 117.38,
      "step": 49585,
      "train_speed(iter/s)": 1.637657
    },
    {
      "acc": 0.66368127,
      "epoch": 1.2579908675799087,
      "grad_norm": 6.6875,
      "learning_rate": 3.3141623799700738e-06,
      "loss": 1.59735012,
      "memory(GiB)": 117.38,
      "step": 49590,
      "train_speed(iter/s)": 1.637675
    },
    {
      "acc": 0.63849754,
      "epoch": 1.2581177067478437,
      "grad_norm": 6.875,
      "learning_rate": 3.3131751958821313e-06,
      "loss": 1.68269958,
      "memory(GiB)": 117.38,
      "step": 49595,
      "train_speed(iter/s)": 1.637693
    },
    {
      "acc": 0.66122522,
      "epoch": 1.2582445459157787,
      "grad_norm": 6.0,
      "learning_rate": 3.312188085988273e-06,
      "loss": 1.552705,
      "memory(GiB)": 117.38,
      "step": 49600,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.66504526,
      "epoch": 1.258371385083714,
      "grad_norm": 5.25,
      "learning_rate": 3.311201050331919e-06,
      "loss": 1.59725103,
      "memory(GiB)": 117.38,
      "step": 49605,
      "train_speed(iter/s)": 1.637731
    },
    {
      "acc": 0.66653099,
      "epoch": 1.258498224251649,
      "grad_norm": 5.5625,
      "learning_rate": 3.310214088956485e-06,
      "loss": 1.58143606,
      "memory(GiB)": 117.38,
      "step": 49610,
      "train_speed(iter/s)": 1.637749
    },
    {
      "acc": 0.67515736,
      "epoch": 1.258625063419584,
      "grad_norm": 5.78125,
      "learning_rate": 3.3092272019053773e-06,
      "loss": 1.47481222,
      "memory(GiB)": 117.38,
      "step": 49615,
      "train_speed(iter/s)": 1.637766
    },
    {
      "acc": 0.68036356,
      "epoch": 1.2587519025875191,
      "grad_norm": 4.75,
      "learning_rate": 3.308240389222006e-06,
      "loss": 1.49470243,
      "memory(GiB)": 117.38,
      "step": 49620,
      "train_speed(iter/s)": 1.637782
    },
    {
      "acc": 0.6625773,
      "epoch": 1.2588787417554541,
      "grad_norm": 5.65625,
      "learning_rate": 3.3072536509497762e-06,
      "loss": 1.57462616,
      "memory(GiB)": 117.38,
      "step": 49625,
      "train_speed(iter/s)": 1.637801
    },
    {
      "acc": 0.65595551,
      "epoch": 1.259005580923389,
      "grad_norm": 5.46875,
      "learning_rate": 3.306266987132089e-06,
      "loss": 1.58381948,
      "memory(GiB)": 117.38,
      "step": 49630,
      "train_speed(iter/s)": 1.63782
    },
    {
      "acc": 0.66937542,
      "epoch": 1.259132420091324,
      "grad_norm": 6.6875,
      "learning_rate": 3.3052803978123405e-06,
      "loss": 1.53871021,
      "memory(GiB)": 117.38,
      "step": 49635,
      "train_speed(iter/s)": 1.637838
    },
    {
      "acc": 0.66464615,
      "epoch": 1.2592592592592593,
      "grad_norm": 4.625,
      "learning_rate": 3.3042938830339264e-06,
      "loss": 1.55959225,
      "memory(GiB)": 117.38,
      "step": 49640,
      "train_speed(iter/s)": 1.637856
    },
    {
      "acc": 0.65394125,
      "epoch": 1.2593860984271943,
      "grad_norm": 5.625,
      "learning_rate": 3.303307442840238e-06,
      "loss": 1.54883499,
      "memory(GiB)": 117.38,
      "step": 49645,
      "train_speed(iter/s)": 1.637873
    },
    {
      "acc": 0.66544194,
      "epoch": 1.2595129375951295,
      "grad_norm": 5.75,
      "learning_rate": 3.302321077274666e-06,
      "loss": 1.52851467,
      "memory(GiB)": 117.38,
      "step": 49650,
      "train_speed(iter/s)": 1.637892
    },
    {
      "acc": 0.63880501,
      "epoch": 1.2596397767630645,
      "grad_norm": 4.96875,
      "learning_rate": 3.30133478638059e-06,
      "loss": 1.67572861,
      "memory(GiB)": 117.38,
      "step": 49655,
      "train_speed(iter/s)": 1.637911
    },
    {
      "acc": 0.65686507,
      "epoch": 1.2597666159309995,
      "grad_norm": 4.6875,
      "learning_rate": 3.300348570201395e-06,
      "loss": 1.56951628,
      "memory(GiB)": 117.38,
      "step": 49660,
      "train_speed(iter/s)": 1.637927
    },
    {
      "acc": 0.64188352,
      "epoch": 1.2598934550989345,
      "grad_norm": 5.375,
      "learning_rate": 3.299362428780457e-06,
      "loss": 1.62201042,
      "memory(GiB)": 117.38,
      "step": 49665,
      "train_speed(iter/s)": 1.637946
    },
    {
      "acc": 0.66775746,
      "epoch": 1.2600202942668697,
      "grad_norm": 5.78125,
      "learning_rate": 3.298376362161154e-06,
      "loss": 1.44218502,
      "memory(GiB)": 117.38,
      "step": 49670,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.66670451,
      "epoch": 1.2601471334348047,
      "grad_norm": 6.28125,
      "learning_rate": 3.297390370386856e-06,
      "loss": 1.61631966,
      "memory(GiB)": 117.38,
      "step": 49675,
      "train_speed(iter/s)": 1.637982
    },
    {
      "acc": 0.64637809,
      "epoch": 1.2602739726027397,
      "grad_norm": 5.28125,
      "learning_rate": 3.2964044535009288e-06,
      "loss": 1.64414711,
      "memory(GiB)": 117.38,
      "step": 49680,
      "train_speed(iter/s)": 1.637999
    },
    {
      "acc": 0.66050034,
      "epoch": 1.2604008117706749,
      "grad_norm": 5.59375,
      "learning_rate": 3.2954186115467412e-06,
      "loss": 1.57813931,
      "memory(GiB)": 117.38,
      "step": 49685,
      "train_speed(iter/s)": 1.638018
    },
    {
      "acc": 0.64761658,
      "epoch": 1.2605276509386099,
      "grad_norm": 6.34375,
      "learning_rate": 3.2944328445676543e-06,
      "loss": 1.64748592,
      "memory(GiB)": 117.38,
      "step": 49690,
      "train_speed(iter/s)": 1.638036
    },
    {
      "acc": 0.64268999,
      "epoch": 1.2606544901065448,
      "grad_norm": 5.875,
      "learning_rate": 3.2934471526070254e-06,
      "loss": 1.6700901,
      "memory(GiB)": 117.38,
      "step": 49695,
      "train_speed(iter/s)": 1.638054
    },
    {
      "acc": 0.66521635,
      "epoch": 1.2607813292744798,
      "grad_norm": 6.96875,
      "learning_rate": 3.2924615357082078e-06,
      "loss": 1.56490726,
      "memory(GiB)": 117.38,
      "step": 49700,
      "train_speed(iter/s)": 1.638072
    },
    {
      "acc": 0.6459219,
      "epoch": 1.260908168442415,
      "grad_norm": 6.28125,
      "learning_rate": 3.2914759939145574e-06,
      "loss": 1.64286423,
      "memory(GiB)": 117.38,
      "step": 49705,
      "train_speed(iter/s)": 1.638089
    },
    {
      "acc": 0.64992886,
      "epoch": 1.26103500761035,
      "grad_norm": 6.65625,
      "learning_rate": 3.2904905272694214e-06,
      "loss": 1.63684788,
      "memory(GiB)": 117.38,
      "step": 49710,
      "train_speed(iter/s)": 1.638107
    },
    {
      "acc": 0.65978479,
      "epoch": 1.2611618467782852,
      "grad_norm": 5.0625,
      "learning_rate": 3.289505135816142e-06,
      "loss": 1.56385193,
      "memory(GiB)": 117.38,
      "step": 49715,
      "train_speed(iter/s)": 1.638124
    },
    {
      "acc": 0.65311298,
      "epoch": 1.2612886859462202,
      "grad_norm": 5.59375,
      "learning_rate": 3.2885198195980653e-06,
      "loss": 1.62536316,
      "memory(GiB)": 117.38,
      "step": 49720,
      "train_speed(iter/s)": 1.638143
    },
    {
      "acc": 0.65693417,
      "epoch": 1.2614155251141552,
      "grad_norm": 5.8125,
      "learning_rate": 3.287534578658527e-06,
      "loss": 1.6061451,
      "memory(GiB)": 117.38,
      "step": 49725,
      "train_speed(iter/s)": 1.638161
    },
    {
      "acc": 0.64645748,
      "epoch": 1.2615423642820902,
      "grad_norm": 5.71875,
      "learning_rate": 3.2865494130408657e-06,
      "loss": 1.60314655,
      "memory(GiB)": 117.38,
      "step": 49730,
      "train_speed(iter/s)": 1.638179
    },
    {
      "acc": 0.64851303,
      "epoch": 1.2616692034500254,
      "grad_norm": 8.0,
      "learning_rate": 3.2855643227884097e-06,
      "loss": 1.64891071,
      "memory(GiB)": 117.38,
      "step": 49735,
      "train_speed(iter/s)": 1.638198
    },
    {
      "acc": 0.66854925,
      "epoch": 1.2617960426179604,
      "grad_norm": 5.59375,
      "learning_rate": 3.28457930794449e-06,
      "loss": 1.54447021,
      "memory(GiB)": 117.38,
      "step": 49740,
      "train_speed(iter/s)": 1.638215
    },
    {
      "acc": 0.67055426,
      "epoch": 1.2619228817858956,
      "grad_norm": 7.75,
      "learning_rate": 3.283594368552429e-06,
      "loss": 1.55973034,
      "memory(GiB)": 117.38,
      "step": 49745,
      "train_speed(iter/s)": 1.638235
    },
    {
      "acc": 0.65528841,
      "epoch": 1.2620497209538306,
      "grad_norm": 6.1875,
      "learning_rate": 3.282609504655554e-06,
      "loss": 1.62334137,
      "memory(GiB)": 117.38,
      "step": 49750,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.65968323,
      "epoch": 1.2621765601217656,
      "grad_norm": 7.28125,
      "learning_rate": 3.281624716297179e-06,
      "loss": 1.55405102,
      "memory(GiB)": 117.38,
      "step": 49755,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.64967728,
      "epoch": 1.2623033992897006,
      "grad_norm": 7.84375,
      "learning_rate": 3.28064000352062e-06,
      "loss": 1.62208481,
      "memory(GiB)": 117.38,
      "step": 49760,
      "train_speed(iter/s)": 1.63829
    },
    {
      "acc": 0.65230274,
      "epoch": 1.2624302384576358,
      "grad_norm": 5.59375,
      "learning_rate": 3.279655366369191e-06,
      "loss": 1.5856945,
      "memory(GiB)": 117.38,
      "step": 49765,
      "train_speed(iter/s)": 1.638308
    },
    {
      "acc": 0.65068178,
      "epoch": 1.2625570776255708,
      "grad_norm": 5.59375,
      "learning_rate": 3.2786708048862e-06,
      "loss": 1.64711056,
      "memory(GiB)": 117.38,
      "step": 49770,
      "train_speed(iter/s)": 1.638326
    },
    {
      "acc": 0.66608849,
      "epoch": 1.2626839167935058,
      "grad_norm": 7.4375,
      "learning_rate": 3.2776863191149517e-06,
      "loss": 1.58934221,
      "memory(GiB)": 117.38,
      "step": 49775,
      "train_speed(iter/s)": 1.638345
    },
    {
      "acc": 0.65145817,
      "epoch": 1.262810755961441,
      "grad_norm": 6.96875,
      "learning_rate": 3.2767019090987483e-06,
      "loss": 1.64121132,
      "memory(GiB)": 117.38,
      "step": 49780,
      "train_speed(iter/s)": 1.638364
    },
    {
      "acc": 0.64563627,
      "epoch": 1.262937595129376,
      "grad_norm": 4.96875,
      "learning_rate": 3.27571757488089e-06,
      "loss": 1.69266663,
      "memory(GiB)": 117.38,
      "step": 49785,
      "train_speed(iter/s)": 1.638381
    },
    {
      "acc": 0.66047411,
      "epoch": 1.263064434297311,
      "grad_norm": 5.5625,
      "learning_rate": 3.274733316504672e-06,
      "loss": 1.55904579,
      "memory(GiB)": 117.38,
      "step": 49790,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.65889931,
      "epoch": 1.263191273465246,
      "grad_norm": 4.8125,
      "learning_rate": 3.273749134013383e-06,
      "loss": 1.52647629,
      "memory(GiB)": 117.38,
      "step": 49795,
      "train_speed(iter/s)": 1.638418
    },
    {
      "acc": 0.66125517,
      "epoch": 1.2633181126331812,
      "grad_norm": 5.125,
      "learning_rate": 3.2727650274503154e-06,
      "loss": 1.56951008,
      "memory(GiB)": 117.38,
      "step": 49800,
      "train_speed(iter/s)": 1.638436
    },
    {
      "acc": 0.65565701,
      "epoch": 1.2634449518011162,
      "grad_norm": 6.75,
      "learning_rate": 3.2717809968587523e-06,
      "loss": 1.5748394,
      "memory(GiB)": 117.38,
      "step": 49805,
      "train_speed(iter/s)": 1.638455
    },
    {
      "acc": 0.65837555,
      "epoch": 1.2635717909690514,
      "grad_norm": 6.0625,
      "learning_rate": 3.270797042281979e-06,
      "loss": 1.57841663,
      "memory(GiB)": 117.38,
      "step": 49810,
      "train_speed(iter/s)": 1.638474
    },
    {
      "acc": 0.64887285,
      "epoch": 1.2636986301369864,
      "grad_norm": 5.09375,
      "learning_rate": 3.269813163763271e-06,
      "loss": 1.68857403,
      "memory(GiB)": 117.38,
      "step": 49815,
      "train_speed(iter/s)": 1.638492
    },
    {
      "acc": 0.65504847,
      "epoch": 1.2638254693049213,
      "grad_norm": 6.0,
      "learning_rate": 3.268829361345904e-06,
      "loss": 1.6133667,
      "memory(GiB)": 117.38,
      "step": 49820,
      "train_speed(iter/s)": 1.63851
    },
    {
      "acc": 0.66069002,
      "epoch": 1.2639523084728563,
      "grad_norm": 6.1875,
      "learning_rate": 3.2678456350731526e-06,
      "loss": 1.56747561,
      "memory(GiB)": 117.38,
      "step": 49825,
      "train_speed(iter/s)": 1.638529
    },
    {
      "acc": 0.67424536,
      "epoch": 1.2640791476407915,
      "grad_norm": 7.3125,
      "learning_rate": 3.266861984988283e-06,
      "loss": 1.52973499,
      "memory(GiB)": 117.38,
      "step": 49830,
      "train_speed(iter/s)": 1.638547
    },
    {
      "acc": 0.66885996,
      "epoch": 1.2642059868087265,
      "grad_norm": 6.65625,
      "learning_rate": 3.2658784111345614e-06,
      "loss": 1.60544052,
      "memory(GiB)": 117.38,
      "step": 49835,
      "train_speed(iter/s)": 1.638566
    },
    {
      "acc": 0.65317492,
      "epoch": 1.2643328259766615,
      "grad_norm": 5.6875,
      "learning_rate": 3.2648949135552482e-06,
      "loss": 1.5562499,
      "memory(GiB)": 117.38,
      "step": 49840,
      "train_speed(iter/s)": 1.638585
    },
    {
      "acc": 0.65379677,
      "epoch": 1.2644596651445967,
      "grad_norm": 9.0,
      "learning_rate": 3.2639114922936045e-06,
      "loss": 1.64710789,
      "memory(GiB)": 117.38,
      "step": 49845,
      "train_speed(iter/s)": 1.638604
    },
    {
      "acc": 0.66617098,
      "epoch": 1.2645865043125317,
      "grad_norm": 7.75,
      "learning_rate": 3.2629281473928855e-06,
      "loss": 1.52495041,
      "memory(GiB)": 117.38,
      "step": 49850,
      "train_speed(iter/s)": 1.638622
    },
    {
      "acc": 0.65657992,
      "epoch": 1.2647133434804667,
      "grad_norm": 4.875,
      "learning_rate": 3.26194487889634e-06,
      "loss": 1.57321072,
      "memory(GiB)": 117.38,
      "step": 49855,
      "train_speed(iter/s)": 1.638642
    },
    {
      "acc": 0.66751165,
      "epoch": 1.2648401826484017,
      "grad_norm": 6.6875,
      "learning_rate": 3.2609616868472192e-06,
      "loss": 1.52704563,
      "memory(GiB)": 117.38,
      "step": 49860,
      "train_speed(iter/s)": 1.63866
    },
    {
      "acc": 0.65924549,
      "epoch": 1.264967021816337,
      "grad_norm": 7.09375,
      "learning_rate": 3.259978571288767e-06,
      "loss": 1.5563652,
      "memory(GiB)": 117.38,
      "step": 49865,
      "train_speed(iter/s)": 1.638678
    },
    {
      "acc": 0.64887381,
      "epoch": 1.265093860984272,
      "grad_norm": 5.0625,
      "learning_rate": 3.2589955322642293e-06,
      "loss": 1.68247833,
      "memory(GiB)": 117.38,
      "step": 49870,
      "train_speed(iter/s)": 1.638697
    },
    {
      "acc": 0.64818001,
      "epoch": 1.2652207001522071,
      "grad_norm": 4.375,
      "learning_rate": 3.2580125698168376e-06,
      "loss": 1.58861504,
      "memory(GiB)": 117.38,
      "step": 49875,
      "train_speed(iter/s)": 1.638716
    },
    {
      "acc": 0.65233889,
      "epoch": 1.265347539320142,
      "grad_norm": 5.46875,
      "learning_rate": 3.2570296839898314e-06,
      "loss": 1.64512596,
      "memory(GiB)": 117.38,
      "step": 49880,
      "train_speed(iter/s)": 1.638734
    },
    {
      "acc": 0.65906014,
      "epoch": 1.265474378488077,
      "grad_norm": 6.8125,
      "learning_rate": 3.2560468748264405e-06,
      "loss": 1.57112417,
      "memory(GiB)": 117.38,
      "step": 49885,
      "train_speed(iter/s)": 1.638752
    },
    {
      "acc": 0.64581718,
      "epoch": 1.265601217656012,
      "grad_norm": 5.5625,
      "learning_rate": 3.2550641423698965e-06,
      "loss": 1.62922554,
      "memory(GiB)": 117.38,
      "step": 49890,
      "train_speed(iter/s)": 1.638771
    },
    {
      "acc": 0.63296289,
      "epoch": 1.2657280568239473,
      "grad_norm": 5.53125,
      "learning_rate": 3.2540814866634206e-06,
      "loss": 1.68817787,
      "memory(GiB)": 117.38,
      "step": 49895,
      "train_speed(iter/s)": 1.638789
    },
    {
      "acc": 0.65115576,
      "epoch": 1.2658548959918823,
      "grad_norm": 5.5,
      "learning_rate": 3.2530989077502355e-06,
      "loss": 1.57354288,
      "memory(GiB)": 117.38,
      "step": 49900,
      "train_speed(iter/s)": 1.638807
    },
    {
      "acc": 0.66715145,
      "epoch": 1.2659817351598175,
      "grad_norm": 7.09375,
      "learning_rate": 3.252116405673561e-06,
      "loss": 1.52467728,
      "memory(GiB)": 117.38,
      "step": 49905,
      "train_speed(iter/s)": 1.638824
    },
    {
      "acc": 0.66883411,
      "epoch": 1.2661085743277525,
      "grad_norm": 6.3125,
      "learning_rate": 3.2511339804766107e-06,
      "loss": 1.53025894,
      "memory(GiB)": 117.38,
      "step": 49910,
      "train_speed(iter/s)": 1.638842
    },
    {
      "acc": 0.67596464,
      "epoch": 1.2662354134956875,
      "grad_norm": 6.125,
      "learning_rate": 3.250151632202596e-06,
      "loss": 1.56433115,
      "memory(GiB)": 117.38,
      "step": 49915,
      "train_speed(iter/s)": 1.63886
    },
    {
      "acc": 0.66348505,
      "epoch": 1.2663622526636225,
      "grad_norm": 6.75,
      "learning_rate": 3.249169360894724e-06,
      "loss": 1.61058025,
      "memory(GiB)": 117.38,
      "step": 49920,
      "train_speed(iter/s)": 1.638878
    },
    {
      "acc": 0.64913659,
      "epoch": 1.2664890918315577,
      "grad_norm": 8.125,
      "learning_rate": 3.2481871665962006e-06,
      "loss": 1.67259274,
      "memory(GiB)": 117.38,
      "step": 49925,
      "train_speed(iter/s)": 1.638896
    },
    {
      "acc": 0.66692619,
      "epoch": 1.2666159309994927,
      "grad_norm": 5.875,
      "learning_rate": 3.2472050493502282e-06,
      "loss": 1.53537865,
      "memory(GiB)": 117.38,
      "step": 49930,
      "train_speed(iter/s)": 1.638915
    },
    {
      "acc": 0.66921182,
      "epoch": 1.2667427701674276,
      "grad_norm": 6.46875,
      "learning_rate": 3.2462230092000017e-06,
      "loss": 1.5311655,
      "memory(GiB)": 117.38,
      "step": 49935,
      "train_speed(iter/s)": 1.638933
    },
    {
      "acc": 0.66158123,
      "epoch": 1.2668696093353629,
      "grad_norm": 6.25,
      "learning_rate": 3.2452410461887184e-06,
      "loss": 1.54242058,
      "memory(GiB)": 117.38,
      "step": 49940,
      "train_speed(iter/s)": 1.63895
    },
    {
      "acc": 0.66830521,
      "epoch": 1.2669964485032978,
      "grad_norm": 5.59375,
      "learning_rate": 3.244259160359567e-06,
      "loss": 1.54080524,
      "memory(GiB)": 117.38,
      "step": 49945,
      "train_speed(iter/s)": 1.638967
    },
    {
      "acc": 0.65941219,
      "epoch": 1.2671232876712328,
      "grad_norm": 5.53125,
      "learning_rate": 3.2432773517557385e-06,
      "loss": 1.61795483,
      "memory(GiB)": 117.38,
      "step": 49950,
      "train_speed(iter/s)": 1.638985
    },
    {
      "acc": 0.6512094,
      "epoch": 1.2672501268391678,
      "grad_norm": 4.8125,
      "learning_rate": 3.2422956204204147e-06,
      "loss": 1.59664774,
      "memory(GiB)": 117.38,
      "step": 49955,
      "train_speed(iter/s)": 1.639002
    },
    {
      "acc": 0.66292543,
      "epoch": 1.267376966007103,
      "grad_norm": 6.03125,
      "learning_rate": 3.2413139663967763e-06,
      "loss": 1.61979713,
      "memory(GiB)": 117.38,
      "step": 49960,
      "train_speed(iter/s)": 1.63902
    },
    {
      "acc": 0.67166414,
      "epoch": 1.267503805175038,
      "grad_norm": 5.46875,
      "learning_rate": 3.2403323897280013e-06,
      "loss": 1.49602633,
      "memory(GiB)": 117.38,
      "step": 49965,
      "train_speed(iter/s)": 1.639039
    },
    {
      "acc": 0.65016065,
      "epoch": 1.2676306443429732,
      "grad_norm": 6.375,
      "learning_rate": 3.2393508904572663e-06,
      "loss": 1.55993385,
      "memory(GiB)": 117.38,
      "step": 49970,
      "train_speed(iter/s)": 1.639057
    },
    {
      "acc": 0.64739633,
      "epoch": 1.2677574835109082,
      "grad_norm": 4.59375,
      "learning_rate": 3.2383694686277382e-06,
      "loss": 1.63961601,
      "memory(GiB)": 117.38,
      "step": 49975,
      "train_speed(iter/s)": 1.639075
    },
    {
      "acc": 0.65471816,
      "epoch": 1.2678843226788432,
      "grad_norm": 5.09375,
      "learning_rate": 3.2373881242825857e-06,
      "loss": 1.61855984,
      "memory(GiB)": 117.38,
      "step": 49980,
      "train_speed(iter/s)": 1.639092
    },
    {
      "acc": 0.65718489,
      "epoch": 1.2680111618467782,
      "grad_norm": 5.125,
      "learning_rate": 3.236406857464973e-06,
      "loss": 1.56859512,
      "memory(GiB)": 117.38,
      "step": 49985,
      "train_speed(iter/s)": 1.63911
    },
    {
      "acc": 0.65096259,
      "epoch": 1.2681380010147134,
      "grad_norm": 5.28125,
      "learning_rate": 3.235425668218063e-06,
      "loss": 1.61637688,
      "memory(GiB)": 117.38,
      "step": 49990,
      "train_speed(iter/s)": 1.639128
    },
    {
      "acc": 0.6488801,
      "epoch": 1.2682648401826484,
      "grad_norm": 5.15625,
      "learning_rate": 3.234444556585007e-06,
      "loss": 1.60163269,
      "memory(GiB)": 117.38,
      "step": 49995,
      "train_speed(iter/s)": 1.639147
    },
    {
      "acc": 0.65126643,
      "epoch": 1.2683916793505834,
      "grad_norm": 5.25,
      "learning_rate": 3.233463522608964e-06,
      "loss": 1.58340807,
      "memory(GiB)": 117.38,
      "step": 50000,
      "train_speed(iter/s)": 1.639165
    },
    {
      "epoch": 1.2683916793505834,
      "eval_acc": 0.6463165412638846,
      "eval_loss": 1.5733541250228882,
      "eval_runtime": 59.0635,
      "eval_samples_per_second": 107.85,
      "eval_steps_per_second": 26.971,
      "step": 50000
    },
    {
      "acc": 0.67112527,
      "epoch": 1.2685185185185186,
      "grad_norm": 6.75,
      "learning_rate": 3.2324825663330818e-06,
      "loss": 1.59025059,
      "memory(GiB)": 117.38,
      "step": 50005,
      "train_speed(iter/s)": 1.635785
    },
    {
      "acc": 0.66425056,
      "epoch": 1.2686453576864536,
      "grad_norm": 6.75,
      "learning_rate": 3.231501687800509e-06,
      "loss": 1.5514925,
      "memory(GiB)": 117.38,
      "step": 50010,
      "train_speed(iter/s)": 1.635802
    },
    {
      "acc": 0.65746942,
      "epoch": 1.2687721968543886,
      "grad_norm": 5.5625,
      "learning_rate": 3.2305208870543857e-06,
      "loss": 1.58647041,
      "memory(GiB)": 117.38,
      "step": 50015,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.66604352,
      "epoch": 1.2688990360223236,
      "grad_norm": 6.21875,
      "learning_rate": 3.2295401641378544e-06,
      "loss": 1.59931889,
      "memory(GiB)": 117.38,
      "step": 50020,
      "train_speed(iter/s)": 1.635837
    },
    {
      "acc": 0.66053581,
      "epoch": 1.2690258751902588,
      "grad_norm": 6.125,
      "learning_rate": 3.2285595190940513e-06,
      "loss": 1.61431999,
      "memory(GiB)": 117.38,
      "step": 50025,
      "train_speed(iter/s)": 1.635854
    },
    {
      "acc": 0.65378671,
      "epoch": 1.2691527143581938,
      "grad_norm": 6.0625,
      "learning_rate": 3.2275789519661103e-06,
      "loss": 1.69963989,
      "memory(GiB)": 117.38,
      "step": 50030,
      "train_speed(iter/s)": 1.635872
    },
    {
      "acc": 0.65933676,
      "epoch": 1.269279553526129,
      "grad_norm": 5.65625,
      "learning_rate": 3.2265984627971595e-06,
      "loss": 1.63584557,
      "memory(GiB)": 117.38,
      "step": 50035,
      "train_speed(iter/s)": 1.63589
    },
    {
      "acc": 0.64069567,
      "epoch": 1.269406392694064,
      "grad_norm": 5.4375,
      "learning_rate": 3.225618051630326e-06,
      "loss": 1.65519104,
      "memory(GiB)": 117.38,
      "step": 50040,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.66593528,
      "epoch": 1.269533231861999,
      "grad_norm": 5.5,
      "learning_rate": 3.2246377185087325e-06,
      "loss": 1.61460819,
      "memory(GiB)": 117.38,
      "step": 50045,
      "train_speed(iter/s)": 1.635926
    },
    {
      "acc": 0.6475225,
      "epoch": 1.269660071029934,
      "grad_norm": 6.75,
      "learning_rate": 3.2236574634755003e-06,
      "loss": 1.66770687,
      "memory(GiB)": 117.38,
      "step": 50050,
      "train_speed(iter/s)": 1.635944
    },
    {
      "acc": 0.63543253,
      "epoch": 1.2697869101978692,
      "grad_norm": 5.78125,
      "learning_rate": 3.222677286573742e-06,
      "loss": 1.72865925,
      "memory(GiB)": 117.38,
      "step": 50055,
      "train_speed(iter/s)": 1.63596
    },
    {
      "acc": 0.65067654,
      "epoch": 1.2699137493658041,
      "grad_norm": 5.9375,
      "learning_rate": 3.221697187846571e-06,
      "loss": 1.6313343,
      "memory(GiB)": 117.38,
      "step": 50060,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.6621747,
      "epoch": 1.2700405885337394,
      "grad_norm": 7.5,
      "learning_rate": 3.2207171673370984e-06,
      "loss": 1.63395157,
      "memory(GiB)": 117.38,
      "step": 50065,
      "train_speed(iter/s)": 1.635993
    },
    {
      "acc": 0.6543478,
      "epoch": 1.2701674277016743,
      "grad_norm": 5.125,
      "learning_rate": 3.2197372250884295e-06,
      "loss": 1.6200882,
      "memory(GiB)": 117.38,
      "step": 50070,
      "train_speed(iter/s)": 1.636011
    },
    {
      "acc": 0.65862293,
      "epoch": 1.2702942668696093,
      "grad_norm": 6.25,
      "learning_rate": 3.218757361143664e-06,
      "loss": 1.62272625,
      "memory(GiB)": 117.38,
      "step": 50075,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.65577483,
      "epoch": 1.2704211060375443,
      "grad_norm": 6.0625,
      "learning_rate": 3.2177775755459034e-06,
      "loss": 1.62437916,
      "memory(GiB)": 117.38,
      "step": 50080,
      "train_speed(iter/s)": 1.636045
    },
    {
      "acc": 0.64790387,
      "epoch": 1.2705479452054795,
      "grad_norm": 6.0625,
      "learning_rate": 3.216797868338241e-06,
      "loss": 1.62575417,
      "memory(GiB)": 117.38,
      "step": 50085,
      "train_speed(iter/s)": 1.636062
    },
    {
      "acc": 0.65712094,
      "epoch": 1.2706747843734145,
      "grad_norm": 5.4375,
      "learning_rate": 3.215818239563773e-06,
      "loss": 1.60138454,
      "memory(GiB)": 117.38,
      "step": 50090,
      "train_speed(iter/s)": 1.636078
    },
    {
      "acc": 0.66571302,
      "epoch": 1.2708016235413495,
      "grad_norm": 5.125,
      "learning_rate": 3.2148386892655814e-06,
      "loss": 1.53991909,
      "memory(GiB)": 117.38,
      "step": 50095,
      "train_speed(iter/s)": 1.63609
    },
    {
      "acc": 0.65168362,
      "epoch": 1.2709284627092847,
      "grad_norm": 6.4375,
      "learning_rate": 3.2138592174867556e-06,
      "loss": 1.66510296,
      "memory(GiB)": 117.38,
      "step": 50100,
      "train_speed(iter/s)": 1.635843
    },
    {
      "acc": 0.66659975,
      "epoch": 1.2710553018772197,
      "grad_norm": 6.9375,
      "learning_rate": 3.2128798242703745e-06,
      "loss": 1.58371181,
      "memory(GiB)": 117.38,
      "step": 50105,
      "train_speed(iter/s)": 1.635857
    },
    {
      "acc": 0.66018152,
      "epoch": 1.2711821410451547,
      "grad_norm": 5.25,
      "learning_rate": 3.2119005096595203e-06,
      "loss": 1.61099701,
      "memory(GiB)": 117.38,
      "step": 50110,
      "train_speed(iter/s)": 1.635874
    },
    {
      "acc": 0.66646771,
      "epoch": 1.2713089802130897,
      "grad_norm": 4.875,
      "learning_rate": 3.2109212736972636e-06,
      "loss": 1.57505798,
      "memory(GiB)": 117.38,
      "step": 50115,
      "train_speed(iter/s)": 1.635891
    },
    {
      "acc": 0.66070452,
      "epoch": 1.271435819381025,
      "grad_norm": 5.84375,
      "learning_rate": 3.2099421164266758e-06,
      "loss": 1.53111658,
      "memory(GiB)": 117.38,
      "step": 50120,
      "train_speed(iter/s)": 1.635908
    },
    {
      "acc": 0.65686388,
      "epoch": 1.27156265854896,
      "grad_norm": 7.09375,
      "learning_rate": 3.2089630378908264e-06,
      "loss": 1.60889378,
      "memory(GiB)": 117.38,
      "step": 50125,
      "train_speed(iter/s)": 1.635925
    },
    {
      "acc": 0.66921763,
      "epoch": 1.271689497716895,
      "grad_norm": 4.4375,
      "learning_rate": 3.207984038132781e-06,
      "loss": 1.54144535,
      "memory(GiB)": 117.38,
      "step": 50130,
      "train_speed(iter/s)": 1.635944
    },
    {
      "acc": 0.66131573,
      "epoch": 1.27181633688483,
      "grad_norm": 5.6875,
      "learning_rate": 3.2070051171955966e-06,
      "loss": 1.59992342,
      "memory(GiB)": 117.38,
      "step": 50135,
      "train_speed(iter/s)": 1.635961
    },
    {
      "acc": 0.68003874,
      "epoch": 1.271943176052765,
      "grad_norm": 5.0625,
      "learning_rate": 3.206026275122332e-06,
      "loss": 1.49340334,
      "memory(GiB)": 117.38,
      "step": 50140,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.67815485,
      "epoch": 1.2720700152207,
      "grad_norm": 5.8125,
      "learning_rate": 3.205047511956042e-06,
      "loss": 1.51377974,
      "memory(GiB)": 117.38,
      "step": 50145,
      "train_speed(iter/s)": 1.635995
    },
    {
      "acc": 0.6564024,
      "epoch": 1.2721968543886353,
      "grad_norm": 5.6875,
      "learning_rate": 3.204068827739777e-06,
      "loss": 1.69637527,
      "memory(GiB)": 117.38,
      "step": 50150,
      "train_speed(iter/s)": 1.636014
    },
    {
      "acc": 0.65682774,
      "epoch": 1.2723236935565703,
      "grad_norm": 5.8125,
      "learning_rate": 3.2030902225165814e-06,
      "loss": 1.63150272,
      "memory(GiB)": 117.38,
      "step": 50155,
      "train_speed(iter/s)": 1.636031
    },
    {
      "acc": 0.66434259,
      "epoch": 1.2724505327245053,
      "grad_norm": 4.625,
      "learning_rate": 3.2021116963295016e-06,
      "loss": 1.57543173,
      "memory(GiB)": 117.38,
      "step": 50160,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.64593782,
      "epoch": 1.2725773718924405,
      "grad_norm": 4.90625,
      "learning_rate": 3.2011332492215753e-06,
      "loss": 1.61526718,
      "memory(GiB)": 117.38,
      "step": 50165,
      "train_speed(iter/s)": 1.636067
    },
    {
      "acc": 0.65929508,
      "epoch": 1.2727042110603755,
      "grad_norm": 5.875,
      "learning_rate": 3.200154881235842e-06,
      "loss": 1.60331402,
      "memory(GiB)": 117.38,
      "step": 50170,
      "train_speed(iter/s)": 1.636085
    },
    {
      "acc": 0.66562729,
      "epoch": 1.2728310502283104,
      "grad_norm": 6.09375,
      "learning_rate": 3.1991765924153316e-06,
      "loss": 1.59988985,
      "memory(GiB)": 117.38,
      "step": 50175,
      "train_speed(iter/s)": 1.636102
    },
    {
      "acc": 0.6452199,
      "epoch": 1.2729578893962454,
      "grad_norm": 5.28125,
      "learning_rate": 3.198198382803075e-06,
      "loss": 1.59723969,
      "memory(GiB)": 117.38,
      "step": 50180,
      "train_speed(iter/s)": 1.636119
    },
    {
      "acc": 0.64359026,
      "epoch": 1.2730847285641806,
      "grad_norm": 6.46875,
      "learning_rate": 3.197220252442097e-06,
      "loss": 1.61370697,
      "memory(GiB)": 117.38,
      "step": 50185,
      "train_speed(iter/s)": 1.636137
    },
    {
      "acc": 0.64292336,
      "epoch": 1.2732115677321156,
      "grad_norm": 5.3125,
      "learning_rate": 3.1962422013754237e-06,
      "loss": 1.63647594,
      "memory(GiB)": 117.38,
      "step": 50190,
      "train_speed(iter/s)": 1.636155
    },
    {
      "acc": 0.67618837,
      "epoch": 1.2733384069000508,
      "grad_norm": 6.5625,
      "learning_rate": 3.1952642296460696e-06,
      "loss": 1.52273808,
      "memory(GiB)": 117.38,
      "step": 50195,
      "train_speed(iter/s)": 1.636173
    },
    {
      "acc": 0.63622956,
      "epoch": 1.2734652460679858,
      "grad_norm": 5.71875,
      "learning_rate": 3.194286337297051e-06,
      "loss": 1.6601965,
      "memory(GiB)": 117.38,
      "step": 50200,
      "train_speed(iter/s)": 1.636191
    },
    {
      "acc": 0.64998941,
      "epoch": 1.2735920852359208,
      "grad_norm": 9.125,
      "learning_rate": 3.1933085243713837e-06,
      "loss": 1.54855881,
      "memory(GiB)": 117.38,
      "step": 50205,
      "train_speed(iter/s)": 1.63621
    },
    {
      "acc": 0.66501269,
      "epoch": 1.2737189244038558,
      "grad_norm": 5.5625,
      "learning_rate": 3.1923307909120736e-06,
      "loss": 1.64308395,
      "memory(GiB)": 117.38,
      "step": 50210,
      "train_speed(iter/s)": 1.636228
    },
    {
      "acc": 0.65833664,
      "epoch": 1.273845763571791,
      "grad_norm": 5.15625,
      "learning_rate": 3.1913531369621243e-06,
      "loss": 1.60546989,
      "memory(GiB)": 117.38,
      "step": 50215,
      "train_speed(iter/s)": 1.636245
    },
    {
      "acc": 0.65798097,
      "epoch": 1.273972602739726,
      "grad_norm": 5.96875,
      "learning_rate": 3.19037556256454e-06,
      "loss": 1.60512352,
      "memory(GiB)": 117.38,
      "step": 50220,
      "train_speed(iter/s)": 1.636263
    },
    {
      "acc": 0.64735341,
      "epoch": 1.2740994419076612,
      "grad_norm": 6.03125,
      "learning_rate": 3.189398067762318e-06,
      "loss": 1.6562561,
      "memory(GiB)": 117.38,
      "step": 50225,
      "train_speed(iter/s)": 1.63628
    },
    {
      "acc": 0.66178579,
      "epoch": 1.2742262810755962,
      "grad_norm": 6.9375,
      "learning_rate": 3.1884206525984535e-06,
      "loss": 1.6252594,
      "memory(GiB)": 117.38,
      "step": 50230,
      "train_speed(iter/s)": 1.636297
    },
    {
      "acc": 0.64683838,
      "epoch": 1.2743531202435312,
      "grad_norm": 5.53125,
      "learning_rate": 3.1874433171159348e-06,
      "loss": 1.64285927,
      "memory(GiB)": 117.38,
      "step": 50235,
      "train_speed(iter/s)": 1.636316
    },
    {
      "acc": 0.65800476,
      "epoch": 1.2744799594114662,
      "grad_norm": 5.15625,
      "learning_rate": 3.1864660613577523e-06,
      "loss": 1.62914963,
      "memory(GiB)": 117.38,
      "step": 50240,
      "train_speed(iter/s)": 1.636333
    },
    {
      "acc": 0.66175122,
      "epoch": 1.2746067985794014,
      "grad_norm": 8.4375,
      "learning_rate": 3.185488885366889e-06,
      "loss": 1.55219488,
      "memory(GiB)": 117.38,
      "step": 50245,
      "train_speed(iter/s)": 1.636349
    },
    {
      "acc": 0.65221691,
      "epoch": 1.2747336377473364,
      "grad_norm": 4.84375,
      "learning_rate": 3.1845117891863274e-06,
      "loss": 1.54298582,
      "memory(GiB)": 117.38,
      "step": 50250,
      "train_speed(iter/s)": 1.636366
    },
    {
      "acc": 0.65278244,
      "epoch": 1.2748604769152714,
      "grad_norm": 5.21875,
      "learning_rate": 3.1835347728590414e-06,
      "loss": 1.60195274,
      "memory(GiB)": 117.38,
      "step": 50255,
      "train_speed(iter/s)": 1.636382
    },
    {
      "acc": 0.65044451,
      "epoch": 1.2749873160832066,
      "grad_norm": 5.53125,
      "learning_rate": 3.1825578364280064e-06,
      "loss": 1.55243683,
      "memory(GiB)": 117.38,
      "step": 50260,
      "train_speed(iter/s)": 1.636398
    },
    {
      "acc": 0.66600599,
      "epoch": 1.2751141552511416,
      "grad_norm": 6.875,
      "learning_rate": 3.181580979936192e-06,
      "loss": 1.59391289,
      "memory(GiB)": 117.38,
      "step": 50265,
      "train_speed(iter/s)": 1.636415
    },
    {
      "acc": 0.64651403,
      "epoch": 1.2752409944190766,
      "grad_norm": 5.375,
      "learning_rate": 3.1806042034265656e-06,
      "loss": 1.64264088,
      "memory(GiB)": 117.38,
      "step": 50270,
      "train_speed(iter/s)": 1.636432
    },
    {
      "acc": 0.65382719,
      "epoch": 1.2753678335870116,
      "grad_norm": 6.375,
      "learning_rate": 3.179627506942089e-06,
      "loss": 1.60444183,
      "memory(GiB)": 117.38,
      "step": 50275,
      "train_speed(iter/s)": 1.63645
    },
    {
      "acc": 0.66716862,
      "epoch": 1.2754946727549468,
      "grad_norm": 5.4375,
      "learning_rate": 3.17865089052572e-06,
      "loss": 1.59919567,
      "memory(GiB)": 117.38,
      "step": 50280,
      "train_speed(iter/s)": 1.636466
    },
    {
      "acc": 0.66760807,
      "epoch": 1.2756215119228818,
      "grad_norm": 5.09375,
      "learning_rate": 3.177674354220418e-06,
      "loss": 1.55120783,
      "memory(GiB)": 117.38,
      "step": 50285,
      "train_speed(iter/s)": 1.636484
    },
    {
      "acc": 0.65816998,
      "epoch": 1.275748351090817,
      "grad_norm": 5.5625,
      "learning_rate": 3.1766978980691355e-06,
      "loss": 1.59152641,
      "memory(GiB)": 117.38,
      "step": 50290,
      "train_speed(iter/s)": 1.636501
    },
    {
      "acc": 0.67199364,
      "epoch": 1.275875190258752,
      "grad_norm": 5.0625,
      "learning_rate": 3.1757215221148173e-06,
      "loss": 1.59214211,
      "memory(GiB)": 117.38,
      "step": 50295,
      "train_speed(iter/s)": 1.636518
    },
    {
      "acc": 0.65828438,
      "epoch": 1.276002029426687,
      "grad_norm": 5.25,
      "learning_rate": 3.174745226400413e-06,
      "loss": 1.60132904,
      "memory(GiB)": 117.38,
      "step": 50300,
      "train_speed(iter/s)": 1.636536
    },
    {
      "acc": 0.66523685,
      "epoch": 1.276128868594622,
      "grad_norm": 5.65625,
      "learning_rate": 3.1737690109688613e-06,
      "loss": 1.55143909,
      "memory(GiB)": 117.38,
      "step": 50305,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.65723853,
      "epoch": 1.2762557077625571,
      "grad_norm": 5.8125,
      "learning_rate": 3.1727928758631054e-06,
      "loss": 1.56019688,
      "memory(GiB)": 117.38,
      "step": 50310,
      "train_speed(iter/s)": 1.636572
    },
    {
      "acc": 0.65220823,
      "epoch": 1.2763825469304921,
      "grad_norm": 5.71875,
      "learning_rate": 3.1718168211260734e-06,
      "loss": 1.66425323,
      "memory(GiB)": 117.38,
      "step": 50315,
      "train_speed(iter/s)": 1.636589
    },
    {
      "acc": 0.64509587,
      "epoch": 1.2765093860984271,
      "grad_norm": 7.40625,
      "learning_rate": 3.1708408468007014e-06,
      "loss": 1.66488075,
      "memory(GiB)": 117.38,
      "step": 50320,
      "train_speed(iter/s)": 1.636607
    },
    {
      "acc": 0.65077329,
      "epoch": 1.2766362252663623,
      "grad_norm": 5.21875,
      "learning_rate": 3.169864952929914e-06,
      "loss": 1.64300232,
      "memory(GiB)": 117.38,
      "step": 50325,
      "train_speed(iter/s)": 1.636624
    },
    {
      "acc": 0.65535846,
      "epoch": 1.2767630644342973,
      "grad_norm": 4.84375,
      "learning_rate": 3.168889139556639e-06,
      "loss": 1.61614685,
      "memory(GiB)": 117.38,
      "step": 50330,
      "train_speed(iter/s)": 1.636641
    },
    {
      "acc": 0.65707207,
      "epoch": 1.2768899036022323,
      "grad_norm": 4.71875,
      "learning_rate": 3.1679134067237942e-06,
      "loss": 1.60960522,
      "memory(GiB)": 117.38,
      "step": 50335,
      "train_speed(iter/s)": 1.636659
    },
    {
      "acc": 0.65848408,
      "epoch": 1.2770167427701673,
      "grad_norm": 5.8125,
      "learning_rate": 3.166937754474296e-06,
      "loss": 1.59378109,
      "memory(GiB)": 117.38,
      "step": 50340,
      "train_speed(iter/s)": 1.636676
    },
    {
      "acc": 0.65967274,
      "epoch": 1.2771435819381025,
      "grad_norm": 5.46875,
      "learning_rate": 3.165962182851061e-06,
      "loss": 1.5828186,
      "memory(GiB)": 117.38,
      "step": 50345,
      "train_speed(iter/s)": 1.636693
    },
    {
      "acc": 0.66474075,
      "epoch": 1.2772704211060375,
      "grad_norm": 6.28125,
      "learning_rate": 3.1649866918969984e-06,
      "loss": 1.57147827,
      "memory(GiB)": 117.38,
      "step": 50350,
      "train_speed(iter/s)": 1.636711
    },
    {
      "acc": 0.64480772,
      "epoch": 1.2773972602739727,
      "grad_norm": 6.40625,
      "learning_rate": 3.164011281655013e-06,
      "loss": 1.60990448,
      "memory(GiB)": 117.38,
      "step": 50355,
      "train_speed(iter/s)": 1.636729
    },
    {
      "acc": 0.64572506,
      "epoch": 1.2775240994419077,
      "grad_norm": 5.75,
      "learning_rate": 3.163035952168007e-06,
      "loss": 1.66291122,
      "memory(GiB)": 117.38,
      "step": 50360,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.66791496,
      "epoch": 1.2776509386098427,
      "grad_norm": 7.625,
      "learning_rate": 3.1620607034788832e-06,
      "loss": 1.64908981,
      "memory(GiB)": 117.38,
      "step": 50365,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.65327721,
      "epoch": 1.2777777777777777,
      "grad_norm": 4.96875,
      "learning_rate": 3.1610855356305354e-06,
      "loss": 1.65682831,
      "memory(GiB)": 117.38,
      "step": 50370,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.64920874,
      "epoch": 1.277904616945713,
      "grad_norm": 5.84375,
      "learning_rate": 3.160110448665854e-06,
      "loss": 1.61971951,
      "memory(GiB)": 117.38,
      "step": 50375,
      "train_speed(iter/s)": 1.636797
    },
    {
      "acc": 0.66433744,
      "epoch": 1.2780314561136479,
      "grad_norm": 4.71875,
      "learning_rate": 3.159135442627731e-06,
      "loss": 1.54368992,
      "memory(GiB)": 117.38,
      "step": 50380,
      "train_speed(iter/s)": 1.636814
    },
    {
      "acc": 0.66488256,
      "epoch": 1.278158295281583,
      "grad_norm": 5.4375,
      "learning_rate": 3.158160517559049e-06,
      "loss": 1.58785248,
      "memory(GiB)": 117.38,
      "step": 50385,
      "train_speed(iter/s)": 1.636831
    },
    {
      "acc": 0.65733833,
      "epoch": 1.278285134449518,
      "grad_norm": 5.375,
      "learning_rate": 3.157185673502693e-06,
      "loss": 1.61690845,
      "memory(GiB)": 117.38,
      "step": 50390,
      "train_speed(iter/s)": 1.636846
    },
    {
      "acc": 0.65631394,
      "epoch": 1.278411973617453,
      "grad_norm": 6.84375,
      "learning_rate": 3.156210910501537e-06,
      "loss": 1.58626194,
      "memory(GiB)": 117.38,
      "step": 50395,
      "train_speed(iter/s)": 1.636864
    },
    {
      "acc": 0.64931765,
      "epoch": 1.278538812785388,
      "grad_norm": 5.0625,
      "learning_rate": 3.155236228598457e-06,
      "loss": 1.59983044,
      "memory(GiB)": 117.38,
      "step": 50400,
      "train_speed(iter/s)": 1.636882
    },
    {
      "acc": 0.65192442,
      "epoch": 1.2786656519533233,
      "grad_norm": 6.0625,
      "learning_rate": 3.1542616278363238e-06,
      "loss": 1.60695801,
      "memory(GiB)": 117.38,
      "step": 50405,
      "train_speed(iter/s)": 1.636898
    },
    {
      "acc": 0.65318117,
      "epoch": 1.2787924911212583,
      "grad_norm": 5.9375,
      "learning_rate": 3.1532871082580064e-06,
      "loss": 1.63932648,
      "memory(GiB)": 117.38,
      "step": 50410,
      "train_speed(iter/s)": 1.636915
    },
    {
      "acc": 0.65757351,
      "epoch": 1.2789193302891932,
      "grad_norm": 5.53125,
      "learning_rate": 3.152312669906366e-06,
      "loss": 1.60349197,
      "memory(GiB)": 117.38,
      "step": 50415,
      "train_speed(iter/s)": 1.636933
    },
    {
      "acc": 0.67174292,
      "epoch": 1.2790461694571285,
      "grad_norm": 5.53125,
      "learning_rate": 3.1513383128242624e-06,
      "loss": 1.52506351,
      "memory(GiB)": 117.38,
      "step": 50420,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.66882143,
      "epoch": 1.2791730086250634,
      "grad_norm": 6.90625,
      "learning_rate": 3.150364037054555e-06,
      "loss": 1.61610775,
      "memory(GiB)": 117.38,
      "step": 50425,
      "train_speed(iter/s)": 1.636967
    },
    {
      "acc": 0.65421519,
      "epoch": 1.2792998477929984,
      "grad_norm": 5.8125,
      "learning_rate": 3.149389842640096e-06,
      "loss": 1.55048885,
      "memory(GiB)": 117.38,
      "step": 50430,
      "train_speed(iter/s)": 1.636984
    },
    {
      "acc": 0.66348271,
      "epoch": 1.2794266869609334,
      "grad_norm": 6.46875,
      "learning_rate": 3.148415729623732e-06,
      "loss": 1.5821991,
      "memory(GiB)": 117.38,
      "step": 50435,
      "train_speed(iter/s)": 1.637002
    },
    {
      "acc": 0.6693789,
      "epoch": 1.2795535261288686,
      "grad_norm": 4.59375,
      "learning_rate": 3.1474416980483126e-06,
      "loss": 1.61463356,
      "memory(GiB)": 117.38,
      "step": 50440,
      "train_speed(iter/s)": 1.63702
    },
    {
      "acc": 0.67107062,
      "epoch": 1.2796803652968036,
      "grad_norm": 6.8125,
      "learning_rate": 3.1464677479566774e-06,
      "loss": 1.58897066,
      "memory(GiB)": 117.38,
      "step": 50445,
      "train_speed(iter/s)": 1.637037
    },
    {
      "acc": 0.65986547,
      "epoch": 1.2798072044647388,
      "grad_norm": 5.875,
      "learning_rate": 3.1454938793916677e-06,
      "loss": 1.62581615,
      "memory(GiB)": 117.38,
      "step": 50450,
      "train_speed(iter/s)": 1.637055
    },
    {
      "acc": 0.64644847,
      "epoch": 1.2799340436326738,
      "grad_norm": 5.8125,
      "learning_rate": 3.144520092396115e-06,
      "loss": 1.63034859,
      "memory(GiB)": 117.38,
      "step": 50455,
      "train_speed(iter/s)": 1.637073
    },
    {
      "acc": 0.64993873,
      "epoch": 1.2800608828006088,
      "grad_norm": 4.84375,
      "learning_rate": 3.1435463870128536e-06,
      "loss": 1.6300766,
      "memory(GiB)": 117.38,
      "step": 50460,
      "train_speed(iter/s)": 1.63709
    },
    {
      "acc": 0.67384334,
      "epoch": 1.2801877219685438,
      "grad_norm": 6.125,
      "learning_rate": 3.142572763284709e-06,
      "loss": 1.54737701,
      "memory(GiB)": 117.38,
      "step": 50465,
      "train_speed(iter/s)": 1.637109
    },
    {
      "acc": 0.65003557,
      "epoch": 1.280314561136479,
      "grad_norm": 5.625,
      "learning_rate": 3.14159922125451e-06,
      "loss": 1.63682899,
      "memory(GiB)": 117.38,
      "step": 50470,
      "train_speed(iter/s)": 1.637128
    },
    {
      "acc": 0.65351772,
      "epoch": 1.280441400304414,
      "grad_norm": 4.6875,
      "learning_rate": 3.1406257609650724e-06,
      "loss": 1.58972874,
      "memory(GiB)": 117.38,
      "step": 50475,
      "train_speed(iter/s)": 1.637145
    },
    {
      "acc": 0.66596484,
      "epoch": 1.280568239472349,
      "grad_norm": 7.40625,
      "learning_rate": 3.139652382459215e-06,
      "loss": 1.60235176,
      "memory(GiB)": 117.38,
      "step": 50480,
      "train_speed(iter/s)": 1.637163
    },
    {
      "acc": 0.64224672,
      "epoch": 1.2806950786402842,
      "grad_norm": 5.96875,
      "learning_rate": 3.1386790857797535e-06,
      "loss": 1.64954071,
      "memory(GiB)": 117.38,
      "step": 50485,
      "train_speed(iter/s)": 1.637181
    },
    {
      "acc": 0.67724533,
      "epoch": 1.2808219178082192,
      "grad_norm": 7.40625,
      "learning_rate": 3.1377058709694957e-06,
      "loss": 1.5474041,
      "memory(GiB)": 117.38,
      "step": 50490,
      "train_speed(iter/s)": 1.637198
    },
    {
      "acc": 0.65323787,
      "epoch": 1.2809487569761542,
      "grad_norm": 5.0625,
      "learning_rate": 3.1367327380712477e-06,
      "loss": 1.54239893,
      "memory(GiB)": 117.38,
      "step": 50495,
      "train_speed(iter/s)": 1.637215
    },
    {
      "acc": 0.64975319,
      "epoch": 1.2810755961440892,
      "grad_norm": 4.875,
      "learning_rate": 3.135759687127812e-06,
      "loss": 1.59814892,
      "memory(GiB)": 117.38,
      "step": 50500,
      "train_speed(iter/s)": 1.637232
    },
    {
      "acc": 0.65838127,
      "epoch": 1.2812024353120244,
      "grad_norm": 4.5625,
      "learning_rate": 3.134786718181989e-06,
      "loss": 1.60836372,
      "memory(GiB)": 117.38,
      "step": 50505,
      "train_speed(iter/s)": 1.637249
    },
    {
      "acc": 0.65520487,
      "epoch": 1.2813292744799594,
      "grad_norm": 5.78125,
      "learning_rate": 3.133813831276575e-06,
      "loss": 1.58764572,
      "memory(GiB)": 117.38,
      "step": 50510,
      "train_speed(iter/s)": 1.637267
    },
    {
      "acc": 0.65381041,
      "epoch": 1.2814561136478946,
      "grad_norm": 6.75,
      "learning_rate": 3.1328410264543585e-06,
      "loss": 1.61964016,
      "memory(GiB)": 117.38,
      "step": 50515,
      "train_speed(iter/s)": 1.637284
    },
    {
      "acc": 0.65839658,
      "epoch": 1.2815829528158296,
      "grad_norm": 5.75,
      "learning_rate": 3.131868303758131e-06,
      "loss": 1.54832602,
      "memory(GiB)": 117.38,
      "step": 50520,
      "train_speed(iter/s)": 1.6373
    },
    {
      "acc": 0.65777388,
      "epoch": 1.2817097919837646,
      "grad_norm": 6.53125,
      "learning_rate": 3.1308956632306754e-06,
      "loss": 1.5596983,
      "memory(GiB)": 117.38,
      "step": 50525,
      "train_speed(iter/s)": 1.637319
    },
    {
      "acc": 0.65868034,
      "epoch": 1.2818366311516995,
      "grad_norm": 5.5,
      "learning_rate": 3.129923104914776e-06,
      "loss": 1.60860634,
      "memory(GiB)": 117.38,
      "step": 50530,
      "train_speed(iter/s)": 1.637337
    },
    {
      "acc": 0.66887999,
      "epoch": 1.2819634703196348,
      "grad_norm": 6.09375,
      "learning_rate": 3.1289506288532045e-06,
      "loss": 1.47076874,
      "memory(GiB)": 117.38,
      "step": 50535,
      "train_speed(iter/s)": 1.637354
    },
    {
      "acc": 0.64477448,
      "epoch": 1.2820903094875697,
      "grad_norm": 6.0,
      "learning_rate": 3.12797823508874e-06,
      "loss": 1.62783394,
      "memory(GiB)": 117.38,
      "step": 50540,
      "train_speed(iter/s)": 1.637372
    },
    {
      "acc": 0.66084676,
      "epoch": 1.282217148655505,
      "grad_norm": 5.5625,
      "learning_rate": 3.127005923664149e-06,
      "loss": 1.59168491,
      "memory(GiB)": 117.38,
      "step": 50545,
      "train_speed(iter/s)": 1.637388
    },
    {
      "acc": 0.65986476,
      "epoch": 1.28234398782344,
      "grad_norm": 5.21875,
      "learning_rate": 3.1260336946222014e-06,
      "loss": 1.55225887,
      "memory(GiB)": 117.38,
      "step": 50550,
      "train_speed(iter/s)": 1.637406
    },
    {
      "acc": 0.6664011,
      "epoch": 1.282470826991375,
      "grad_norm": 6.3125,
      "learning_rate": 3.1250615480056584e-06,
      "loss": 1.51310654,
      "memory(GiB)": 117.38,
      "step": 50555,
      "train_speed(iter/s)": 1.637423
    },
    {
      "acc": 0.67312164,
      "epoch": 1.28259766615931,
      "grad_norm": 5.0,
      "learning_rate": 3.124089483857278e-06,
      "loss": 1.48702517,
      "memory(GiB)": 117.38,
      "step": 50560,
      "train_speed(iter/s)": 1.637439
    },
    {
      "acc": 0.65216436,
      "epoch": 1.2827245053272451,
      "grad_norm": 8.0625,
      "learning_rate": 3.123117502219819e-06,
      "loss": 1.62879372,
      "memory(GiB)": 117.38,
      "step": 50565,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.65934901,
      "epoch": 1.2828513444951801,
      "grad_norm": 6.375,
      "learning_rate": 3.122145603136032e-06,
      "loss": 1.59183836,
      "memory(GiB)": 117.38,
      "step": 50570,
      "train_speed(iter/s)": 1.637474
    },
    {
      "acc": 0.66513715,
      "epoch": 1.2829781836631151,
      "grad_norm": 6.5,
      "learning_rate": 3.1211737866486653e-06,
      "loss": 1.63434715,
      "memory(GiB)": 117.38,
      "step": 50575,
      "train_speed(iter/s)": 1.63749
    },
    {
      "acc": 0.66325121,
      "epoch": 1.2831050228310503,
      "grad_norm": 4.84375,
      "learning_rate": 3.1202020528004627e-06,
      "loss": 1.57187872,
      "memory(GiB)": 117.38,
      "step": 50580,
      "train_speed(iter/s)": 1.637507
    },
    {
      "acc": 0.65429831,
      "epoch": 1.2832318619989853,
      "grad_norm": 5.1875,
      "learning_rate": 3.119230401634167e-06,
      "loss": 1.59203625,
      "memory(GiB)": 117.38,
      "step": 50585,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.65885706,
      "epoch": 1.2833587011669203,
      "grad_norm": 5.4375,
      "learning_rate": 3.118258833192517e-06,
      "loss": 1.55022364,
      "memory(GiB)": 117.38,
      "step": 50590,
      "train_speed(iter/s)": 1.637541
    },
    {
      "acc": 0.64973927,
      "epoch": 1.2834855403348553,
      "grad_norm": 6.875,
      "learning_rate": 3.117287347518242e-06,
      "loss": 1.59403229,
      "memory(GiB)": 117.38,
      "step": 50595,
      "train_speed(iter/s)": 1.637558
    },
    {
      "acc": 0.65729885,
      "epoch": 1.2836123795027905,
      "grad_norm": 6.71875,
      "learning_rate": 3.116315944654077e-06,
      "loss": 1.69018822,
      "memory(GiB)": 117.38,
      "step": 50600,
      "train_speed(iter/s)": 1.637575
    },
    {
      "acc": 0.64665127,
      "epoch": 1.2837392186707255,
      "grad_norm": 6.09375,
      "learning_rate": 3.115344624642745e-06,
      "loss": 1.64469833,
      "memory(GiB)": 117.38,
      "step": 50605,
      "train_speed(iter/s)": 1.637592
    },
    {
      "acc": 0.67346849,
      "epoch": 1.2838660578386607,
      "grad_norm": 7.34375,
      "learning_rate": 3.1143733875269734e-06,
      "loss": 1.61861629,
      "memory(GiB)": 117.38,
      "step": 50610,
      "train_speed(iter/s)": 1.63761
    },
    {
      "acc": 0.67737298,
      "epoch": 1.2839928970065957,
      "grad_norm": 5.8125,
      "learning_rate": 3.1134022333494774e-06,
      "loss": 1.52774324,
      "memory(GiB)": 117.38,
      "step": 50615,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.67497382,
      "epoch": 1.2841197361745307,
      "grad_norm": 5.8125,
      "learning_rate": 3.1124311621529744e-06,
      "loss": 1.51304722,
      "memory(GiB)": 117.38,
      "step": 50620,
      "train_speed(iter/s)": 1.637644
    },
    {
      "acc": 0.65862589,
      "epoch": 1.2842465753424657,
      "grad_norm": 6.5,
      "learning_rate": 3.111460173980175e-06,
      "loss": 1.63301544,
      "memory(GiB)": 117.38,
      "step": 50625,
      "train_speed(iter/s)": 1.637661
    },
    {
      "acc": 0.65544624,
      "epoch": 1.2843734145104009,
      "grad_norm": 5.34375,
      "learning_rate": 3.110489268873792e-06,
      "loss": 1.55189438,
      "memory(GiB)": 117.38,
      "step": 50630,
      "train_speed(iter/s)": 1.637679
    },
    {
      "acc": 0.66510191,
      "epoch": 1.2845002536783359,
      "grad_norm": 6.3125,
      "learning_rate": 3.1095184468765248e-06,
      "loss": 1.61021194,
      "memory(GiB)": 117.38,
      "step": 50635,
      "train_speed(iter/s)": 1.637696
    },
    {
      "acc": 0.65029812,
      "epoch": 1.2846270928462709,
      "grad_norm": 4.625,
      "learning_rate": 3.1085477080310766e-06,
      "loss": 1.59244518,
      "memory(GiB)": 117.38,
      "step": 50640,
      "train_speed(iter/s)": 1.637714
    },
    {
      "acc": 0.66593041,
      "epoch": 1.284753932014206,
      "grad_norm": 5.75,
      "learning_rate": 3.1075770523801453e-06,
      "loss": 1.52028494,
      "memory(GiB)": 117.38,
      "step": 50645,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.65899096,
      "epoch": 1.284880771182141,
      "grad_norm": 5.34375,
      "learning_rate": 3.106606479966426e-06,
      "loss": 1.58623772,
      "memory(GiB)": 117.38,
      "step": 50650,
      "train_speed(iter/s)": 1.637747
    },
    {
      "acc": 0.67684488,
      "epoch": 1.285007610350076,
      "grad_norm": 6.0,
      "learning_rate": 3.1056359908326044e-06,
      "loss": 1.57980261,
      "memory(GiB)": 117.38,
      "step": 50655,
      "train_speed(iter/s)": 1.637765
    },
    {
      "acc": 0.64775853,
      "epoch": 1.285134449518011,
      "grad_norm": 6.84375,
      "learning_rate": 3.1046655850213707e-06,
      "loss": 1.59221067,
      "memory(GiB)": 117.38,
      "step": 50660,
      "train_speed(iter/s)": 1.637783
    },
    {
      "acc": 0.64618044,
      "epoch": 1.2852612886859462,
      "grad_norm": 6.03125,
      "learning_rate": 3.103695262575407e-06,
      "loss": 1.6291193,
      "memory(GiB)": 117.38,
      "step": 50665,
      "train_speed(iter/s)": 1.6378
    },
    {
      "acc": 0.65867229,
      "epoch": 1.2853881278538812,
      "grad_norm": 5.75,
      "learning_rate": 3.102725023537393e-06,
      "loss": 1.61899948,
      "memory(GiB)": 117.38,
      "step": 50670,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.66438456,
      "epoch": 1.2855149670218164,
      "grad_norm": 5.46875,
      "learning_rate": 3.1017548679500008e-06,
      "loss": 1.52022285,
      "memory(GiB)": 117.38,
      "step": 50675,
      "train_speed(iter/s)": 1.637835
    },
    {
      "acc": 0.64881101,
      "epoch": 1.2856418061897514,
      "grad_norm": 7.46875,
      "learning_rate": 3.1007847958559057e-06,
      "loss": 1.61680336,
      "memory(GiB)": 117.38,
      "step": 50680,
      "train_speed(iter/s)": 1.637853
    },
    {
      "acc": 0.64892354,
      "epoch": 1.2857686453576864,
      "grad_norm": 5.5625,
      "learning_rate": 3.099814807297774e-06,
      "loss": 1.63759174,
      "memory(GiB)": 117.38,
      "step": 50685,
      "train_speed(iter/s)": 1.637871
    },
    {
      "acc": 0.67104535,
      "epoch": 1.2858954845256214,
      "grad_norm": 5.5625,
      "learning_rate": 3.098844902318272e-06,
      "loss": 1.51733265,
      "memory(GiB)": 117.38,
      "step": 50690,
      "train_speed(iter/s)": 1.637888
    },
    {
      "acc": 0.66727409,
      "epoch": 1.2860223236935566,
      "grad_norm": 5.78125,
      "learning_rate": 3.0978750809600596e-06,
      "loss": 1.59357586,
      "memory(GiB)": 117.38,
      "step": 50695,
      "train_speed(iter/s)": 1.637903
    },
    {
      "acc": 0.66900167,
      "epoch": 1.2861491628614916,
      "grad_norm": 5.8125,
      "learning_rate": 3.0969053432657913e-06,
      "loss": 1.56347475,
      "memory(GiB)": 117.38,
      "step": 50700,
      "train_speed(iter/s)": 1.63792
    },
    {
      "acc": 0.66129742,
      "epoch": 1.2862760020294268,
      "grad_norm": 5.65625,
      "learning_rate": 3.0959356892781246e-06,
      "loss": 1.57301655,
      "memory(GiB)": 117.38,
      "step": 50705,
      "train_speed(iter/s)": 1.637937
    },
    {
      "acc": 0.66649795,
      "epoch": 1.2864028411973618,
      "grad_norm": 6.5,
      "learning_rate": 3.0949661190397072e-06,
      "loss": 1.5851676,
      "memory(GiB)": 117.38,
      "step": 50710,
      "train_speed(iter/s)": 1.637955
    },
    {
      "acc": 0.65414267,
      "epoch": 1.2865296803652968,
      "grad_norm": 6.09375,
      "learning_rate": 3.0939966325931852e-06,
      "loss": 1.64785309,
      "memory(GiB)": 117.38,
      "step": 50715,
      "train_speed(iter/s)": 1.637972
    },
    {
      "acc": 0.63294878,
      "epoch": 1.2866565195332318,
      "grad_norm": 5.1875,
      "learning_rate": 3.0930272299812e-06,
      "loss": 1.68220959,
      "memory(GiB)": 117.38,
      "step": 50720,
      "train_speed(iter/s)": 1.63799
    },
    {
      "acc": 0.65390005,
      "epoch": 1.286783358701167,
      "grad_norm": 5.21875,
      "learning_rate": 3.0920579112463916e-06,
      "loss": 1.59368076,
      "memory(GiB)": 117.38,
      "step": 50725,
      "train_speed(iter/s)": 1.638009
    },
    {
      "acc": 0.65120931,
      "epoch": 1.286910197869102,
      "grad_norm": 5.40625,
      "learning_rate": 3.0910886764313964e-06,
      "loss": 1.68505592,
      "memory(GiB)": 117.38,
      "step": 50730,
      "train_speed(iter/s)": 1.638026
    },
    {
      "acc": 0.66721745,
      "epoch": 1.287037037037037,
      "grad_norm": 5.34375,
      "learning_rate": 3.0901195255788406e-06,
      "loss": 1.54817085,
      "memory(GiB)": 117.38,
      "step": 50735,
      "train_speed(iter/s)": 1.638044
    },
    {
      "acc": 0.6673748,
      "epoch": 1.2871638762049722,
      "grad_norm": 5.9375,
      "learning_rate": 3.089150458731357e-06,
      "loss": 1.54632559,
      "memory(GiB)": 117.38,
      "step": 50740,
      "train_speed(iter/s)": 1.638061
    },
    {
      "acc": 0.67838879,
      "epoch": 1.2872907153729072,
      "grad_norm": 5.4375,
      "learning_rate": 3.0881814759315666e-06,
      "loss": 1.48835278,
      "memory(GiB)": 117.38,
      "step": 50745,
      "train_speed(iter/s)": 1.638078
    },
    {
      "acc": 0.65278931,
      "epoch": 1.2874175545408422,
      "grad_norm": 5.625,
      "learning_rate": 3.0872125772220934e-06,
      "loss": 1.64299603,
      "memory(GiB)": 117.38,
      "step": 50750,
      "train_speed(iter/s)": 1.638094
    },
    {
      "acc": 0.65849876,
      "epoch": 1.2875443937087772,
      "grad_norm": 6.15625,
      "learning_rate": 3.0862437626455483e-06,
      "loss": 1.62352982,
      "memory(GiB)": 117.38,
      "step": 50755,
      "train_speed(iter/s)": 1.638111
    },
    {
      "acc": 0.66881943,
      "epoch": 1.2876712328767124,
      "grad_norm": 6.1875,
      "learning_rate": 3.0852750322445473e-06,
      "loss": 1.51064625,
      "memory(GiB)": 117.38,
      "step": 50760,
      "train_speed(iter/s)": 1.638128
    },
    {
      "acc": 0.66159663,
      "epoch": 1.2877980720446474,
      "grad_norm": 5.84375,
      "learning_rate": 3.0843063860616982e-06,
      "loss": 1.55973902,
      "memory(GiB)": 117.38,
      "step": 50765,
      "train_speed(iter/s)": 1.638146
    },
    {
      "acc": 0.66133451,
      "epoch": 1.2879249112125826,
      "grad_norm": 5.59375,
      "learning_rate": 3.0833378241396094e-06,
      "loss": 1.58226347,
      "memory(GiB)": 117.38,
      "step": 50770,
      "train_speed(iter/s)": 1.638165
    },
    {
      "acc": 0.64204926,
      "epoch": 1.2880517503805176,
      "grad_norm": 5.09375,
      "learning_rate": 3.0823693465208794e-06,
      "loss": 1.62418671,
      "memory(GiB)": 117.38,
      "step": 50775,
      "train_speed(iter/s)": 1.638182
    },
    {
      "acc": 0.65554996,
      "epoch": 1.2881785895484525,
      "grad_norm": 7.96875,
      "learning_rate": 3.081400953248106e-06,
      "loss": 1.58124094,
      "memory(GiB)": 117.38,
      "step": 50780,
      "train_speed(iter/s)": 1.6382
    },
    {
      "acc": 0.65164385,
      "epoch": 1.2883054287163875,
      "grad_norm": 7.875,
      "learning_rate": 3.0804326443638854e-06,
      "loss": 1.59955006,
      "memory(GiB)": 117.38,
      "step": 50785,
      "train_speed(iter/s)": 1.638218
    },
    {
      "acc": 0.65098739,
      "epoch": 1.2884322678843227,
      "grad_norm": 6.96875,
      "learning_rate": 3.0794644199108087e-06,
      "loss": 1.67862701,
      "memory(GiB)": 117.38,
      "step": 50790,
      "train_speed(iter/s)": 1.638236
    },
    {
      "acc": 0.65335464,
      "epoch": 1.2885591070522577,
      "grad_norm": 5.875,
      "learning_rate": 3.07849627993146e-06,
      "loss": 1.6175621,
      "memory(GiB)": 117.38,
      "step": 50795,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.65353198,
      "epoch": 1.2886859462201927,
      "grad_norm": 5.03125,
      "learning_rate": 3.0775282244684233e-06,
      "loss": 1.56455421,
      "memory(GiB)": 117.38,
      "step": 50800,
      "train_speed(iter/s)": 1.63827
    },
    {
      "acc": 0.64816279,
      "epoch": 1.288812785388128,
      "grad_norm": 7.34375,
      "learning_rate": 3.076560253564279e-06,
      "loss": 1.63729172,
      "memory(GiB)": 117.38,
      "step": 50805,
      "train_speed(iter/s)": 1.638288
    },
    {
      "acc": 0.65356383,
      "epoch": 1.288939624556063,
      "grad_norm": 7.46875,
      "learning_rate": 3.0755923672616038e-06,
      "loss": 1.6340889,
      "memory(GiB)": 117.38,
      "step": 50810,
      "train_speed(iter/s)": 1.638305
    },
    {
      "acc": 0.6475235,
      "epoch": 1.289066463723998,
      "grad_norm": 6.4375,
      "learning_rate": 3.074624565602966e-06,
      "loss": 1.68175354,
      "memory(GiB)": 117.38,
      "step": 50815,
      "train_speed(iter/s)": 1.638322
    },
    {
      "acc": 0.6499897,
      "epoch": 1.289193302891933,
      "grad_norm": 5.8125,
      "learning_rate": 3.073656848630937e-06,
      "loss": 1.59709034,
      "memory(GiB)": 117.38,
      "step": 50820,
      "train_speed(iter/s)": 1.638339
    },
    {
      "acc": 0.64909759,
      "epoch": 1.2893201420598681,
      "grad_norm": 5.59375,
      "learning_rate": 3.0726892163880784e-06,
      "loss": 1.61072235,
      "memory(GiB)": 117.38,
      "step": 50825,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.65183229,
      "epoch": 1.289446981227803,
      "grad_norm": 4.90625,
      "learning_rate": 3.071721668916956e-06,
      "loss": 1.61939106,
      "memory(GiB)": 117.38,
      "step": 50830,
      "train_speed(iter/s)": 1.638372
    },
    {
      "acc": 0.67048807,
      "epoch": 1.2895738203957383,
      "grad_norm": 6.71875,
      "learning_rate": 3.0707542062601225e-06,
      "loss": 1.52533302,
      "memory(GiB)": 117.38,
      "step": 50835,
      "train_speed(iter/s)": 1.63839
    },
    {
      "acc": 0.645679,
      "epoch": 1.2897006595636733,
      "grad_norm": 5.125,
      "learning_rate": 3.0697868284601323e-06,
      "loss": 1.6127409,
      "memory(GiB)": 117.38,
      "step": 50840,
      "train_speed(iter/s)": 1.638408
    },
    {
      "acc": 0.6747376,
      "epoch": 1.2898274987316083,
      "grad_norm": 6.34375,
      "learning_rate": 3.068819535559534e-06,
      "loss": 1.50929184,
      "memory(GiB)": 117.38,
      "step": 50845,
      "train_speed(iter/s)": 1.638425
    },
    {
      "acc": 0.65932112,
      "epoch": 1.2899543378995433,
      "grad_norm": 5.78125,
      "learning_rate": 3.0678523276008774e-06,
      "loss": 1.56526995,
      "memory(GiB)": 117.38,
      "step": 50850,
      "train_speed(iter/s)": 1.638443
    },
    {
      "acc": 0.6631443,
      "epoch": 1.2900811770674785,
      "grad_norm": 6.21875,
      "learning_rate": 3.0668852046267e-06,
      "loss": 1.56042929,
      "memory(GiB)": 117.38,
      "step": 50855,
      "train_speed(iter/s)": 1.63846
    },
    {
      "acc": 0.65926046,
      "epoch": 1.2902080162354135,
      "grad_norm": 6.3125,
      "learning_rate": 3.0659181666795413e-06,
      "loss": 1.56487341,
      "memory(GiB)": 117.38,
      "step": 50860,
      "train_speed(iter/s)": 1.638477
    },
    {
      "acc": 0.65164356,
      "epoch": 1.2903348554033487,
      "grad_norm": 6.0625,
      "learning_rate": 3.0649512138019376e-06,
      "loss": 1.61396122,
      "memory(GiB)": 117.38,
      "step": 50865,
      "train_speed(iter/s)": 1.638495
    },
    {
      "acc": 0.65864949,
      "epoch": 1.2904616945712837,
      "grad_norm": 5.71875,
      "learning_rate": 3.0639843460364203e-06,
      "loss": 1.58406963,
      "memory(GiB)": 117.38,
      "step": 50870,
      "train_speed(iter/s)": 1.638241
    },
    {
      "acc": 0.64596262,
      "epoch": 1.2905885337392187,
      "grad_norm": 6.0625,
      "learning_rate": 3.0630175634255134e-06,
      "loss": 1.66630001,
      "memory(GiB)": 117.38,
      "step": 50875,
      "train_speed(iter/s)": 1.638257
    },
    {
      "acc": 0.64331188,
      "epoch": 1.2907153729071537,
      "grad_norm": 5.34375,
      "learning_rate": 3.062050866011742e-06,
      "loss": 1.6839241,
      "memory(GiB)": 117.38,
      "step": 50880,
      "train_speed(iter/s)": 1.638274
    },
    {
      "acc": 0.646241,
      "epoch": 1.2908422120750889,
      "grad_norm": 7.71875,
      "learning_rate": 3.0610842538376264e-06,
      "loss": 1.71417999,
      "memory(GiB)": 117.38,
      "step": 50885,
      "train_speed(iter/s)": 1.638291
    },
    {
      "acc": 0.66505051,
      "epoch": 1.2909690512430239,
      "grad_norm": 5.4375,
      "learning_rate": 3.060117726945683e-06,
      "loss": 1.60677147,
      "memory(GiB)": 117.38,
      "step": 50890,
      "train_speed(iter/s)": 1.638309
    },
    {
      "acc": 0.63203182,
      "epoch": 1.2910958904109588,
      "grad_norm": 5.53125,
      "learning_rate": 3.059151285378421e-06,
      "loss": 1.70354424,
      "memory(GiB)": 117.38,
      "step": 50895,
      "train_speed(iter/s)": 1.638327
    },
    {
      "acc": 0.66179781,
      "epoch": 1.291222729578894,
      "grad_norm": 6.0,
      "learning_rate": 3.0581849291783518e-06,
      "loss": 1.54617586,
      "memory(GiB)": 117.38,
      "step": 50900,
      "train_speed(iter/s)": 1.638345
    },
    {
      "acc": 0.65977721,
      "epoch": 1.291349568746829,
      "grad_norm": 5.90625,
      "learning_rate": 3.057218658387977e-06,
      "loss": 1.56992302,
      "memory(GiB)": 117.38,
      "step": 50905,
      "train_speed(iter/s)": 1.638363
    },
    {
      "acc": 0.65622191,
      "epoch": 1.291476407914764,
      "grad_norm": 6.96875,
      "learning_rate": 3.0562524730498023e-06,
      "loss": 1.62710361,
      "memory(GiB)": 117.38,
      "step": 50910,
      "train_speed(iter/s)": 1.638382
    },
    {
      "acc": 0.64522934,
      "epoch": 1.291603247082699,
      "grad_norm": 6.125,
      "learning_rate": 3.055286373206321e-06,
      "loss": 1.67231064,
      "memory(GiB)": 117.38,
      "step": 50915,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.66228938,
      "epoch": 1.2917300862506342,
      "grad_norm": 5.0625,
      "learning_rate": 3.054320358900027e-06,
      "loss": 1.62173042,
      "memory(GiB)": 117.38,
      "step": 50920,
      "train_speed(iter/s)": 1.638417
    },
    {
      "acc": 0.64072666,
      "epoch": 1.2918569254185692,
      "grad_norm": 5.6875,
      "learning_rate": 3.053354430173411e-06,
      "loss": 1.66871567,
      "memory(GiB)": 117.38,
      "step": 50925,
      "train_speed(iter/s)": 1.638435
    },
    {
      "acc": 0.67633533,
      "epoch": 1.2919837645865044,
      "grad_norm": 7.8125,
      "learning_rate": 3.0523885870689595e-06,
      "loss": 1.53640633,
      "memory(GiB)": 117.38,
      "step": 50930,
      "train_speed(iter/s)": 1.638454
    },
    {
      "acc": 0.66682391,
      "epoch": 1.2921106037544394,
      "grad_norm": 6.75,
      "learning_rate": 3.051422829629152e-06,
      "loss": 1.53191719,
      "memory(GiB)": 117.38,
      "step": 50935,
      "train_speed(iter/s)": 1.638472
    },
    {
      "acc": 0.65137725,
      "epoch": 1.2922374429223744,
      "grad_norm": 5.21875,
      "learning_rate": 3.050457157896467e-06,
      "loss": 1.6059185,
      "memory(GiB)": 117.38,
      "step": 50940,
      "train_speed(iter/s)": 1.63849
    },
    {
      "acc": 0.65863485,
      "epoch": 1.2923642820903094,
      "grad_norm": 4.75,
      "learning_rate": 3.049491571913382e-06,
      "loss": 1.56069822,
      "memory(GiB)": 117.38,
      "step": 50945,
      "train_speed(iter/s)": 1.638508
    },
    {
      "acc": 0.64616456,
      "epoch": 1.2924911212582446,
      "grad_norm": 5.40625,
      "learning_rate": 3.048526071722367e-06,
      "loss": 1.61379967,
      "memory(GiB)": 117.38,
      "step": 50950,
      "train_speed(iter/s)": 1.638527
    },
    {
      "acc": 0.65950646,
      "epoch": 1.2926179604261796,
      "grad_norm": 6.125,
      "learning_rate": 3.047560657365886e-06,
      "loss": 1.59002857,
      "memory(GiB)": 117.38,
      "step": 50955,
      "train_speed(iter/s)": 1.638546
    },
    {
      "acc": 0.63919864,
      "epoch": 1.2927447995941146,
      "grad_norm": 5.71875,
      "learning_rate": 3.046595328886405e-06,
      "loss": 1.69314194,
      "memory(GiB)": 117.38,
      "step": 50960,
      "train_speed(iter/s)": 1.638565
    },
    {
      "acc": 0.65888104,
      "epoch": 1.2928716387620498,
      "grad_norm": 5.53125,
      "learning_rate": 3.0456300863263815e-06,
      "loss": 1.5710393,
      "memory(GiB)": 117.38,
      "step": 50965,
      "train_speed(iter/s)": 1.638583
    },
    {
      "acc": 0.65923815,
      "epoch": 1.2929984779299848,
      "grad_norm": 5.0625,
      "learning_rate": 3.044664929728276e-06,
      "loss": 1.55587101,
      "memory(GiB)": 117.38,
      "step": 50970,
      "train_speed(iter/s)": 1.638602
    },
    {
      "acc": 0.67046094,
      "epoch": 1.2931253170979198,
      "grad_norm": 6.0625,
      "learning_rate": 3.0436998591345336e-06,
      "loss": 1.5219985,
      "memory(GiB)": 117.38,
      "step": 50975,
      "train_speed(iter/s)": 1.63862
    },
    {
      "acc": 0.64892979,
      "epoch": 1.2932521562658548,
      "grad_norm": 8.0,
      "learning_rate": 3.042734874587607e-06,
      "loss": 1.6151371,
      "memory(GiB)": 117.38,
      "step": 50980,
      "train_speed(iter/s)": 1.638638
    },
    {
      "acc": 0.64846973,
      "epoch": 1.29337899543379,
      "grad_norm": 5.46875,
      "learning_rate": 3.041769976129938e-06,
      "loss": 1.56504183,
      "memory(GiB)": 117.38,
      "step": 50985,
      "train_speed(iter/s)": 1.638656
    },
    {
      "acc": 0.66654482,
      "epoch": 1.293505834601725,
      "grad_norm": 5.375,
      "learning_rate": 3.0408051638039697e-06,
      "loss": 1.59809179,
      "memory(GiB)": 117.38,
      "step": 50990,
      "train_speed(iter/s)": 1.638672
    },
    {
      "acc": 0.63057394,
      "epoch": 1.2936326737696602,
      "grad_norm": 9.0,
      "learning_rate": 3.039840437652137e-06,
      "loss": 1.69659061,
      "memory(GiB)": 117.38,
      "step": 50995,
      "train_speed(iter/s)": 1.63869
    },
    {
      "acc": 0.65516634,
      "epoch": 1.2937595129375952,
      "grad_norm": 5.9375,
      "learning_rate": 3.0388757977168724e-06,
      "loss": 1.63966293,
      "memory(GiB)": 117.38,
      "step": 51000,
      "train_speed(iter/s)": 1.638707
    },
    {
      "epoch": 1.2937595129375952,
      "eval_acc": 0.6462405269213982,
      "eval_loss": 1.573349118232727,
      "eval_runtime": 58.8372,
      "eval_samples_per_second": 108.265,
      "eval_steps_per_second": 27.075,
      "step": 51000
    },
    {
      "acc": 0.66557379,
      "epoch": 1.2938863521055302,
      "grad_norm": 6.84375,
      "learning_rate": 3.0379112440406066e-06,
      "loss": 1.547052,
      "memory(GiB)": 117.38,
      "step": 51005,
      "train_speed(iter/s)": 1.635404
    },
    {
      "acc": 0.6597537,
      "epoch": 1.2940131912734651,
      "grad_norm": 4.96875,
      "learning_rate": 3.036946776665766e-06,
      "loss": 1.57173166,
      "memory(GiB)": 117.38,
      "step": 51010,
      "train_speed(iter/s)": 1.63542
    },
    {
      "acc": 0.65995388,
      "epoch": 1.2941400304414004,
      "grad_norm": 6.125,
      "learning_rate": 3.0359823956347695e-06,
      "loss": 1.56989346,
      "memory(GiB)": 117.38,
      "step": 51015,
      "train_speed(iter/s)": 1.635435
    },
    {
      "acc": 0.65556507,
      "epoch": 1.2942668696093353,
      "grad_norm": 4.875,
      "learning_rate": 3.0350181009900347e-06,
      "loss": 1.68399258,
      "memory(GiB)": 117.38,
      "step": 51020,
      "train_speed(iter/s)": 1.635452
    },
    {
      "acc": 0.67168455,
      "epoch": 1.2943937087772706,
      "grad_norm": 5.6875,
      "learning_rate": 3.0340538927739784e-06,
      "loss": 1.53910589,
      "memory(GiB)": 117.38,
      "step": 51025,
      "train_speed(iter/s)": 1.635468
    },
    {
      "acc": 0.67344546,
      "epoch": 1.2945205479452055,
      "grad_norm": 5.5,
      "learning_rate": 3.0330897710290093e-06,
      "loss": 1.60014553,
      "memory(GiB)": 117.38,
      "step": 51030,
      "train_speed(iter/s)": 1.635485
    },
    {
      "acc": 0.64574738,
      "epoch": 1.2946473871131405,
      "grad_norm": 5.4375,
      "learning_rate": 3.032125735797532e-06,
      "loss": 1.60611382,
      "memory(GiB)": 117.38,
      "step": 51035,
      "train_speed(iter/s)": 1.6355
    },
    {
      "acc": 0.64362578,
      "epoch": 1.2947742262810755,
      "grad_norm": 6.40625,
      "learning_rate": 3.031161787121952e-06,
      "loss": 1.67409859,
      "memory(GiB)": 117.38,
      "step": 51040,
      "train_speed(iter/s)": 1.635516
    },
    {
      "acc": 0.67234669,
      "epoch": 1.2949010654490107,
      "grad_norm": 4.71875,
      "learning_rate": 3.0301979250446655e-06,
      "loss": 1.63208065,
      "memory(GiB)": 117.38,
      "step": 51045,
      "train_speed(iter/s)": 1.635533
    },
    {
      "acc": 0.68056087,
      "epoch": 1.2950279046169457,
      "grad_norm": 9.1875,
      "learning_rate": 3.029234149608071e-06,
      "loss": 1.54029303,
      "memory(GiB)": 117.38,
      "step": 51050,
      "train_speed(iter/s)": 1.635548
    },
    {
      "acc": 0.668679,
      "epoch": 1.2951547437848807,
      "grad_norm": 5.9375,
      "learning_rate": 3.0282704608545566e-06,
      "loss": 1.55561857,
      "memory(GiB)": 117.38,
      "step": 51055,
      "train_speed(iter/s)": 1.635565
    },
    {
      "acc": 0.66406183,
      "epoch": 1.295281582952816,
      "grad_norm": 5.8125,
      "learning_rate": 3.0273068588265097e-06,
      "loss": 1.59801769,
      "memory(GiB)": 117.38,
      "step": 51060,
      "train_speed(iter/s)": 1.635581
    },
    {
      "acc": 0.64535794,
      "epoch": 1.295408422120751,
      "grad_norm": 4.84375,
      "learning_rate": 3.0263433435663143e-06,
      "loss": 1.64406738,
      "memory(GiB)": 117.38,
      "step": 51065,
      "train_speed(iter/s)": 1.635595
    },
    {
      "acc": 0.64878621,
      "epoch": 1.295535261288686,
      "grad_norm": 5.03125,
      "learning_rate": 3.0253799151163522e-06,
      "loss": 1.63740387,
      "memory(GiB)": 117.38,
      "step": 51070,
      "train_speed(iter/s)": 1.635612
    },
    {
      "acc": 0.64827609,
      "epoch": 1.295662100456621,
      "grad_norm": 5.46875,
      "learning_rate": 3.0244165735189967e-06,
      "loss": 1.59974203,
      "memory(GiB)": 117.38,
      "step": 51075,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.64578218,
      "epoch": 1.295788939624556,
      "grad_norm": 5.40625,
      "learning_rate": 3.023453318816619e-06,
      "loss": 1.65839882,
      "memory(GiB)": 117.38,
      "step": 51080,
      "train_speed(iter/s)": 1.635643
    },
    {
      "acc": 0.64653873,
      "epoch": 1.295915778792491,
      "grad_norm": 6.625,
      "learning_rate": 3.022490151051591e-06,
      "loss": 1.58992701,
      "memory(GiB)": 117.38,
      "step": 51085,
      "train_speed(iter/s)": 1.635658
    },
    {
      "acc": 0.65987306,
      "epoch": 1.2960426179604263,
      "grad_norm": 8.3125,
      "learning_rate": 3.0215270702662753e-06,
      "loss": 1.61896954,
      "memory(GiB)": 117.38,
      "step": 51090,
      "train_speed(iter/s)": 1.635674
    },
    {
      "acc": 0.67181969,
      "epoch": 1.2961694571283613,
      "grad_norm": 5.4375,
      "learning_rate": 3.020564076503031e-06,
      "loss": 1.50718708,
      "memory(GiB)": 117.38,
      "step": 51095,
      "train_speed(iter/s)": 1.635689
    },
    {
      "acc": 0.65677891,
      "epoch": 1.2962962962962963,
      "grad_norm": 4.9375,
      "learning_rate": 3.019601169804216e-06,
      "loss": 1.60704212,
      "memory(GiB)": 117.38,
      "step": 51100,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.66070251,
      "epoch": 1.2964231354642313,
      "grad_norm": 4.65625,
      "learning_rate": 3.018638350212184e-06,
      "loss": 1.6213335,
      "memory(GiB)": 117.38,
      "step": 51105,
      "train_speed(iter/s)": 1.635721
    },
    {
      "acc": 0.65107613,
      "epoch": 1.2965499746321665,
      "grad_norm": 5.09375,
      "learning_rate": 3.0176756177692845e-06,
      "loss": 1.57734585,
      "memory(GiB)": 117.38,
      "step": 51110,
      "train_speed(iter/s)": 1.635738
    },
    {
      "acc": 0.65010014,
      "epoch": 1.2966768138001015,
      "grad_norm": 5.625,
      "learning_rate": 3.01671297251786e-06,
      "loss": 1.63011723,
      "memory(GiB)": 117.38,
      "step": 51115,
      "train_speed(iter/s)": 1.635754
    },
    {
      "acc": 0.6633954,
      "epoch": 1.2968036529680365,
      "grad_norm": 5.21875,
      "learning_rate": 3.0157504145002546e-06,
      "loss": 1.56812382,
      "memory(GiB)": 117.38,
      "step": 51120,
      "train_speed(iter/s)": 1.63577
    },
    {
      "acc": 0.64187508,
      "epoch": 1.2969304921359717,
      "grad_norm": 5.625,
      "learning_rate": 3.0147879437588046e-06,
      "loss": 1.62867241,
      "memory(GiB)": 117.38,
      "step": 51125,
      "train_speed(iter/s)": 1.635786
    },
    {
      "acc": 0.65675321,
      "epoch": 1.2970573313039067,
      "grad_norm": 7.375,
      "learning_rate": 3.013825560335845e-06,
      "loss": 1.60088959,
      "memory(GiB)": 117.38,
      "step": 51130,
      "train_speed(iter/s)": 1.635803
    },
    {
      "acc": 0.65545635,
      "epoch": 1.2971841704718416,
      "grad_norm": 7.1875,
      "learning_rate": 3.0128632642737044e-06,
      "loss": 1.56130943,
      "memory(GiB)": 117.38,
      "step": 51135,
      "train_speed(iter/s)": 1.63582
    },
    {
      "acc": 0.67316637,
      "epoch": 1.2973110096397766,
      "grad_norm": 6.9375,
      "learning_rate": 3.0119010556147088e-06,
      "loss": 1.57268896,
      "memory(GiB)": 117.38,
      "step": 51140,
      "train_speed(iter/s)": 1.635835
    },
    {
      "acc": 0.6693522,
      "epoch": 1.2974378488077118,
      "grad_norm": 5.46875,
      "learning_rate": 3.0109389344011813e-06,
      "loss": 1.5877161,
      "memory(GiB)": 117.38,
      "step": 51145,
      "train_speed(iter/s)": 1.635851
    },
    {
      "acc": 0.65841799,
      "epoch": 1.2975646879756468,
      "grad_norm": 5.6875,
      "learning_rate": 3.0099769006754415e-06,
      "loss": 1.57503414,
      "memory(GiB)": 117.38,
      "step": 51150,
      "train_speed(iter/s)": 1.635869
    },
    {
      "acc": 0.65028243,
      "epoch": 1.297691527143582,
      "grad_norm": 5.625,
      "learning_rate": 3.0090149544798007e-06,
      "loss": 1.5683279,
      "memory(GiB)": 117.38,
      "step": 51155,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.64297285,
      "epoch": 1.297818366311517,
      "grad_norm": 6.875,
      "learning_rate": 3.0080530958565712e-06,
      "loss": 1.63781223,
      "memory(GiB)": 117.38,
      "step": 51160,
      "train_speed(iter/s)": 1.6359
    },
    {
      "acc": 0.66097555,
      "epoch": 1.297945205479452,
      "grad_norm": 6.5625,
      "learning_rate": 3.0070913248480602e-06,
      "loss": 1.58337421,
      "memory(GiB)": 117.38,
      "step": 51165,
      "train_speed(iter/s)": 1.635916
    },
    {
      "acc": 0.66105242,
      "epoch": 1.298072044647387,
      "grad_norm": 5.4375,
      "learning_rate": 3.0061296414965724e-06,
      "loss": 1.59135027,
      "memory(GiB)": 117.38,
      "step": 51170,
      "train_speed(iter/s)": 1.635933
    },
    {
      "acc": 0.67142873,
      "epoch": 1.2981988838153222,
      "grad_norm": 4.6875,
      "learning_rate": 3.005168045844402e-06,
      "loss": 1.58040609,
      "memory(GiB)": 117.38,
      "step": 51175,
      "train_speed(iter/s)": 1.635949
    },
    {
      "acc": 0.65379391,
      "epoch": 1.2983257229832572,
      "grad_norm": 6.0625,
      "learning_rate": 3.0042065379338486e-06,
      "loss": 1.61228695,
      "memory(GiB)": 117.38,
      "step": 51180,
      "train_speed(iter/s)": 1.635966
    },
    {
      "acc": 0.65575447,
      "epoch": 1.2984525621511924,
      "grad_norm": 6.96875,
      "learning_rate": 3.003245117807201e-06,
      "loss": 1.61400337,
      "memory(GiB)": 117.38,
      "step": 51185,
      "train_speed(iter/s)": 1.635983
    },
    {
      "acc": 0.66946974,
      "epoch": 1.2985794013191274,
      "grad_norm": 5.3125,
      "learning_rate": 3.0022837855067514e-06,
      "loss": 1.51799068,
      "memory(GiB)": 117.38,
      "step": 51190,
      "train_speed(iter/s)": 1.636
    },
    {
      "acc": 0.6639267,
      "epoch": 1.2987062404870624,
      "grad_norm": 5.0625,
      "learning_rate": 3.0013225410747772e-06,
      "loss": 1.53750954,
      "memory(GiB)": 117.38,
      "step": 51195,
      "train_speed(iter/s)": 1.636015
    },
    {
      "acc": 0.64466162,
      "epoch": 1.2988330796549974,
      "grad_norm": 6.09375,
      "learning_rate": 3.0003613845535617e-06,
      "loss": 1.62723694,
      "memory(GiB)": 117.38,
      "step": 51200,
      "train_speed(iter/s)": 1.636033
    },
    {
      "acc": 0.65698643,
      "epoch": 1.2989599188229326,
      "grad_norm": 6.4375,
      "learning_rate": 2.9994003159853793e-06,
      "loss": 1.61211662,
      "memory(GiB)": 117.38,
      "step": 51205,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.65932074,
      "epoch": 1.2990867579908676,
      "grad_norm": 5.8125,
      "learning_rate": 2.998439335412505e-06,
      "loss": 1.54120035,
      "memory(GiB)": 117.38,
      "step": 51210,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.65672312,
      "epoch": 1.2992135971588026,
      "grad_norm": 4.6875,
      "learning_rate": 2.9974784428772043e-06,
      "loss": 1.56844044,
      "memory(GiB)": 117.38,
      "step": 51215,
      "train_speed(iter/s)": 1.636079
    },
    {
      "acc": 0.66167655,
      "epoch": 1.2993404363267378,
      "grad_norm": 5.3125,
      "learning_rate": 2.996517638421741e-06,
      "loss": 1.61957092,
      "memory(GiB)": 117.38,
      "step": 51220,
      "train_speed(iter/s)": 1.636097
    },
    {
      "acc": 0.658776,
      "epoch": 1.2994672754946728,
      "grad_norm": 5.375,
      "learning_rate": 2.9955569220883777e-06,
      "loss": 1.58248234,
      "memory(GiB)": 117.38,
      "step": 51225,
      "train_speed(iter/s)": 1.636112
    },
    {
      "acc": 0.6524682,
      "epoch": 1.2995941146626078,
      "grad_norm": 6.28125,
      "learning_rate": 2.9945962939193718e-06,
      "loss": 1.6409996,
      "memory(GiB)": 117.38,
      "step": 51230,
      "train_speed(iter/s)": 1.636128
    },
    {
      "acc": 0.66665049,
      "epoch": 1.2997209538305428,
      "grad_norm": 4.96875,
      "learning_rate": 2.9936357539569728e-06,
      "loss": 1.54585724,
      "memory(GiB)": 117.38,
      "step": 51235,
      "train_speed(iter/s)": 1.636146
    },
    {
      "acc": 0.6669415,
      "epoch": 1.299847792998478,
      "grad_norm": 6.1875,
      "learning_rate": 2.9926753022434306e-06,
      "loss": 1.53547039,
      "memory(GiB)": 117.38,
      "step": 51240,
      "train_speed(iter/s)": 1.63616
    },
    {
      "acc": 0.67339115,
      "epoch": 1.299974632166413,
      "grad_norm": 5.25,
      "learning_rate": 2.9917149388209908e-06,
      "loss": 1.58321886,
      "memory(GiB)": 117.38,
      "step": 51245,
      "train_speed(iter/s)": 1.636175
    },
    {
      "acc": 0.65516682,
      "epoch": 1.3001014713343482,
      "grad_norm": 5.15625,
      "learning_rate": 2.9907546637318964e-06,
      "loss": 1.6069809,
      "memory(GiB)": 117.38,
      "step": 51250,
      "train_speed(iter/s)": 1.636192
    },
    {
      "acc": 0.65986772,
      "epoch": 1.3002283105022832,
      "grad_norm": 8.25,
      "learning_rate": 2.98979447701838e-06,
      "loss": 1.60593185,
      "memory(GiB)": 117.38,
      "step": 51255,
      "train_speed(iter/s)": 1.636209
    },
    {
      "acc": 0.66070147,
      "epoch": 1.3003551496702181,
      "grad_norm": 10.1875,
      "learning_rate": 2.988834378722679e-06,
      "loss": 1.61141033,
      "memory(GiB)": 117.38,
      "step": 51260,
      "train_speed(iter/s)": 1.636227
    },
    {
      "acc": 0.65299373,
      "epoch": 1.3004819888381531,
      "grad_norm": 5.0625,
      "learning_rate": 2.9878743688870193e-06,
      "loss": 1.58934889,
      "memory(GiB)": 117.38,
      "step": 51265,
      "train_speed(iter/s)": 1.636243
    },
    {
      "acc": 0.65360303,
      "epoch": 1.3006088280060883,
      "grad_norm": 5.03125,
      "learning_rate": 2.9869144475536306e-06,
      "loss": 1.61109886,
      "memory(GiB)": 117.38,
      "step": 51270,
      "train_speed(iter/s)": 1.63626
    },
    {
      "acc": 0.64588294,
      "epoch": 1.3007356671740233,
      "grad_norm": 5.1875,
      "learning_rate": 2.9859546147647316e-06,
      "loss": 1.69761391,
      "memory(GiB)": 117.38,
      "step": 51275,
      "train_speed(iter/s)": 1.636276
    },
    {
      "acc": 0.650596,
      "epoch": 1.3008625063419583,
      "grad_norm": 5.4375,
      "learning_rate": 2.984994870562541e-06,
      "loss": 1.64292259,
      "memory(GiB)": 117.38,
      "step": 51280,
      "train_speed(iter/s)": 1.636291
    },
    {
      "acc": 0.65538807,
      "epoch": 1.3009893455098935,
      "grad_norm": 5.375,
      "learning_rate": 2.9840352149892703e-06,
      "loss": 1.59412556,
      "memory(GiB)": 117.38,
      "step": 51285,
      "train_speed(iter/s)": 1.636307
    },
    {
      "acc": 0.67346067,
      "epoch": 1.3011161846778285,
      "grad_norm": 5.6875,
      "learning_rate": 2.9830756480871343e-06,
      "loss": 1.61384506,
      "memory(GiB)": 117.38,
      "step": 51290,
      "train_speed(iter/s)": 1.636323
    },
    {
      "acc": 0.65672727,
      "epoch": 1.3012430238457635,
      "grad_norm": 5.6875,
      "learning_rate": 2.9821161698983347e-06,
      "loss": 1.62566128,
      "memory(GiB)": 117.38,
      "step": 51295,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65628948,
      "epoch": 1.3013698630136985,
      "grad_norm": 5.78125,
      "learning_rate": 2.9811567804650747e-06,
      "loss": 1.59102726,
      "memory(GiB)": 117.38,
      "step": 51300,
      "train_speed(iter/s)": 1.636359
    },
    {
      "acc": 0.65287066,
      "epoch": 1.3014967021816337,
      "grad_norm": 5.96875,
      "learning_rate": 2.980197479829554e-06,
      "loss": 1.59878273,
      "memory(GiB)": 117.38,
      "step": 51305,
      "train_speed(iter/s)": 1.636376
    },
    {
      "acc": 0.66711597,
      "epoch": 1.3016235413495687,
      "grad_norm": 6.65625,
      "learning_rate": 2.9792382680339666e-06,
      "loss": 1.55396156,
      "memory(GiB)": 117.38,
      "step": 51310,
      "train_speed(iter/s)": 1.636393
    },
    {
      "acc": 0.64026814,
      "epoch": 1.301750380517504,
      "grad_norm": 6.03125,
      "learning_rate": 2.9782791451205e-06,
      "loss": 1.62613811,
      "memory(GiB)": 117.38,
      "step": 51315,
      "train_speed(iter/s)": 1.636409
    },
    {
      "acc": 0.63746824,
      "epoch": 1.301877219685439,
      "grad_norm": 6.96875,
      "learning_rate": 2.9773201111313444e-06,
      "loss": 1.66666603,
      "memory(GiB)": 117.38,
      "step": 51320,
      "train_speed(iter/s)": 1.636427
    },
    {
      "acc": 0.66756029,
      "epoch": 1.302004058853374,
      "grad_norm": 6.25,
      "learning_rate": 2.9763611661086806e-06,
      "loss": 1.58938808,
      "memory(GiB)": 117.38,
      "step": 51325,
      "train_speed(iter/s)": 1.636443
    },
    {
      "acc": 0.66998334,
      "epoch": 1.3021308980213089,
      "grad_norm": 6.34375,
      "learning_rate": 2.975402310094689e-06,
      "loss": 1.55547972,
      "memory(GiB)": 117.38,
      "step": 51330,
      "train_speed(iter/s)": 1.636461
    },
    {
      "acc": 0.64836669,
      "epoch": 1.302257737189244,
      "grad_norm": 4.71875,
      "learning_rate": 2.9744435431315403e-06,
      "loss": 1.61106319,
      "memory(GiB)": 117.38,
      "step": 51335,
      "train_speed(iter/s)": 1.636476
    },
    {
      "acc": 0.65225735,
      "epoch": 1.302384576357179,
      "grad_norm": 7.40625,
      "learning_rate": 2.9734848652614097e-06,
      "loss": 1.69089603,
      "memory(GiB)": 117.38,
      "step": 51340,
      "train_speed(iter/s)": 1.636494
    },
    {
      "acc": 0.66204529,
      "epoch": 1.3025114155251143,
      "grad_norm": 6.46875,
      "learning_rate": 2.972526276526461e-06,
      "loss": 1.60239105,
      "memory(GiB)": 117.38,
      "step": 51345,
      "train_speed(iter/s)": 1.63651
    },
    {
      "acc": 0.66751919,
      "epoch": 1.3026382546930493,
      "grad_norm": 6.21875,
      "learning_rate": 2.971567776968861e-06,
      "loss": 1.55322838,
      "memory(GiB)": 117.38,
      "step": 51350,
      "train_speed(iter/s)": 1.636527
    },
    {
      "acc": 0.67617989,
      "epoch": 1.3027650938609843,
      "grad_norm": 5.5,
      "learning_rate": 2.9706093666307654e-06,
      "loss": 1.56108465,
      "memory(GiB)": 117.38,
      "step": 51355,
      "train_speed(iter/s)": 1.636542
    },
    {
      "acc": 0.65380955,
      "epoch": 1.3028919330289193,
      "grad_norm": 5.25,
      "learning_rate": 2.969651045554329e-06,
      "loss": 1.57948246,
      "memory(GiB)": 117.38,
      "step": 51360,
      "train_speed(iter/s)": 1.63656
    },
    {
      "acc": 0.65414896,
      "epoch": 1.3030187721968545,
      "grad_norm": 5.3125,
      "learning_rate": 2.968692813781706e-06,
      "loss": 1.61331062,
      "memory(GiB)": 117.38,
      "step": 51365,
      "train_speed(iter/s)": 1.636576
    },
    {
      "acc": 0.6445148,
      "epoch": 1.3031456113647895,
      "grad_norm": 5.9375,
      "learning_rate": 2.9677346713550437e-06,
      "loss": 1.61052322,
      "memory(GiB)": 117.38,
      "step": 51370,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.67251377,
      "epoch": 1.3032724505327244,
      "grad_norm": 4.1875,
      "learning_rate": 2.966776618316482e-06,
      "loss": 1.51120205,
      "memory(GiB)": 117.38,
      "step": 51375,
      "train_speed(iter/s)": 1.636611
    },
    {
      "acc": 0.64410119,
      "epoch": 1.3033992897006597,
      "grad_norm": 5.75,
      "learning_rate": 2.9658186547081612e-06,
      "loss": 1.64027081,
      "memory(GiB)": 117.38,
      "step": 51380,
      "train_speed(iter/s)": 1.636626
    },
    {
      "acc": 0.67041311,
      "epoch": 1.3035261288685946,
      "grad_norm": 6.1875,
      "learning_rate": 2.9648607805722197e-06,
      "loss": 1.56244469,
      "memory(GiB)": 117.38,
      "step": 51385,
      "train_speed(iter/s)": 1.636642
    },
    {
      "acc": 0.65798521,
      "epoch": 1.3036529680365296,
      "grad_norm": 5.53125,
      "learning_rate": 2.963902995950788e-06,
      "loss": 1.5554759,
      "memory(GiB)": 117.38,
      "step": 51390,
      "train_speed(iter/s)": 1.636659
    },
    {
      "acc": 0.64611349,
      "epoch": 1.3037798072044646,
      "grad_norm": 6.25,
      "learning_rate": 2.962945300885991e-06,
      "loss": 1.65860405,
      "memory(GiB)": 117.38,
      "step": 51395,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.65637007,
      "epoch": 1.3039066463723998,
      "grad_norm": 6.71875,
      "learning_rate": 2.9619876954199557e-06,
      "loss": 1.57655602,
      "memory(GiB)": 117.38,
      "step": 51400,
      "train_speed(iter/s)": 1.636692
    },
    {
      "acc": 0.67129002,
      "epoch": 1.3040334855403348,
      "grad_norm": 4.90625,
      "learning_rate": 2.9610301795947992e-06,
      "loss": 1.54279785,
      "memory(GiB)": 117.38,
      "step": 51405,
      "train_speed(iter/s)": 1.636707
    },
    {
      "acc": 0.65228286,
      "epoch": 1.30416032470827,
      "grad_norm": 6.53125,
      "learning_rate": 2.9600727534526417e-06,
      "loss": 1.64160862,
      "memory(GiB)": 117.38,
      "step": 51410,
      "train_speed(iter/s)": 1.636724
    },
    {
      "acc": 0.65549097,
      "epoch": 1.304287163876205,
      "grad_norm": 6.25,
      "learning_rate": 2.9591154170355895e-06,
      "loss": 1.67537041,
      "memory(GiB)": 117.38,
      "step": 51415,
      "train_speed(iter/s)": 1.63674
    },
    {
      "acc": 0.64177289,
      "epoch": 1.30441400304414,
      "grad_norm": 5.5,
      "learning_rate": 2.9581581703857545e-06,
      "loss": 1.66877747,
      "memory(GiB)": 117.38,
      "step": 51420,
      "train_speed(iter/s)": 1.636757
    },
    {
      "acc": 0.65517545,
      "epoch": 1.304540842212075,
      "grad_norm": 6.15625,
      "learning_rate": 2.9572010135452377e-06,
      "loss": 1.57466717,
      "memory(GiB)": 117.38,
      "step": 51425,
      "train_speed(iter/s)": 1.636772
    },
    {
      "acc": 0.65859098,
      "epoch": 1.3046676813800102,
      "grad_norm": 8.625,
      "learning_rate": 2.9562439465561425e-06,
      "loss": 1.63903542,
      "memory(GiB)": 117.38,
      "step": 51430,
      "train_speed(iter/s)": 1.636789
    },
    {
      "acc": 0.66389856,
      "epoch": 1.3047945205479452,
      "grad_norm": 6.40625,
      "learning_rate": 2.955286969460563e-06,
      "loss": 1.55800343,
      "memory(GiB)": 117.38,
      "step": 51435,
      "train_speed(iter/s)": 1.636804
    },
    {
      "acc": 0.64793568,
      "epoch": 1.3049213597158802,
      "grad_norm": 6.40625,
      "learning_rate": 2.9543300823005903e-06,
      "loss": 1.67616043,
      "memory(GiB)": 117.38,
      "step": 51440,
      "train_speed(iter/s)": 1.63682
    },
    {
      "acc": 0.6390697,
      "epoch": 1.3050481988838154,
      "grad_norm": 5.65625,
      "learning_rate": 2.953373285118315e-06,
      "loss": 1.63063946,
      "memory(GiB)": 117.38,
      "step": 51445,
      "train_speed(iter/s)": 1.636836
    },
    {
      "acc": 0.65421939,
      "epoch": 1.3051750380517504,
      "grad_norm": 6.15625,
      "learning_rate": 2.9524165779558206e-06,
      "loss": 1.65491867,
      "memory(GiB)": 117.38,
      "step": 51450,
      "train_speed(iter/s)": 1.636853
    },
    {
      "acc": 0.6562438,
      "epoch": 1.3053018772196854,
      "grad_norm": 5.78125,
      "learning_rate": 2.9514599608551865e-06,
      "loss": 1.60461826,
      "memory(GiB)": 117.38,
      "step": 51455,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.64334412,
      "epoch": 1.3054287163876204,
      "grad_norm": 5.1875,
      "learning_rate": 2.9505034338584882e-06,
      "loss": 1.72497978,
      "memory(GiB)": 117.38,
      "step": 51460,
      "train_speed(iter/s)": 1.636884
    },
    {
      "acc": 0.67794304,
      "epoch": 1.3055555555555556,
      "grad_norm": 5.875,
      "learning_rate": 2.9495469970078e-06,
      "loss": 1.58000813,
      "memory(GiB)": 117.38,
      "step": 51465,
      "train_speed(iter/s)": 1.6369
    },
    {
      "acc": 0.64928265,
      "epoch": 1.3056823947234906,
      "grad_norm": 5.6875,
      "learning_rate": 2.9485906503451907e-06,
      "loss": 1.66637421,
      "memory(GiB)": 117.38,
      "step": 51470,
      "train_speed(iter/s)": 1.636916
    },
    {
      "acc": 0.64091649,
      "epoch": 1.3058092338914258,
      "grad_norm": 5.21875,
      "learning_rate": 2.9476343939127217e-06,
      "loss": 1.62044888,
      "memory(GiB)": 117.38,
      "step": 51475,
      "train_speed(iter/s)": 1.636931
    },
    {
      "acc": 0.67019229,
      "epoch": 1.3059360730593608,
      "grad_norm": 5.625,
      "learning_rate": 2.9466782277524554e-06,
      "loss": 1.48893566,
      "memory(GiB)": 117.38,
      "step": 51480,
      "train_speed(iter/s)": 1.636948
    },
    {
      "acc": 0.65354128,
      "epoch": 1.3060629122272958,
      "grad_norm": 9.5625,
      "learning_rate": 2.9457221519064477e-06,
      "loss": 1.64324951,
      "memory(GiB)": 117.38,
      "step": 51485,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.66596785,
      "epoch": 1.3061897513952307,
      "grad_norm": 5.6875,
      "learning_rate": 2.944766166416754e-06,
      "loss": 1.57936859,
      "memory(GiB)": 117.38,
      "step": 51490,
      "train_speed(iter/s)": 1.636981
    },
    {
      "acc": 0.65529156,
      "epoch": 1.306316590563166,
      "grad_norm": 5.21875,
      "learning_rate": 2.943810271325418e-06,
      "loss": 1.59710217,
      "memory(GiB)": 117.38,
      "step": 51495,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.64954548,
      "epoch": 1.306443429731101,
      "grad_norm": 5.78125,
      "learning_rate": 2.9428544666744873e-06,
      "loss": 1.57202291,
      "memory(GiB)": 117.38,
      "step": 51500,
      "train_speed(iter/s)": 1.637012
    },
    {
      "acc": 0.63954611,
      "epoch": 1.3065702688990362,
      "grad_norm": 5.96875,
      "learning_rate": 2.9418987525060004e-06,
      "loss": 1.74219055,
      "memory(GiB)": 117.38,
      "step": 51505,
      "train_speed(iter/s)": 1.637028
    },
    {
      "acc": 0.6643631,
      "epoch": 1.3066971080669711,
      "grad_norm": 7.1875,
      "learning_rate": 2.9409431288619973e-06,
      "loss": 1.60146503,
      "memory(GiB)": 117.38,
      "step": 51510,
      "train_speed(iter/s)": 1.637044
    },
    {
      "acc": 0.65021467,
      "epoch": 1.3068239472349061,
      "grad_norm": 5.5,
      "learning_rate": 2.939987595784507e-06,
      "loss": 1.6500515,
      "memory(GiB)": 117.38,
      "step": 51515,
      "train_speed(iter/s)": 1.637061
    },
    {
      "acc": 0.6537488,
      "epoch": 1.3069507864028411,
      "grad_norm": 5.59375,
      "learning_rate": 2.9390321533155585e-06,
      "loss": 1.68441391,
      "memory(GiB)": 117.38,
      "step": 51520,
      "train_speed(iter/s)": 1.637076
    },
    {
      "acc": 0.66419053,
      "epoch": 1.3070776255707763,
      "grad_norm": 5.375,
      "learning_rate": 2.9380768014971794e-06,
      "loss": 1.63850346,
      "memory(GiB)": 117.38,
      "step": 51525,
      "train_speed(iter/s)": 1.637094
    },
    {
      "acc": 0.67292109,
      "epoch": 1.3072044647387113,
      "grad_norm": 6.21875,
      "learning_rate": 2.937121540371389e-06,
      "loss": 1.52047367,
      "memory(GiB)": 117.38,
      "step": 51530,
      "train_speed(iter/s)": 1.637111
    },
    {
      "acc": 0.66494627,
      "epoch": 1.3073313039066463,
      "grad_norm": 7.5,
      "learning_rate": 2.9361663699802007e-06,
      "loss": 1.60929298,
      "memory(GiB)": 117.38,
      "step": 51535,
      "train_speed(iter/s)": 1.637126
    },
    {
      "acc": 0.65330281,
      "epoch": 1.3074581430745815,
      "grad_norm": 5.90625,
      "learning_rate": 2.9352112903656315e-06,
      "loss": 1.59436474,
      "memory(GiB)": 117.38,
      "step": 51540,
      "train_speed(iter/s)": 1.637144
    },
    {
      "acc": 0.65280647,
      "epoch": 1.3075849822425165,
      "grad_norm": 5.15625,
      "learning_rate": 2.9342563015696866e-06,
      "loss": 1.68462639,
      "memory(GiB)": 117.38,
      "step": 51545,
      "train_speed(iter/s)": 1.637159
    },
    {
      "acc": 0.67475739,
      "epoch": 1.3077118214104515,
      "grad_norm": 6.28125,
      "learning_rate": 2.9333014036343765e-06,
      "loss": 1.53416767,
      "memory(GiB)": 117.38,
      "step": 51550,
      "train_speed(iter/s)": 1.637177
    },
    {
      "acc": 0.65203185,
      "epoch": 1.3078386605783865,
      "grad_norm": 5.3125,
      "learning_rate": 2.932346596601694e-06,
      "loss": 1.5928648,
      "memory(GiB)": 117.38,
      "step": 51555,
      "train_speed(iter/s)": 1.637192
    },
    {
      "acc": 0.64538231,
      "epoch": 1.3079654997463217,
      "grad_norm": 5.46875,
      "learning_rate": 2.931391880513641e-06,
      "loss": 1.5798933,
      "memory(GiB)": 117.38,
      "step": 51560,
      "train_speed(iter/s)": 1.637209
    },
    {
      "acc": 0.64531078,
      "epoch": 1.3080923389142567,
      "grad_norm": 6.28125,
      "learning_rate": 2.9304372554122074e-06,
      "loss": 1.65907631,
      "memory(GiB)": 117.38,
      "step": 51565,
      "train_speed(iter/s)": 1.637225
    },
    {
      "acc": 0.64039068,
      "epoch": 1.308219178082192,
      "grad_norm": 5.875,
      "learning_rate": 2.9294827213393857e-06,
      "loss": 1.64682217,
      "memory(GiB)": 117.38,
      "step": 51570,
      "train_speed(iter/s)": 1.637241
    },
    {
      "acc": 0.66979671,
      "epoch": 1.308346017250127,
      "grad_norm": 5.25,
      "learning_rate": 2.9285282783371567e-06,
      "loss": 1.50388784,
      "memory(GiB)": 117.38,
      "step": 51575,
      "train_speed(iter/s)": 1.637257
    },
    {
      "acc": 0.65458665,
      "epoch": 1.3084728564180619,
      "grad_norm": 6.0,
      "learning_rate": 2.9275739264475013e-06,
      "loss": 1.57106981,
      "memory(GiB)": 117.38,
      "step": 51580,
      "train_speed(iter/s)": 1.637274
    },
    {
      "acc": 0.64749331,
      "epoch": 1.3085996955859969,
      "grad_norm": 5.5,
      "learning_rate": 2.926619665712399e-06,
      "loss": 1.5768364,
      "memory(GiB)": 117.38,
      "step": 51585,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.64772253,
      "epoch": 1.308726534753932,
      "grad_norm": 7.84375,
      "learning_rate": 2.9256654961738217e-06,
      "loss": 1.61024647,
      "memory(GiB)": 117.38,
      "step": 51590,
      "train_speed(iter/s)": 1.637306
    },
    {
      "acc": 0.66716747,
      "epoch": 1.308853373921867,
      "grad_norm": 5.5,
      "learning_rate": 2.9247114178737356e-06,
      "loss": 1.52597733,
      "memory(GiB)": 117.38,
      "step": 51595,
      "train_speed(iter/s)": 1.637322
    },
    {
      "acc": 0.65529041,
      "epoch": 1.308980213089802,
      "grad_norm": 5.03125,
      "learning_rate": 2.9237574308541063e-06,
      "loss": 1.59420547,
      "memory(GiB)": 117.38,
      "step": 51600,
      "train_speed(iter/s)": 1.637338
    },
    {
      "acc": 0.64674397,
      "epoch": 1.3091070522577373,
      "grad_norm": 5.0,
      "learning_rate": 2.9228035351568955e-06,
      "loss": 1.63080368,
      "memory(GiB)": 117.38,
      "step": 51605,
      "train_speed(iter/s)": 1.637354
    },
    {
      "acc": 0.67508087,
      "epoch": 1.3092338914256723,
      "grad_norm": 6.90625,
      "learning_rate": 2.92184973082406e-06,
      "loss": 1.47398796,
      "memory(GiB)": 117.38,
      "step": 51610,
      "train_speed(iter/s)": 1.63737
    },
    {
      "acc": 0.66170506,
      "epoch": 1.3093607305936072,
      "grad_norm": 5.8125,
      "learning_rate": 2.920896017897551e-06,
      "loss": 1.55876312,
      "memory(GiB)": 117.38,
      "step": 51615,
      "train_speed(iter/s)": 1.637386
    },
    {
      "acc": 0.64708872,
      "epoch": 1.3094875697615422,
      "grad_norm": 4.71875,
      "learning_rate": 2.9199423964193176e-06,
      "loss": 1.59538851,
      "memory(GiB)": 117.38,
      "step": 51620,
      "train_speed(iter/s)": 1.637402
    },
    {
      "acc": 0.65633287,
      "epoch": 1.3096144089294774,
      "grad_norm": 5.78125,
      "learning_rate": 2.9189888664313045e-06,
      "loss": 1.60734482,
      "memory(GiB)": 117.38,
      "step": 51625,
      "train_speed(iter/s)": 1.637418
    },
    {
      "acc": 0.65228448,
      "epoch": 1.3097412480974124,
      "grad_norm": 5.40625,
      "learning_rate": 2.9180354279754517e-06,
      "loss": 1.60209885,
      "memory(GiB)": 117.38,
      "step": 51630,
      "train_speed(iter/s)": 1.637434
    },
    {
      "acc": 0.65717149,
      "epoch": 1.3098680872653476,
      "grad_norm": 7.21875,
      "learning_rate": 2.9170820810936968e-06,
      "loss": 1.57594357,
      "memory(GiB)": 117.38,
      "step": 51635,
      "train_speed(iter/s)": 1.637449
    },
    {
      "acc": 0.65055823,
      "epoch": 1.3099949264332826,
      "grad_norm": 5.75,
      "learning_rate": 2.9161288258279715e-06,
      "loss": 1.61533432,
      "memory(GiB)": 117.38,
      "step": 51640,
      "train_speed(iter/s)": 1.637465
    },
    {
      "acc": 0.65223579,
      "epoch": 1.3101217656012176,
      "grad_norm": 6.0625,
      "learning_rate": 2.9151756622202037e-06,
      "loss": 1.6044939,
      "memory(GiB)": 117.38,
      "step": 51645,
      "train_speed(iter/s)": 1.637481
    },
    {
      "acc": 0.65353446,
      "epoch": 1.3102486047691526,
      "grad_norm": 9.0625,
      "learning_rate": 2.914222590312319e-06,
      "loss": 1.61331139,
      "memory(GiB)": 117.38,
      "step": 51650,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.65009022,
      "epoch": 1.3103754439370878,
      "grad_norm": 8.25,
      "learning_rate": 2.9132696101462366e-06,
      "loss": 1.60575142,
      "memory(GiB)": 117.38,
      "step": 51655,
      "train_speed(iter/s)": 1.637515
    },
    {
      "acc": 0.64684038,
      "epoch": 1.3105022831050228,
      "grad_norm": 5.96875,
      "learning_rate": 2.912316721763874e-06,
      "loss": 1.70278587,
      "memory(GiB)": 117.38,
      "step": 51660,
      "train_speed(iter/s)": 1.637532
    },
    {
      "acc": 0.66801019,
      "epoch": 1.310629122272958,
      "grad_norm": 5.78125,
      "learning_rate": 2.9113639252071395e-06,
      "loss": 1.50198326,
      "memory(GiB)": 117.38,
      "step": 51665,
      "train_speed(iter/s)": 1.637549
    },
    {
      "acc": 0.64791107,
      "epoch": 1.310755961440893,
      "grad_norm": 6.5,
      "learning_rate": 2.91041122051795e-06,
      "loss": 1.61493797,
      "memory(GiB)": 117.38,
      "step": 51670,
      "train_speed(iter/s)": 1.637565
    },
    {
      "acc": 0.64566798,
      "epoch": 1.310882800608828,
      "grad_norm": 5.0,
      "learning_rate": 2.9094586077382016e-06,
      "loss": 1.63597908,
      "memory(GiB)": 117.38,
      "step": 51675,
      "train_speed(iter/s)": 1.63758
    },
    {
      "acc": 0.65538793,
      "epoch": 1.311009639776763,
      "grad_norm": 6.78125,
      "learning_rate": 2.9085060869097977e-06,
      "loss": 1.59805403,
      "memory(GiB)": 117.38,
      "step": 51680,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.66358232,
      "epoch": 1.3111364789446982,
      "grad_norm": 5.25,
      "learning_rate": 2.907553658074631e-06,
      "loss": 1.52724018,
      "memory(GiB)": 117.38,
      "step": 51685,
      "train_speed(iter/s)": 1.63761
    },
    {
      "acc": 0.65128002,
      "epoch": 1.3112633181126332,
      "grad_norm": 5.21875,
      "learning_rate": 2.906601321274601e-06,
      "loss": 1.66135139,
      "memory(GiB)": 117.38,
      "step": 51690,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.6644032,
      "epoch": 1.3113901572805682,
      "grad_norm": 4.9375,
      "learning_rate": 2.90564907655159e-06,
      "loss": 1.61644211,
      "memory(GiB)": 117.38,
      "step": 51695,
      "train_speed(iter/s)": 1.637643
    },
    {
      "acc": 0.66594381,
      "epoch": 1.3115169964485034,
      "grad_norm": 5.53125,
      "learning_rate": 2.9046969239474808e-06,
      "loss": 1.53055096,
      "memory(GiB)": 117.38,
      "step": 51700,
      "train_speed(iter/s)": 1.63766
    },
    {
      "acc": 0.65734358,
      "epoch": 1.3116438356164384,
      "grad_norm": 5.25,
      "learning_rate": 2.9037448635041574e-06,
      "loss": 1.58900261,
      "memory(GiB)": 117.38,
      "step": 51705,
      "train_speed(iter/s)": 1.637677
    },
    {
      "acc": 0.6529006,
      "epoch": 1.3117706747843734,
      "grad_norm": 5.375,
      "learning_rate": 2.9027928952634964e-06,
      "loss": 1.57910213,
      "memory(GiB)": 117.38,
      "step": 51710,
      "train_speed(iter/s)": 1.637694
    },
    {
      "acc": 0.66224136,
      "epoch": 1.3118975139523084,
      "grad_norm": 5.4375,
      "learning_rate": 2.901841019267363e-06,
      "loss": 1.53208561,
      "memory(GiB)": 117.38,
      "step": 51715,
      "train_speed(iter/s)": 1.637709
    },
    {
      "acc": 0.6627141,
      "epoch": 1.3120243531202436,
      "grad_norm": 5.65625,
      "learning_rate": 2.900889235557631e-06,
      "loss": 1.63631554,
      "memory(GiB)": 117.38,
      "step": 51720,
      "train_speed(iter/s)": 1.637727
    },
    {
      "acc": 0.6683043,
      "epoch": 1.3121511922881786,
      "grad_norm": 5.625,
      "learning_rate": 2.8999375441761627e-06,
      "loss": 1.57904625,
      "memory(GiB)": 117.38,
      "step": 51725,
      "train_speed(iter/s)": 1.637741
    },
    {
      "acc": 0.66034756,
      "epoch": 1.3122780314561138,
      "grad_norm": 5.25,
      "learning_rate": 2.8989859451648193e-06,
      "loss": 1.54935474,
      "memory(GiB)": 117.38,
      "step": 51730,
      "train_speed(iter/s)": 1.637758
    },
    {
      "acc": 0.66374736,
      "epoch": 1.3124048706240488,
      "grad_norm": 7.125,
      "learning_rate": 2.89803443856545e-06,
      "loss": 1.50767307,
      "memory(GiB)": 117.38,
      "step": 51735,
      "train_speed(iter/s)": 1.637773
    },
    {
      "acc": 0.65610485,
      "epoch": 1.3125317097919837,
      "grad_norm": 5.28125,
      "learning_rate": 2.897083024419913e-06,
      "loss": 1.63265266,
      "memory(GiB)": 117.38,
      "step": 51740,
      "train_speed(iter/s)": 1.637789
    },
    {
      "acc": 0.65229473,
      "epoch": 1.3126585489599187,
      "grad_norm": 6.28125,
      "learning_rate": 2.8961317027700534e-06,
      "loss": 1.58904819,
      "memory(GiB)": 117.38,
      "step": 51745,
      "train_speed(iter/s)": 1.637805
    },
    {
      "acc": 0.65377903,
      "epoch": 1.312785388127854,
      "grad_norm": 6.9375,
      "learning_rate": 2.8951804736577148e-06,
      "loss": 1.61027641,
      "memory(GiB)": 117.38,
      "step": 51750,
      "train_speed(iter/s)": 1.637821
    },
    {
      "acc": 0.67582321,
      "epoch": 1.312912227295789,
      "grad_norm": 6.09375,
      "learning_rate": 2.894229337124736e-06,
      "loss": 1.57504196,
      "memory(GiB)": 117.38,
      "step": 51755,
      "train_speed(iter/s)": 1.637836
    },
    {
      "acc": 0.63971281,
      "epoch": 1.313039066463724,
      "grad_norm": 5.5,
      "learning_rate": 2.8932782932129524e-06,
      "loss": 1.60785809,
      "memory(GiB)": 117.38,
      "step": 51760,
      "train_speed(iter/s)": 1.637852
    },
    {
      "acc": 0.6622704,
      "epoch": 1.3131659056316591,
      "grad_norm": 5.875,
      "learning_rate": 2.8923273419641956e-06,
      "loss": 1.50102777,
      "memory(GiB)": 117.38,
      "step": 51765,
      "train_speed(iter/s)": 1.637869
    },
    {
      "acc": 0.66209106,
      "epoch": 1.3132927447995941,
      "grad_norm": 6.28125,
      "learning_rate": 2.891376483420292e-06,
      "loss": 1.58476448,
      "memory(GiB)": 117.38,
      "step": 51770,
      "train_speed(iter/s)": 1.637884
    },
    {
      "acc": 0.66455984,
      "epoch": 1.3134195839675291,
      "grad_norm": 6.28125,
      "learning_rate": 2.8904257176230655e-06,
      "loss": 1.56132154,
      "memory(GiB)": 117.38,
      "step": 51775,
      "train_speed(iter/s)": 1.637901
    },
    {
      "acc": 0.66377859,
      "epoch": 1.313546423135464,
      "grad_norm": 5.8125,
      "learning_rate": 2.8894750446143345e-06,
      "loss": 1.5216713,
      "memory(GiB)": 117.38,
      "step": 51780,
      "train_speed(iter/s)": 1.637918
    },
    {
      "acc": 0.64870396,
      "epoch": 1.3136732623033993,
      "grad_norm": 6.53125,
      "learning_rate": 2.8885244644359134e-06,
      "loss": 1.59811602,
      "memory(GiB)": 117.38,
      "step": 51785,
      "train_speed(iter/s)": 1.637935
    },
    {
      "acc": 0.64730048,
      "epoch": 1.3138001014713343,
      "grad_norm": 6.28125,
      "learning_rate": 2.887573977129614e-06,
      "loss": 1.649617,
      "memory(GiB)": 117.38,
      "step": 51790,
      "train_speed(iter/s)": 1.637953
    },
    {
      "acc": 0.66811953,
      "epoch": 1.3139269406392695,
      "grad_norm": 6.96875,
      "learning_rate": 2.886623582737242e-06,
      "loss": 1.58251591,
      "memory(GiB)": 117.38,
      "step": 51795,
      "train_speed(iter/s)": 1.637968
    },
    {
      "acc": 0.66126637,
      "epoch": 1.3140537798072045,
      "grad_norm": 5.5,
      "learning_rate": 2.8856732813006007e-06,
      "loss": 1.64648476,
      "memory(GiB)": 117.38,
      "step": 51800,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.65193205,
      "epoch": 1.3141806189751395,
      "grad_norm": 5.4375,
      "learning_rate": 2.8847230728614854e-06,
      "loss": 1.66526375,
      "memory(GiB)": 117.38,
      "step": 51805,
      "train_speed(iter/s)": 1.638
    },
    {
      "acc": 0.67225418,
      "epoch": 1.3143074581430745,
      "grad_norm": 5.5,
      "learning_rate": 2.883772957461698e-06,
      "loss": 1.50958004,
      "memory(GiB)": 117.38,
      "step": 51810,
      "train_speed(iter/s)": 1.638016
    },
    {
      "acc": 0.64741755,
      "epoch": 1.3144342973110097,
      "grad_norm": 6.96875,
      "learning_rate": 2.8828229351430224e-06,
      "loss": 1.67167568,
      "memory(GiB)": 117.38,
      "step": 51815,
      "train_speed(iter/s)": 1.638033
    },
    {
      "acc": 0.66516385,
      "epoch": 1.3145611364789447,
      "grad_norm": 6.28125,
      "learning_rate": 2.881873005947247e-06,
      "loss": 1.56833324,
      "memory(GiB)": 117.38,
      "step": 51820,
      "train_speed(iter/s)": 1.638048
    },
    {
      "acc": 0.64847803,
      "epoch": 1.31468797564688,
      "grad_norm": 5.25,
      "learning_rate": 2.88092316991615e-06,
      "loss": 1.6051199,
      "memory(GiB)": 117.38,
      "step": 51825,
      "train_speed(iter/s)": 1.638064
    },
    {
      "acc": 0.66874151,
      "epoch": 1.3148148148148149,
      "grad_norm": 5.78125,
      "learning_rate": 2.879973427091518e-06,
      "loss": 1.56490974,
      "memory(GiB)": 117.38,
      "step": 51830,
      "train_speed(iter/s)": 1.638081
    },
    {
      "acc": 0.66776762,
      "epoch": 1.3149416539827499,
      "grad_norm": 7.03125,
      "learning_rate": 2.879023777515118e-06,
      "loss": 1.54060535,
      "memory(GiB)": 117.38,
      "step": 51835,
      "train_speed(iter/s)": 1.638097
    },
    {
      "acc": 0.64110527,
      "epoch": 1.3150684931506849,
      "grad_norm": 5.5625,
      "learning_rate": 2.8780742212287192e-06,
      "loss": 1.59418535,
      "memory(GiB)": 117.38,
      "step": 51840,
      "train_speed(iter/s)": 1.638114
    },
    {
      "acc": 0.64460273,
      "epoch": 1.31519533231862,
      "grad_norm": 5.96875,
      "learning_rate": 2.8771247582740924e-06,
      "loss": 1.63289852,
      "memory(GiB)": 117.38,
      "step": 51845,
      "train_speed(iter/s)": 1.63813
    },
    {
      "acc": 0.66308203,
      "epoch": 1.315322171486555,
      "grad_norm": 5.625,
      "learning_rate": 2.876175388692999e-06,
      "loss": 1.57990704,
      "memory(GiB)": 117.38,
      "step": 51850,
      "train_speed(iter/s)": 1.638147
    },
    {
      "acc": 0.64651594,
      "epoch": 1.31544901065449,
      "grad_norm": 7.03125,
      "learning_rate": 2.875226112527192e-06,
      "loss": 1.67886295,
      "memory(GiB)": 117.38,
      "step": 51855,
      "train_speed(iter/s)": 1.638163
    },
    {
      "acc": 0.66003923,
      "epoch": 1.3155758498224253,
      "grad_norm": 5.75,
      "learning_rate": 2.8742769298184246e-06,
      "loss": 1.6152523,
      "memory(GiB)": 117.38,
      "step": 51860,
      "train_speed(iter/s)": 1.63818
    },
    {
      "acc": 0.65803847,
      "epoch": 1.3157026889903602,
      "grad_norm": 4.875,
      "learning_rate": 2.8733278406084507e-06,
      "loss": 1.60375576,
      "memory(GiB)": 117.38,
      "step": 51865,
      "train_speed(iter/s)": 1.638195
    },
    {
      "acc": 0.66233673,
      "epoch": 1.3158295281582952,
      "grad_norm": 5.90625,
      "learning_rate": 2.872378844939015e-06,
      "loss": 1.6214613,
      "memory(GiB)": 117.38,
      "step": 51870,
      "train_speed(iter/s)": 1.638212
    },
    {
      "acc": 0.65733314,
      "epoch": 1.3159563673262302,
      "grad_norm": 5.59375,
      "learning_rate": 2.871429942851853e-06,
      "loss": 1.66374626,
      "memory(GiB)": 117.38,
      "step": 51875,
      "train_speed(iter/s)": 1.638227
    },
    {
      "acc": 0.66757278,
      "epoch": 1.3160832064941654,
      "grad_norm": 7.0625,
      "learning_rate": 2.8704811343887075e-06,
      "loss": 1.53596821,
      "memory(GiB)": 117.38,
      "step": 51880,
      "train_speed(iter/s)": 1.638243
    },
    {
      "acc": 0.66010361,
      "epoch": 1.3162100456621004,
      "grad_norm": 6.25,
      "learning_rate": 2.86953241959131e-06,
      "loss": 1.58630199,
      "memory(GiB)": 117.38,
      "step": 51885,
      "train_speed(iter/s)": 1.638259
    },
    {
      "acc": 0.67056479,
      "epoch": 1.3163368848300356,
      "grad_norm": 5.71875,
      "learning_rate": 2.8685837985013874e-06,
      "loss": 1.56700478,
      "memory(GiB)": 117.38,
      "step": 51890,
      "train_speed(iter/s)": 1.638275
    },
    {
      "acc": 0.66395802,
      "epoch": 1.3164637239979706,
      "grad_norm": 6.0,
      "learning_rate": 2.867635271160666e-06,
      "loss": 1.55505075,
      "memory(GiB)": 117.38,
      "step": 51895,
      "train_speed(iter/s)": 1.638291
    },
    {
      "acc": 0.64235859,
      "epoch": 1.3165905631659056,
      "grad_norm": 6.4375,
      "learning_rate": 2.8666868376108658e-06,
      "loss": 1.71422424,
      "memory(GiB)": 117.38,
      "step": 51900,
      "train_speed(iter/s)": 1.638308
    },
    {
      "acc": 0.6542592,
      "epoch": 1.3167174023338406,
      "grad_norm": 6.46875,
      "learning_rate": 2.865738497893703e-06,
      "loss": 1.5798089,
      "memory(GiB)": 117.38,
      "step": 51905,
      "train_speed(iter/s)": 1.638324
    },
    {
      "acc": 0.64640779,
      "epoch": 1.3168442415017758,
      "grad_norm": 5.90625,
      "learning_rate": 2.8647902520508896e-06,
      "loss": 1.576754,
      "memory(GiB)": 117.38,
      "step": 51910,
      "train_speed(iter/s)": 1.638341
    },
    {
      "acc": 0.64876933,
      "epoch": 1.3169710806697108,
      "grad_norm": 5.4375,
      "learning_rate": 2.8638421001241346e-06,
      "loss": 1.61733742,
      "memory(GiB)": 117.38,
      "step": 51915,
      "train_speed(iter/s)": 1.638357
    },
    {
      "acc": 0.65364203,
      "epoch": 1.3170979198376458,
      "grad_norm": 6.40625,
      "learning_rate": 2.8628940421551404e-06,
      "loss": 1.6147459,
      "memory(GiB)": 117.38,
      "step": 51920,
      "train_speed(iter/s)": 1.638373
    },
    {
      "acc": 0.64499273,
      "epoch": 1.317224759005581,
      "grad_norm": 5.9375,
      "learning_rate": 2.861946078185608e-06,
      "loss": 1.66424904,
      "memory(GiB)": 117.38,
      "step": 51925,
      "train_speed(iter/s)": 1.638388
    },
    {
      "acc": 0.66102796,
      "epoch": 1.317351598173516,
      "grad_norm": 5.34375,
      "learning_rate": 2.860998208257233e-06,
      "loss": 1.56621389,
      "memory(GiB)": 117.38,
      "step": 51930,
      "train_speed(iter/s)": 1.638406
    },
    {
      "acc": 0.65145807,
      "epoch": 1.317478437341451,
      "grad_norm": 5.3125,
      "learning_rate": 2.860050432411707e-06,
      "loss": 1.654772,
      "memory(GiB)": 117.38,
      "step": 51935,
      "train_speed(iter/s)": 1.638422
    },
    {
      "acc": 0.64750786,
      "epoch": 1.317605276509386,
      "grad_norm": 5.375,
      "learning_rate": 2.8591027506907167e-06,
      "loss": 1.66390533,
      "memory(GiB)": 117.38,
      "step": 51940,
      "train_speed(iter/s)": 1.638437
    },
    {
      "acc": 0.6699873,
      "epoch": 1.3177321156773212,
      "grad_norm": 5.09375,
      "learning_rate": 2.858155163135946e-06,
      "loss": 1.59997644,
      "memory(GiB)": 117.38,
      "step": 51945,
      "train_speed(iter/s)": 1.638453
    },
    {
      "acc": 0.64035473,
      "epoch": 1.3178589548452562,
      "grad_norm": 5.375,
      "learning_rate": 2.857207669789074e-06,
      "loss": 1.64997883,
      "memory(GiB)": 117.38,
      "step": 51950,
      "train_speed(iter/s)": 1.63847
    },
    {
      "acc": 0.64072218,
      "epoch": 1.3179857940131914,
      "grad_norm": 5.40625,
      "learning_rate": 2.8562602706917754e-06,
      "loss": 1.65395164,
      "memory(GiB)": 117.38,
      "step": 51955,
      "train_speed(iter/s)": 1.638486
    },
    {
      "acc": 0.66478128,
      "epoch": 1.3181126331811264,
      "grad_norm": 5.15625,
      "learning_rate": 2.8553129658857215e-06,
      "loss": 1.61188622,
      "memory(GiB)": 117.38,
      "step": 51960,
      "train_speed(iter/s)": 1.638503
    },
    {
      "acc": 0.6422164,
      "epoch": 1.3182394723490614,
      "grad_norm": 5.84375,
      "learning_rate": 2.854365755412576e-06,
      "loss": 1.59453106,
      "memory(GiB)": 117.38,
      "step": 51965,
      "train_speed(iter/s)": 1.638517
    },
    {
      "acc": 0.66423454,
      "epoch": 1.3183663115169963,
      "grad_norm": 6.28125,
      "learning_rate": 2.8534186393140083e-06,
      "loss": 1.53146667,
      "memory(GiB)": 117.38,
      "step": 51970,
      "train_speed(iter/s)": 1.638534
    },
    {
      "acc": 0.65198679,
      "epoch": 1.3184931506849316,
      "grad_norm": 5.34375,
      "learning_rate": 2.8524716176316715e-06,
      "loss": 1.55946875,
      "memory(GiB)": 117.38,
      "step": 51975,
      "train_speed(iter/s)": 1.638548
    },
    {
      "acc": 0.63573198,
      "epoch": 1.3186199898528665,
      "grad_norm": 6.84375,
      "learning_rate": 2.851524690407218e-06,
      "loss": 1.69592514,
      "memory(GiB)": 117.38,
      "step": 51980,
      "train_speed(iter/s)": 1.638565
    },
    {
      "acc": 0.65509243,
      "epoch": 1.3187468290208018,
      "grad_norm": 4.59375,
      "learning_rate": 2.8505778576823036e-06,
      "loss": 1.6254961,
      "memory(GiB)": 117.38,
      "step": 51985,
      "train_speed(iter/s)": 1.638581
    },
    {
      "acc": 0.65358362,
      "epoch": 1.3188736681887367,
      "grad_norm": 6.5,
      "learning_rate": 2.849631119498573e-06,
      "loss": 1.65867023,
      "memory(GiB)": 117.38,
      "step": 51990,
      "train_speed(iter/s)": 1.638597
    },
    {
      "acc": 0.64058247,
      "epoch": 1.3190005073566717,
      "grad_norm": 5.21875,
      "learning_rate": 2.8486844758976652e-06,
      "loss": 1.633815,
      "memory(GiB)": 117.38,
      "step": 51995,
      "train_speed(iter/s)": 1.638613
    },
    {
      "acc": 0.66000299,
      "epoch": 1.3191273465246067,
      "grad_norm": 5.90625,
      "learning_rate": 2.8477379269212157e-06,
      "loss": 1.57531452,
      "memory(GiB)": 117.38,
      "step": 52000,
      "train_speed(iter/s)": 1.638631
    },
    {
      "epoch": 1.3191273465246067,
      "eval_acc": 0.6463119469904376,
      "eval_loss": 1.5733332633972168,
      "eval_runtime": 58.9081,
      "eval_samples_per_second": 108.134,
      "eval_steps_per_second": 27.042,
      "step": 52000
    },
    {
      "acc": 0.65874939,
      "epoch": 1.319254185692542,
      "grad_norm": 5.3125,
      "learning_rate": 2.846791472610865e-06,
      "loss": 1.58662806,
      "memory(GiB)": 117.38,
      "step": 52005,
      "train_speed(iter/s)": 1.635387
    },
    {
      "acc": 0.64676714,
      "epoch": 1.319381024860477,
      "grad_norm": 6.0,
      "learning_rate": 2.845845113008239e-06,
      "loss": 1.65532303,
      "memory(GiB)": 117.38,
      "step": 52010,
      "train_speed(iter/s)": 1.635404
    },
    {
      "acc": 0.65876184,
      "epoch": 1.319507864028412,
      "grad_norm": 5.6875,
      "learning_rate": 2.84489884815496e-06,
      "loss": 1.59165421,
      "memory(GiB)": 117.38,
      "step": 52015,
      "train_speed(iter/s)": 1.635421
    },
    {
      "acc": 0.64427605,
      "epoch": 1.3196347031963471,
      "grad_norm": 5.0625,
      "learning_rate": 2.843952678092653e-06,
      "loss": 1.57793026,
      "memory(GiB)": 117.38,
      "step": 52020,
      "train_speed(iter/s)": 1.635437
    },
    {
      "acc": 0.64241514,
      "epoch": 1.3197615423642821,
      "grad_norm": 6.3125,
      "learning_rate": 2.8430066028629328e-06,
      "loss": 1.65505943,
      "memory(GiB)": 117.38,
      "step": 52025,
      "train_speed(iter/s)": 1.635453
    },
    {
      "acc": 0.66073937,
      "epoch": 1.319888381532217,
      "grad_norm": 5.4375,
      "learning_rate": 2.842060622507415e-06,
      "loss": 1.54876308,
      "memory(GiB)": 117.38,
      "step": 52030,
      "train_speed(iter/s)": 1.63547
    },
    {
      "acc": 0.66686354,
      "epoch": 1.320015220700152,
      "grad_norm": 5.53125,
      "learning_rate": 2.841114737067702e-06,
      "loss": 1.53359814,
      "memory(GiB)": 117.38,
      "step": 52035,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.67591119,
      "epoch": 1.3201420598680873,
      "grad_norm": 5.09375,
      "learning_rate": 2.840168946585402e-06,
      "loss": 1.49785595,
      "memory(GiB)": 117.38,
      "step": 52040,
      "train_speed(iter/s)": 1.635503
    },
    {
      "acc": 0.66978974,
      "epoch": 1.3202688990360223,
      "grad_norm": 6.15625,
      "learning_rate": 2.8392232511021158e-06,
      "loss": 1.56766491,
      "memory(GiB)": 117.38,
      "step": 52045,
      "train_speed(iter/s)": 1.635518
    },
    {
      "acc": 0.64868822,
      "epoch": 1.3203957382039575,
      "grad_norm": 7.75,
      "learning_rate": 2.8382776506594385e-06,
      "loss": 1.64825268,
      "memory(GiB)": 117.38,
      "step": 52050,
      "train_speed(iter/s)": 1.635534
    },
    {
      "acc": 0.65092087,
      "epoch": 1.3205225773718925,
      "grad_norm": 5.75,
      "learning_rate": 2.837332145298961e-06,
      "loss": 1.60266342,
      "memory(GiB)": 117.38,
      "step": 52055,
      "train_speed(iter/s)": 1.63555
    },
    {
      "acc": 0.662743,
      "epoch": 1.3206494165398275,
      "grad_norm": 7.34375,
      "learning_rate": 2.836386735062271e-06,
      "loss": 1.61657543,
      "memory(GiB)": 117.38,
      "step": 52060,
      "train_speed(iter/s)": 1.635565
    },
    {
      "acc": 0.67379699,
      "epoch": 1.3207762557077625,
      "grad_norm": 6.03125,
      "learning_rate": 2.835441419990953e-06,
      "loss": 1.52454882,
      "memory(GiB)": 117.38,
      "step": 52065,
      "train_speed(iter/s)": 1.635581
    },
    {
      "acc": 0.64366484,
      "epoch": 1.3209030948756977,
      "grad_norm": 4.90625,
      "learning_rate": 2.834496200126585e-06,
      "loss": 1.60783329,
      "memory(GiB)": 117.38,
      "step": 52070,
      "train_speed(iter/s)": 1.635597
    },
    {
      "acc": 0.63625898,
      "epoch": 1.3210299340436327,
      "grad_norm": 6.75,
      "learning_rate": 2.8335510755107426e-06,
      "loss": 1.69429092,
      "memory(GiB)": 117.38,
      "step": 52075,
      "train_speed(iter/s)": 1.635613
    },
    {
      "acc": 0.66968241,
      "epoch": 1.3211567732115677,
      "grad_norm": 7.3125,
      "learning_rate": 2.8326060461849966e-06,
      "loss": 1.59520359,
      "memory(GiB)": 117.38,
      "step": 52080,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.6709352,
      "epoch": 1.3212836123795029,
      "grad_norm": 7.25,
      "learning_rate": 2.8316611121909126e-06,
      "loss": 1.57514286,
      "memory(GiB)": 117.38,
      "step": 52085,
      "train_speed(iter/s)": 1.635643
    },
    {
      "acc": 0.65509658,
      "epoch": 1.3214104515474379,
      "grad_norm": 5.34375,
      "learning_rate": 2.8307162735700544e-06,
      "loss": 1.7069088,
      "memory(GiB)": 117.38,
      "step": 52090,
      "train_speed(iter/s)": 1.635659
    },
    {
      "acc": 0.66510782,
      "epoch": 1.3215372907153728,
      "grad_norm": 5.84375,
      "learning_rate": 2.8297715303639796e-06,
      "loss": 1.53103218,
      "memory(GiB)": 117.38,
      "step": 52095,
      "train_speed(iter/s)": 1.635675
    },
    {
      "acc": 0.64029341,
      "epoch": 1.3216641298833078,
      "grad_norm": 5.0,
      "learning_rate": 2.8288268826142423e-06,
      "loss": 1.67757034,
      "memory(GiB)": 117.38,
      "step": 52100,
      "train_speed(iter/s)": 1.635691
    },
    {
      "acc": 0.6592268,
      "epoch": 1.321790969051243,
      "grad_norm": 7.21875,
      "learning_rate": 2.8278823303623905e-06,
      "loss": 1.60833187,
      "memory(GiB)": 117.38,
      "step": 52105,
      "train_speed(iter/s)": 1.635707
    },
    {
      "acc": 0.64504409,
      "epoch": 1.321917808219178,
      "grad_norm": 5.625,
      "learning_rate": 2.8269378736499754e-06,
      "loss": 1.65360146,
      "memory(GiB)": 117.38,
      "step": 52110,
      "train_speed(iter/s)": 1.635724
    },
    {
      "acc": 0.65946589,
      "epoch": 1.3220446473871132,
      "grad_norm": 5.84375,
      "learning_rate": 2.8259935125185323e-06,
      "loss": 1.64948692,
      "memory(GiB)": 117.38,
      "step": 52115,
      "train_speed(iter/s)": 1.635741
    },
    {
      "acc": 0.6404623,
      "epoch": 1.3221714865550482,
      "grad_norm": 6.8125,
      "learning_rate": 2.8250492470096008e-06,
      "loss": 1.69871597,
      "memory(GiB)": 117.38,
      "step": 52120,
      "train_speed(iter/s)": 1.635757
    },
    {
      "acc": 0.65764208,
      "epoch": 1.3222983257229832,
      "grad_norm": 7.8125,
      "learning_rate": 2.824105077164712e-06,
      "loss": 1.57595005,
      "memory(GiB)": 117.38,
      "step": 52125,
      "train_speed(iter/s)": 1.635769
    },
    {
      "acc": 0.64921436,
      "epoch": 1.3224251648909182,
      "grad_norm": 5.03125,
      "learning_rate": 2.823161003025401e-06,
      "loss": 1.62484474,
      "memory(GiB)": 117.38,
      "step": 52130,
      "train_speed(iter/s)": 1.635785
    },
    {
      "acc": 0.6612052,
      "epoch": 1.3225520040588534,
      "grad_norm": 9.8125,
      "learning_rate": 2.822217024633186e-06,
      "loss": 1.67833023,
      "memory(GiB)": 117.38,
      "step": 52135,
      "train_speed(iter/s)": 1.635801
    },
    {
      "acc": 0.65908689,
      "epoch": 1.3226788432267884,
      "grad_norm": 5.59375,
      "learning_rate": 2.821273142029587e-06,
      "loss": 1.60943222,
      "memory(GiB)": 117.38,
      "step": 52140,
      "train_speed(iter/s)": 1.635815
    },
    {
      "acc": 0.64429984,
      "epoch": 1.3228056823947236,
      "grad_norm": 6.0,
      "learning_rate": 2.820329355256124e-06,
      "loss": 1.5831439,
      "memory(GiB)": 117.38,
      "step": 52145,
      "train_speed(iter/s)": 1.635831
    },
    {
      "acc": 0.65499277,
      "epoch": 1.3229325215626586,
      "grad_norm": 5.65625,
      "learning_rate": 2.8193856643543106e-06,
      "loss": 1.58729792,
      "memory(GiB)": 117.38,
      "step": 52150,
      "train_speed(iter/s)": 1.635846
    },
    {
      "acc": 0.65308132,
      "epoch": 1.3230593607305936,
      "grad_norm": 7.09375,
      "learning_rate": 2.8184420693656468e-06,
      "loss": 1.62044601,
      "memory(GiB)": 117.38,
      "step": 52155,
      "train_speed(iter/s)": 1.635862
    },
    {
      "acc": 0.66029449,
      "epoch": 1.3231861998985286,
      "grad_norm": 5.15625,
      "learning_rate": 2.817498570331643e-06,
      "loss": 1.56551914,
      "memory(GiB)": 117.38,
      "step": 52160,
      "train_speed(iter/s)": 1.635878
    },
    {
      "acc": 0.66244607,
      "epoch": 1.3233130390664638,
      "grad_norm": 5.0,
      "learning_rate": 2.816555167293795e-06,
      "loss": 1.54772625,
      "memory(GiB)": 117.38,
      "step": 52165,
      "train_speed(iter/s)": 1.635894
    },
    {
      "acc": 0.65782738,
      "epoch": 1.3234398782343988,
      "grad_norm": 5.25,
      "learning_rate": 2.815611860293603e-06,
      "loss": 1.65788116,
      "memory(GiB)": 117.38,
      "step": 52170,
      "train_speed(iter/s)": 1.635909
    },
    {
      "acc": 0.65940089,
      "epoch": 1.3235667174023338,
      "grad_norm": 5.15625,
      "learning_rate": 2.814668649372549e-06,
      "loss": 1.61913071,
      "memory(GiB)": 117.38,
      "step": 52175,
      "train_speed(iter/s)": 1.635924
    },
    {
      "acc": 0.65202379,
      "epoch": 1.323693556570269,
      "grad_norm": 6.25,
      "learning_rate": 2.8137255345721266e-06,
      "loss": 1.64482841,
      "memory(GiB)": 117.38,
      "step": 52180,
      "train_speed(iter/s)": 1.635939
    },
    {
      "acc": 0.65637221,
      "epoch": 1.323820395738204,
      "grad_norm": 6.65625,
      "learning_rate": 2.8127825159338163e-06,
      "loss": 1.63288727,
      "memory(GiB)": 117.38,
      "step": 52185,
      "train_speed(iter/s)": 1.635954
    },
    {
      "acc": 0.64874058,
      "epoch": 1.323947234906139,
      "grad_norm": 5.96875,
      "learning_rate": 2.8118395934990962e-06,
      "loss": 1.69786911,
      "memory(GiB)": 117.38,
      "step": 52190,
      "train_speed(iter/s)": 1.635968
    },
    {
      "acc": 0.63527327,
      "epoch": 1.324074074074074,
      "grad_norm": 5.53125,
      "learning_rate": 2.81089676730944e-06,
      "loss": 1.64451103,
      "memory(GiB)": 117.38,
      "step": 52195,
      "train_speed(iter/s)": 1.635982
    },
    {
      "acc": 0.66795893,
      "epoch": 1.3242009132420092,
      "grad_norm": 4.875,
      "learning_rate": 2.8099540374063185e-06,
      "loss": 1.54161692,
      "memory(GiB)": 117.38,
      "step": 52200,
      "train_speed(iter/s)": 1.635996
    },
    {
      "acc": 0.64185085,
      "epoch": 1.3243277524099442,
      "grad_norm": 4.65625,
      "learning_rate": 2.8090114038311956e-06,
      "loss": 1.660499,
      "memory(GiB)": 117.38,
      "step": 52205,
      "train_speed(iter/s)": 1.636011
    },
    {
      "acc": 0.65227818,
      "epoch": 1.3244545915778794,
      "grad_norm": 6.125,
      "learning_rate": 2.8080688666255328e-06,
      "loss": 1.58447628,
      "memory(GiB)": 117.38,
      "step": 52210,
      "train_speed(iter/s)": 1.636026
    },
    {
      "acc": 0.6516387,
      "epoch": 1.3245814307458144,
      "grad_norm": 5.40625,
      "learning_rate": 2.8071264258307884e-06,
      "loss": 1.58347549,
      "memory(GiB)": 117.38,
      "step": 52215,
      "train_speed(iter/s)": 1.636038
    },
    {
      "acc": 0.66553926,
      "epoch": 1.3247082699137493,
      "grad_norm": 5.1875,
      "learning_rate": 2.8061840814884133e-06,
      "loss": 1.56262131,
      "memory(GiB)": 117.38,
      "step": 52220,
      "train_speed(iter/s)": 1.636053
    },
    {
      "acc": 0.66843071,
      "epoch": 1.3248351090816843,
      "grad_norm": 4.84375,
      "learning_rate": 2.805241833639858e-06,
      "loss": 1.57691097,
      "memory(GiB)": 117.38,
      "step": 52225,
      "train_speed(iter/s)": 1.636068
    },
    {
      "acc": 0.6546845,
      "epoch": 1.3249619482496195,
      "grad_norm": 5.09375,
      "learning_rate": 2.804299682326565e-06,
      "loss": 1.62780075,
      "memory(GiB)": 117.38,
      "step": 52230,
      "train_speed(iter/s)": 1.636082
    },
    {
      "acc": 0.65772219,
      "epoch": 1.3250887874175545,
      "grad_norm": 5.5625,
      "learning_rate": 2.8033576275899752e-06,
      "loss": 1.62959042,
      "memory(GiB)": 117.38,
      "step": 52235,
      "train_speed(iter/s)": 1.636096
    },
    {
      "acc": 0.65651793,
      "epoch": 1.3252156265854895,
      "grad_norm": 5.125,
      "learning_rate": 2.8024156694715242e-06,
      "loss": 1.62317276,
      "memory(GiB)": 117.38,
      "step": 52240,
      "train_speed(iter/s)": 1.63611
    },
    {
      "acc": 0.64496408,
      "epoch": 1.3253424657534247,
      "grad_norm": 10.4375,
      "learning_rate": 2.8014738080126424e-06,
      "loss": 1.64377098,
      "memory(GiB)": 117.38,
      "step": 52245,
      "train_speed(iter/s)": 1.636124
    },
    {
      "acc": 0.65751052,
      "epoch": 1.3254693049213597,
      "grad_norm": 4.46875,
      "learning_rate": 2.8005320432547612e-06,
      "loss": 1.6021944,
      "memory(GiB)": 117.38,
      "step": 52250,
      "train_speed(iter/s)": 1.636139
    },
    {
      "acc": 0.64199843,
      "epoch": 1.3255961440892947,
      "grad_norm": 6.3125,
      "learning_rate": 2.7995903752392993e-06,
      "loss": 1.63557053,
      "memory(GiB)": 117.38,
      "step": 52255,
      "train_speed(iter/s)": 1.636154
    },
    {
      "acc": 0.67634807,
      "epoch": 1.3257229832572297,
      "grad_norm": 7.09375,
      "learning_rate": 2.7986488040076764e-06,
      "loss": 1.50680571,
      "memory(GiB)": 117.38,
      "step": 52260,
      "train_speed(iter/s)": 1.63617
    },
    {
      "acc": 0.65127783,
      "epoch": 1.325849822425165,
      "grad_norm": 7.1875,
      "learning_rate": 2.797707329601306e-06,
      "loss": 1.66753712,
      "memory(GiB)": 117.38,
      "step": 52265,
      "train_speed(iter/s)": 1.636185
    },
    {
      "acc": 0.654738,
      "epoch": 1.3259766615931,
      "grad_norm": 7.28125,
      "learning_rate": 2.7967659520616032e-06,
      "loss": 1.60343819,
      "memory(GiB)": 117.38,
      "step": 52270,
      "train_speed(iter/s)": 1.636201
    },
    {
      "acc": 0.66296921,
      "epoch": 1.3261035007610351,
      "grad_norm": 5.09375,
      "learning_rate": 2.7958246714299685e-06,
      "loss": 1.48076677,
      "memory(GiB)": 117.38,
      "step": 52275,
      "train_speed(iter/s)": 1.636215
    },
    {
      "acc": 0.64425993,
      "epoch": 1.32623033992897,
      "grad_norm": 4.84375,
      "learning_rate": 2.7948834877478035e-06,
      "loss": 1.64650574,
      "memory(GiB)": 117.38,
      "step": 52280,
      "train_speed(iter/s)": 1.636231
    },
    {
      "acc": 0.63907843,
      "epoch": 1.326357179096905,
      "grad_norm": 6.8125,
      "learning_rate": 2.7939424010565107e-06,
      "loss": 1.67418308,
      "memory(GiB)": 117.38,
      "step": 52285,
      "train_speed(iter/s)": 1.636246
    },
    {
      "acc": 0.66446128,
      "epoch": 1.32648401826484,
      "grad_norm": 5.4375,
      "learning_rate": 2.793001411397482e-06,
      "loss": 1.63926468,
      "memory(GiB)": 117.38,
      "step": 52290,
      "train_speed(iter/s)": 1.636262
    },
    {
      "acc": 0.65369778,
      "epoch": 1.3266108574327753,
      "grad_norm": 6.8125,
      "learning_rate": 2.792060518812103e-06,
      "loss": 1.65218544,
      "memory(GiB)": 117.38,
      "step": 52295,
      "train_speed(iter/s)": 1.636278
    },
    {
      "acc": 0.66207428,
      "epoch": 1.3267376966007103,
      "grad_norm": 6.46875,
      "learning_rate": 2.7911197233417574e-06,
      "loss": 1.60361862,
      "memory(GiB)": 117.38,
      "step": 52300,
      "train_speed(iter/s)": 1.636293
    },
    {
      "acc": 0.65392046,
      "epoch": 1.3268645357686455,
      "grad_norm": 6.125,
      "learning_rate": 2.790179025027831e-06,
      "loss": 1.62812309,
      "memory(GiB)": 117.38,
      "step": 52305,
      "train_speed(iter/s)": 1.63631
    },
    {
      "acc": 0.64088154,
      "epoch": 1.3269913749365805,
      "grad_norm": 6.5625,
      "learning_rate": 2.789238423911699e-06,
      "loss": 1.64108238,
      "memory(GiB)": 117.38,
      "step": 52310,
      "train_speed(iter/s)": 1.636328
    },
    {
      "acc": 0.66546278,
      "epoch": 1.3271182141045155,
      "grad_norm": 5.5625,
      "learning_rate": 2.788297920034727e-06,
      "loss": 1.53426723,
      "memory(GiB)": 117.38,
      "step": 52315,
      "train_speed(iter/s)": 1.636344
    },
    {
      "acc": 0.66275949,
      "epoch": 1.3272450532724505,
      "grad_norm": 6.5,
      "learning_rate": 2.78735751343829e-06,
      "loss": 1.59943008,
      "memory(GiB)": 117.38,
      "step": 52320,
      "train_speed(iter/s)": 1.636362
    },
    {
      "acc": 0.65351934,
      "epoch": 1.3273718924403857,
      "grad_norm": 6.1875,
      "learning_rate": 2.786417204163748e-06,
      "loss": 1.6237751,
      "memory(GiB)": 117.38,
      "step": 52325,
      "train_speed(iter/s)": 1.636378
    },
    {
      "acc": 0.64910564,
      "epoch": 1.3274987316083207,
      "grad_norm": 4.9375,
      "learning_rate": 2.7854769922524593e-06,
      "loss": 1.59807625,
      "memory(GiB)": 117.38,
      "step": 52330,
      "train_speed(iter/s)": 1.636395
    },
    {
      "acc": 0.6614953,
      "epoch": 1.3276255707762556,
      "grad_norm": 5.65625,
      "learning_rate": 2.7845368777457803e-06,
      "loss": 1.61577263,
      "memory(GiB)": 117.38,
      "step": 52335,
      "train_speed(iter/s)": 1.636412
    },
    {
      "acc": 0.65810719,
      "epoch": 1.3277524099441909,
      "grad_norm": 5.40625,
      "learning_rate": 2.7835968606850616e-06,
      "loss": 1.60805588,
      "memory(GiB)": 117.38,
      "step": 52340,
      "train_speed(iter/s)": 1.636428
    },
    {
      "acc": 0.64208531,
      "epoch": 1.3278792491121258,
      "grad_norm": 7.03125,
      "learning_rate": 2.782656941111648e-06,
      "loss": 1.63031311,
      "memory(GiB)": 117.38,
      "step": 52345,
      "train_speed(iter/s)": 1.636446
    },
    {
      "acc": 0.67260337,
      "epoch": 1.3280060882800608,
      "grad_norm": 8.0625,
      "learning_rate": 2.7817171190668812e-06,
      "loss": 1.49973927,
      "memory(GiB)": 117.38,
      "step": 52350,
      "train_speed(iter/s)": 1.636464
    },
    {
      "acc": 0.65336275,
      "epoch": 1.3281329274479958,
      "grad_norm": 6.34375,
      "learning_rate": 2.7807773945921e-06,
      "loss": 1.60625591,
      "memory(GiB)": 117.38,
      "step": 52355,
      "train_speed(iter/s)": 1.636481
    },
    {
      "acc": 0.66692867,
      "epoch": 1.328259766615931,
      "grad_norm": 5.28125,
      "learning_rate": 2.7798377677286363e-06,
      "loss": 1.57732439,
      "memory(GiB)": 117.38,
      "step": 52360,
      "train_speed(iter/s)": 1.636497
    },
    {
      "acc": 0.66101351,
      "epoch": 1.328386605783866,
      "grad_norm": 5.375,
      "learning_rate": 2.778898238517821e-06,
      "loss": 1.53063793,
      "memory(GiB)": 117.38,
      "step": 52365,
      "train_speed(iter/s)": 1.636512
    },
    {
      "acc": 0.65004921,
      "epoch": 1.3285134449518012,
      "grad_norm": 5.46875,
      "learning_rate": 2.7779588070009767e-06,
      "loss": 1.58471775,
      "memory(GiB)": 117.38,
      "step": 52370,
      "train_speed(iter/s)": 1.636528
    },
    {
      "acc": 0.66250114,
      "epoch": 1.3286402841197362,
      "grad_norm": 5.5,
      "learning_rate": 2.7770194732194256e-06,
      "loss": 1.57993488,
      "memory(GiB)": 117.38,
      "step": 52375,
      "train_speed(iter/s)": 1.636545
    },
    {
      "acc": 0.65336142,
      "epoch": 1.3287671232876712,
      "grad_norm": 4.78125,
      "learning_rate": 2.7760802372144825e-06,
      "loss": 1.65320168,
      "memory(GiB)": 117.38,
      "step": 52380,
      "train_speed(iter/s)": 1.636563
    },
    {
      "acc": 0.66621313,
      "epoch": 1.3288939624556062,
      "grad_norm": 6.65625,
      "learning_rate": 2.7751410990274596e-06,
      "loss": 1.56870127,
      "memory(GiB)": 117.38,
      "step": 52385,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.65803304,
      "epoch": 1.3290208016235414,
      "grad_norm": 6.625,
      "learning_rate": 2.774202058699664e-06,
      "loss": 1.53034534,
      "memory(GiB)": 117.38,
      "step": 52390,
      "train_speed(iter/s)": 1.636596
    },
    {
      "acc": 0.6521492,
      "epoch": 1.3291476407914764,
      "grad_norm": 4.96875,
      "learning_rate": 2.7732631162724005e-06,
      "loss": 1.57482338,
      "memory(GiB)": 117.38,
      "step": 52395,
      "train_speed(iter/s)": 1.636611
    },
    {
      "acc": 0.65294418,
      "epoch": 1.3292744799594114,
      "grad_norm": 4.59375,
      "learning_rate": 2.772324271786966e-06,
      "loss": 1.61312847,
      "memory(GiB)": 117.38,
      "step": 52400,
      "train_speed(iter/s)": 1.636627
    },
    {
      "acc": 0.65981736,
      "epoch": 1.3294013191273466,
      "grad_norm": 6.0,
      "learning_rate": 2.7713855252846545e-06,
      "loss": 1.58860378,
      "memory(GiB)": 117.38,
      "step": 52405,
      "train_speed(iter/s)": 1.636643
    },
    {
      "acc": 0.66145015,
      "epoch": 1.3295281582952816,
      "grad_norm": 7.4375,
      "learning_rate": 2.7704468768067616e-06,
      "loss": 1.55884542,
      "memory(GiB)": 117.38,
      "step": 52410,
      "train_speed(iter/s)": 1.63666
    },
    {
      "acc": 0.64659901,
      "epoch": 1.3296549974632166,
      "grad_norm": 7.34375,
      "learning_rate": 2.7695083263945664e-06,
      "loss": 1.66289177,
      "memory(GiB)": 117.38,
      "step": 52415,
      "train_speed(iter/s)": 1.636676
    },
    {
      "acc": 0.64792838,
      "epoch": 1.3297818366311516,
      "grad_norm": 6.09375,
      "learning_rate": 2.7685698740893516e-06,
      "loss": 1.68667831,
      "memory(GiB)": 117.38,
      "step": 52420,
      "train_speed(iter/s)": 1.636694
    },
    {
      "acc": 0.65250673,
      "epoch": 1.3299086757990868,
      "grad_norm": 6.28125,
      "learning_rate": 2.7676315199323995e-06,
      "loss": 1.5916996,
      "memory(GiB)": 117.38,
      "step": 52425,
      "train_speed(iter/s)": 1.636711
    },
    {
      "acc": 0.6755506,
      "epoch": 1.3300355149670218,
      "grad_norm": 5.8125,
      "learning_rate": 2.7666932639649814e-06,
      "loss": 1.50242176,
      "memory(GiB)": 117.38,
      "step": 52430,
      "train_speed(iter/s)": 1.636729
    },
    {
      "acc": 0.64364338,
      "epoch": 1.330162354134957,
      "grad_norm": 5.375,
      "learning_rate": 2.765755106228362e-06,
      "loss": 1.60989609,
      "memory(GiB)": 117.38,
      "step": 52435,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.64933214,
      "epoch": 1.330289193302892,
      "grad_norm": 6.1875,
      "learning_rate": 2.764817046763807e-06,
      "loss": 1.59384193,
      "memory(GiB)": 117.38,
      "step": 52440,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.6595407,
      "epoch": 1.330416032470827,
      "grad_norm": 5.78125,
      "learning_rate": 2.7638790856125786e-06,
      "loss": 1.64224262,
      "memory(GiB)": 117.38,
      "step": 52445,
      "train_speed(iter/s)": 1.636779
    },
    {
      "acc": 0.65388603,
      "epoch": 1.330542871638762,
      "grad_norm": 5.84375,
      "learning_rate": 2.7629412228159346e-06,
      "loss": 1.64019585,
      "memory(GiB)": 117.38,
      "step": 52450,
      "train_speed(iter/s)": 1.636795
    },
    {
      "acc": 0.64686003,
      "epoch": 1.3306697108066972,
      "grad_norm": 7.28125,
      "learning_rate": 2.762003458415119e-06,
      "loss": 1.65340157,
      "memory(GiB)": 117.38,
      "step": 52455,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.66051583,
      "epoch": 1.3307965499746321,
      "grad_norm": 5.375,
      "learning_rate": 2.7610657924513853e-06,
      "loss": 1.54742651,
      "memory(GiB)": 117.38,
      "step": 52460,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.66580324,
      "epoch": 1.3309233891425674,
      "grad_norm": 5.84375,
      "learning_rate": 2.7601282249659737e-06,
      "loss": 1.61788673,
      "memory(GiB)": 117.38,
      "step": 52465,
      "train_speed(iter/s)": 1.636846
    },
    {
      "acc": 0.64221544,
      "epoch": 1.3310502283105023,
      "grad_norm": 5.90625,
      "learning_rate": 2.759190756000126e-06,
      "loss": 1.67348537,
      "memory(GiB)": 117.38,
      "step": 52470,
      "train_speed(iter/s)": 1.636862
    },
    {
      "acc": 0.64889803,
      "epoch": 1.3311770674784373,
      "grad_norm": 6.9375,
      "learning_rate": 2.7582533855950687e-06,
      "loss": 1.63913956,
      "memory(GiB)": 117.38,
      "step": 52475,
      "train_speed(iter/s)": 1.636878
    },
    {
      "acc": 0.6537281,
      "epoch": 1.3313039066463723,
      "grad_norm": 5.34375,
      "learning_rate": 2.757316113792038e-06,
      "loss": 1.63948975,
      "memory(GiB)": 117.38,
      "step": 52480,
      "train_speed(iter/s)": 1.636896
    },
    {
      "acc": 0.63775034,
      "epoch": 1.3314307458143075,
      "grad_norm": 4.8125,
      "learning_rate": 2.756378940632258e-06,
      "loss": 1.65937729,
      "memory(GiB)": 117.38,
      "step": 52485,
      "train_speed(iter/s)": 1.636914
    },
    {
      "acc": 0.64764094,
      "epoch": 1.3315575849822425,
      "grad_norm": 6.71875,
      "learning_rate": 2.755441866156949e-06,
      "loss": 1.65567741,
      "memory(GiB)": 117.38,
      "step": 52490,
      "train_speed(iter/s)": 1.63693
    },
    {
      "acc": 0.66028323,
      "epoch": 1.3316844241501775,
      "grad_norm": 5.40625,
      "learning_rate": 2.7545048904073278e-06,
      "loss": 1.62102776,
      "memory(GiB)": 117.38,
      "step": 52495,
      "train_speed(iter/s)": 1.636947
    },
    {
      "acc": 0.66409416,
      "epoch": 1.3318112633181127,
      "grad_norm": 5.53125,
      "learning_rate": 2.7535680134246067e-06,
      "loss": 1.57515907,
      "memory(GiB)": 117.38,
      "step": 52500,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.67666497,
      "epoch": 1.3319381024860477,
      "grad_norm": 5.15625,
      "learning_rate": 2.752631235249995e-06,
      "loss": 1.61476402,
      "memory(GiB)": 117.38,
      "step": 52505,
      "train_speed(iter/s)": 1.63698
    },
    {
      "acc": 0.66616154,
      "epoch": 1.3320649416539827,
      "grad_norm": 8.1875,
      "learning_rate": 2.7516945559246945e-06,
      "loss": 1.49781132,
      "memory(GiB)": 117.38,
      "step": 52510,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.66098967,
      "epoch": 1.3321917808219177,
      "grad_norm": 5.5,
      "learning_rate": 2.7507579754899053e-06,
      "loss": 1.57758045,
      "memory(GiB)": 117.38,
      "step": 52515,
      "train_speed(iter/s)": 1.637014
    },
    {
      "acc": 0.67135019,
      "epoch": 1.332318619989853,
      "grad_norm": 6.15625,
      "learning_rate": 2.749821493986823e-06,
      "loss": 1.50583439,
      "memory(GiB)": 117.38,
      "step": 52520,
      "train_speed(iter/s)": 1.637031
    },
    {
      "acc": 0.66961088,
      "epoch": 1.332445459157788,
      "grad_norm": 6.4375,
      "learning_rate": 2.748885111456637e-06,
      "loss": 1.60621929,
      "memory(GiB)": 117.38,
      "step": 52525,
      "train_speed(iter/s)": 1.637049
    },
    {
      "acc": 0.66691132,
      "epoch": 1.332572298325723,
      "grad_norm": 6.375,
      "learning_rate": 2.7479488279405354e-06,
      "loss": 1.57460241,
      "memory(GiB)": 117.38,
      "step": 52530,
      "train_speed(iter/s)": 1.637067
    },
    {
      "acc": 0.65898952,
      "epoch": 1.332699137493658,
      "grad_norm": 6.15625,
      "learning_rate": 2.7470126434796984e-06,
      "loss": 1.57970161,
      "memory(GiB)": 117.38,
      "step": 52535,
      "train_speed(iter/s)": 1.637085
    },
    {
      "acc": 0.63932762,
      "epoch": 1.332825976661593,
      "grad_norm": 5.625,
      "learning_rate": 2.746076558115304e-06,
      "loss": 1.62135773,
      "memory(GiB)": 117.38,
      "step": 52540,
      "train_speed(iter/s)": 1.6371
    },
    {
      "acc": 0.66982231,
      "epoch": 1.332952815829528,
      "grad_norm": 5.375,
      "learning_rate": 2.7451405718885237e-06,
      "loss": 1.54981756,
      "memory(GiB)": 117.38,
      "step": 52545,
      "train_speed(iter/s)": 1.637117
    },
    {
      "acc": 0.65233712,
      "epoch": 1.3330796549974633,
      "grad_norm": 5.625,
      "learning_rate": 2.7442046848405328e-06,
      "loss": 1.5745225,
      "memory(GiB)": 117.38,
      "step": 52550,
      "train_speed(iter/s)": 1.637134
    },
    {
      "acc": 0.66620231,
      "epoch": 1.3332064941653983,
      "grad_norm": 5.15625,
      "learning_rate": 2.743268897012489e-06,
      "loss": 1.54645576,
      "memory(GiB)": 117.38,
      "step": 52555,
      "train_speed(iter/s)": 1.637151
    },
    {
      "acc": 0.65001745,
      "epoch": 1.3333333333333333,
      "grad_norm": 6.0,
      "learning_rate": 2.7423332084455543e-06,
      "loss": 1.62158928,
      "memory(GiB)": 117.38,
      "step": 52560,
      "train_speed(iter/s)": 1.637168
    },
    {
      "acc": 0.67157116,
      "epoch": 1.3334601725012685,
      "grad_norm": 5.78125,
      "learning_rate": 2.741397619180883e-06,
      "loss": 1.49711065,
      "memory(GiB)": 117.38,
      "step": 52565,
      "train_speed(iter/s)": 1.637186
    },
    {
      "acc": 0.64873924,
      "epoch": 1.3335870116692035,
      "grad_norm": 5.5,
      "learning_rate": 2.740462129259633e-06,
      "loss": 1.65579453,
      "memory(GiB)": 117.38,
      "step": 52570,
      "train_speed(iter/s)": 1.637204
    },
    {
      "acc": 0.65199451,
      "epoch": 1.3337138508371384,
      "grad_norm": 5.28125,
      "learning_rate": 2.739526738722944e-06,
      "loss": 1.66223907,
      "memory(GiB)": 117.38,
      "step": 52575,
      "train_speed(iter/s)": 1.637221
    },
    {
      "acc": 0.65989714,
      "epoch": 1.3338406900050734,
      "grad_norm": 5.3125,
      "learning_rate": 2.738591447611959e-06,
      "loss": 1.54689894,
      "memory(GiB)": 117.38,
      "step": 52580,
      "train_speed(iter/s)": 1.637238
    },
    {
      "acc": 0.6383184,
      "epoch": 1.3339675291730086,
      "grad_norm": 5.78125,
      "learning_rate": 2.7376562559678214e-06,
      "loss": 1.65324364,
      "memory(GiB)": 117.38,
      "step": 52585,
      "train_speed(iter/s)": 1.637255
    },
    {
      "acc": 0.65521321,
      "epoch": 1.3340943683409436,
      "grad_norm": 6.53125,
      "learning_rate": 2.7367211638316637e-06,
      "loss": 1.64562035,
      "memory(GiB)": 117.38,
      "step": 52590,
      "train_speed(iter/s)": 1.637272
    },
    {
      "acc": 0.65895061,
      "epoch": 1.3342212075088788,
      "grad_norm": 5.0625,
      "learning_rate": 2.735786171244611e-06,
      "loss": 1.60423737,
      "memory(GiB)": 117.38,
      "step": 52595,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.64557638,
      "epoch": 1.3343480466768138,
      "grad_norm": 5.4375,
      "learning_rate": 2.7348512782477922e-06,
      "loss": 1.61355,
      "memory(GiB)": 117.38,
      "step": 52600,
      "train_speed(iter/s)": 1.637306
    },
    {
      "acc": 0.66050453,
      "epoch": 1.3344748858447488,
      "grad_norm": 7.15625,
      "learning_rate": 2.7339164848823287e-06,
      "loss": 1.58573818,
      "memory(GiB)": 117.38,
      "step": 52605,
      "train_speed(iter/s)": 1.637323
    },
    {
      "acc": 0.65355721,
      "epoch": 1.3346017250126838,
      "grad_norm": 5.78125,
      "learning_rate": 2.7329817911893365e-06,
      "loss": 1.63428097,
      "memory(GiB)": 117.38,
      "step": 52610,
      "train_speed(iter/s)": 1.637341
    },
    {
      "acc": 0.65812111,
      "epoch": 1.334728564180619,
      "grad_norm": 6.09375,
      "learning_rate": 2.7320471972099226e-06,
      "loss": 1.58594646,
      "memory(GiB)": 117.38,
      "step": 52615,
      "train_speed(iter/s)": 1.637358
    },
    {
      "acc": 0.6591898,
      "epoch": 1.334855403348554,
      "grad_norm": 5.5625,
      "learning_rate": 2.7311127029852007e-06,
      "loss": 1.59654484,
      "memory(GiB)": 117.38,
      "step": 52620,
      "train_speed(iter/s)": 1.637376
    },
    {
      "acc": 0.64393187,
      "epoch": 1.3349822425164892,
      "grad_norm": 5.40625,
      "learning_rate": 2.7301783085562726e-06,
      "loss": 1.65579853,
      "memory(GiB)": 117.38,
      "step": 52625,
      "train_speed(iter/s)": 1.637393
    },
    {
      "acc": 0.65317993,
      "epoch": 1.3351090816844242,
      "grad_norm": 5.59375,
      "learning_rate": 2.7292440139642364e-06,
      "loss": 1.60785484,
      "memory(GiB)": 117.38,
      "step": 52630,
      "train_speed(iter/s)": 1.637409
    },
    {
      "acc": 0.64856644,
      "epoch": 1.3352359208523592,
      "grad_norm": 5.96875,
      "learning_rate": 2.7283098192501855e-06,
      "loss": 1.59053631,
      "memory(GiB)": 117.38,
      "step": 52635,
      "train_speed(iter/s)": 1.637425
    },
    {
      "acc": 0.66258683,
      "epoch": 1.3353627600202942,
      "grad_norm": 6.0,
      "learning_rate": 2.7273757244552124e-06,
      "loss": 1.53788548,
      "memory(GiB)": 117.38,
      "step": 52640,
      "train_speed(iter/s)": 1.637442
    },
    {
      "acc": 0.65966253,
      "epoch": 1.3354895991882294,
      "grad_norm": 5.40625,
      "learning_rate": 2.726441729620401e-06,
      "loss": 1.56639271,
      "memory(GiB)": 117.38,
      "step": 52645,
      "train_speed(iter/s)": 1.637458
    },
    {
      "acc": 0.66232243,
      "epoch": 1.3356164383561644,
      "grad_norm": 5.5,
      "learning_rate": 2.725507834786833e-06,
      "loss": 1.65225945,
      "memory(GiB)": 117.38,
      "step": 52650,
      "train_speed(iter/s)": 1.637477
    },
    {
      "acc": 0.6623137,
      "epoch": 1.3357432775240994,
      "grad_norm": 6.71875,
      "learning_rate": 2.7245740399955857e-06,
      "loss": 1.59502268,
      "memory(GiB)": 117.38,
      "step": 52655,
      "train_speed(iter/s)": 1.637494
    },
    {
      "acc": 0.65993319,
      "epoch": 1.3358701166920346,
      "grad_norm": 5.375,
      "learning_rate": 2.72364034528773e-06,
      "loss": 1.58483095,
      "memory(GiB)": 117.38,
      "step": 52660,
      "train_speed(iter/s)": 1.637512
    },
    {
      "acc": 0.65926423,
      "epoch": 1.3359969558599696,
      "grad_norm": 5.625,
      "learning_rate": 2.722706750704337e-06,
      "loss": 1.59527245,
      "memory(GiB)": 117.38,
      "step": 52665,
      "train_speed(iter/s)": 1.637528
    },
    {
      "acc": 0.65151391,
      "epoch": 1.3361237950279046,
      "grad_norm": 7.1875,
      "learning_rate": 2.7217732562864673e-06,
      "loss": 1.61700287,
      "memory(GiB)": 117.38,
      "step": 52670,
      "train_speed(iter/s)": 1.637545
    },
    {
      "acc": 0.64332972,
      "epoch": 1.3362506341958396,
      "grad_norm": 6.9375,
      "learning_rate": 2.720839862075181e-06,
      "loss": 1.65035,
      "memory(GiB)": 117.38,
      "step": 52675,
      "train_speed(iter/s)": 1.637561
    },
    {
      "acc": 0.66069026,
      "epoch": 1.3363774733637748,
      "grad_norm": 5.0625,
      "learning_rate": 2.7199065681115344e-06,
      "loss": 1.61837101,
      "memory(GiB)": 117.38,
      "step": 52680,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.66102395,
      "epoch": 1.3365043125317098,
      "grad_norm": 6.1875,
      "learning_rate": 2.7189733744365742e-06,
      "loss": 1.60382118,
      "memory(GiB)": 117.38,
      "step": 52685,
      "train_speed(iter/s)": 1.637594
    },
    {
      "acc": 0.66155758,
      "epoch": 1.336631151699645,
      "grad_norm": 5.625,
      "learning_rate": 2.718040281091353e-06,
      "loss": 1.52586889,
      "memory(GiB)": 117.38,
      "step": 52690,
      "train_speed(iter/s)": 1.637611
    },
    {
      "acc": 0.65605564,
      "epoch": 1.33675799086758,
      "grad_norm": 5.53125,
      "learning_rate": 2.717107288116906e-06,
      "loss": 1.567558,
      "memory(GiB)": 117.38,
      "step": 52695,
      "train_speed(iter/s)": 1.637629
    },
    {
      "acc": 0.65056314,
      "epoch": 1.336884830035515,
      "grad_norm": 6.0,
      "learning_rate": 2.716174395554274e-06,
      "loss": 1.62484398,
      "memory(GiB)": 117.38,
      "step": 52700,
      "train_speed(iter/s)": 1.637645
    },
    {
      "acc": 0.6628602,
      "epoch": 1.33701166920345,
      "grad_norm": 4.90625,
      "learning_rate": 2.715241603444486e-06,
      "loss": 1.62239742,
      "memory(GiB)": 117.38,
      "step": 52705,
      "train_speed(iter/s)": 1.637663
    },
    {
      "acc": 0.65894337,
      "epoch": 1.3371385083713851,
      "grad_norm": 6.3125,
      "learning_rate": 2.714308911828577e-06,
      "loss": 1.56882267,
      "memory(GiB)": 117.38,
      "step": 52710,
      "train_speed(iter/s)": 1.637678
    },
    {
      "acc": 0.64816999,
      "epoch": 1.3372653475393201,
      "grad_norm": 6.03125,
      "learning_rate": 2.713376320747565e-06,
      "loss": 1.66252346,
      "memory(GiB)": 117.38,
      "step": 52715,
      "train_speed(iter/s)": 1.637694
    },
    {
      "acc": 0.65537429,
      "epoch": 1.3373921867072551,
      "grad_norm": 5.71875,
      "learning_rate": 2.7124438302424696e-06,
      "loss": 1.58291273,
      "memory(GiB)": 117.38,
      "step": 52720,
      "train_speed(iter/s)": 1.637711
    },
    {
      "acc": 0.65098572,
      "epoch": 1.3375190258751903,
      "grad_norm": 7.25,
      "learning_rate": 2.711511440354309e-06,
      "loss": 1.61649456,
      "memory(GiB)": 117.38,
      "step": 52725,
      "train_speed(iter/s)": 1.637727
    },
    {
      "acc": 0.65720758,
      "epoch": 1.3376458650431253,
      "grad_norm": 6.71875,
      "learning_rate": 2.710579151124095e-06,
      "loss": 1.6403759,
      "memory(GiB)": 117.38,
      "step": 52730,
      "train_speed(iter/s)": 1.637744
    },
    {
      "acc": 0.65557075,
      "epoch": 1.3377727042110603,
      "grad_norm": 5.5,
      "learning_rate": 2.70964696259283e-06,
      "loss": 1.56805096,
      "memory(GiB)": 117.38,
      "step": 52735,
      "train_speed(iter/s)": 1.637761
    },
    {
      "acc": 0.6476429,
      "epoch": 1.3378995433789953,
      "grad_norm": 5.375,
      "learning_rate": 2.7087148748015146e-06,
      "loss": 1.57338219,
      "memory(GiB)": 117.38,
      "step": 52740,
      "train_speed(iter/s)": 1.637777
    },
    {
      "acc": 0.63488717,
      "epoch": 1.3380263825469305,
      "grad_norm": 5.28125,
      "learning_rate": 2.7077828877911517e-06,
      "loss": 1.73790627,
      "memory(GiB)": 117.38,
      "step": 52745,
      "train_speed(iter/s)": 1.637794
    },
    {
      "acc": 0.64949837,
      "epoch": 1.3381532217148655,
      "grad_norm": 6.84375,
      "learning_rate": 2.706851001602733e-06,
      "loss": 1.55197601,
      "memory(GiB)": 117.38,
      "step": 52750,
      "train_speed(iter/s)": 1.637811
    },
    {
      "acc": 0.6568469,
      "epoch": 1.3382800608828007,
      "grad_norm": 5.03125,
      "learning_rate": 2.7059192162772407e-06,
      "loss": 1.60443554,
      "memory(GiB)": 117.38,
      "step": 52755,
      "train_speed(iter/s)": 1.637829
    },
    {
      "acc": 0.65990729,
      "epoch": 1.3384069000507357,
      "grad_norm": 6.3125,
      "learning_rate": 2.704987531855666e-06,
      "loss": 1.61475754,
      "memory(GiB)": 117.38,
      "step": 52760,
      "train_speed(iter/s)": 1.637847
    },
    {
      "acc": 0.66679478,
      "epoch": 1.3385337392186707,
      "grad_norm": 5.34375,
      "learning_rate": 2.704055948378986e-06,
      "loss": 1.60053635,
      "memory(GiB)": 117.38,
      "step": 52765,
      "train_speed(iter/s)": 1.637865
    },
    {
      "acc": 0.65497775,
      "epoch": 1.3386605783866057,
      "grad_norm": 5.9375,
      "learning_rate": 2.7031244658881773e-06,
      "loss": 1.557059,
      "memory(GiB)": 117.38,
      "step": 52770,
      "train_speed(iter/s)": 1.637883
    },
    {
      "acc": 0.65754519,
      "epoch": 1.338787417554541,
      "grad_norm": 7.65625,
      "learning_rate": 2.7021930844242085e-06,
      "loss": 1.62008839,
      "memory(GiB)": 117.38,
      "step": 52775,
      "train_speed(iter/s)": 1.637901
    },
    {
      "acc": 0.65888691,
      "epoch": 1.3389142567224759,
      "grad_norm": 7.21875,
      "learning_rate": 2.7012618040280463e-06,
      "loss": 1.61328239,
      "memory(GiB)": 117.38,
      "step": 52780,
      "train_speed(iter/s)": 1.637918
    },
    {
      "acc": 0.67575898,
      "epoch": 1.339041095890411,
      "grad_norm": 6.3125,
      "learning_rate": 2.7003306247406536e-06,
      "loss": 1.54389858,
      "memory(GiB)": 117.38,
      "step": 52785,
      "train_speed(iter/s)": 1.637936
    },
    {
      "acc": 0.65001764,
      "epoch": 1.339167935058346,
      "grad_norm": 5.40625,
      "learning_rate": 2.6993995466029877e-06,
      "loss": 1.69043427,
      "memory(GiB)": 117.38,
      "step": 52790,
      "train_speed(iter/s)": 1.637954
    },
    {
      "acc": 0.6416666,
      "epoch": 1.339294774226281,
      "grad_norm": 4.84375,
      "learning_rate": 2.6984685696560002e-06,
      "loss": 1.62393398,
      "memory(GiB)": 117.38,
      "step": 52795,
      "train_speed(iter/s)": 1.637971
    },
    {
      "acc": 0.65619736,
      "epoch": 1.339421613394216,
      "grad_norm": 5.28125,
      "learning_rate": 2.6975376939406418e-06,
      "loss": 1.5741333,
      "memory(GiB)": 117.38,
      "step": 52800,
      "train_speed(iter/s)": 1.637988
    },
    {
      "acc": 0.65987396,
      "epoch": 1.3395484525621513,
      "grad_norm": 5.21875,
      "learning_rate": 2.6966069194978537e-06,
      "loss": 1.56193695,
      "memory(GiB)": 117.38,
      "step": 52805,
      "train_speed(iter/s)": 1.638006
    },
    {
      "acc": 0.64126892,
      "epoch": 1.3396752917300863,
      "grad_norm": 6.15625,
      "learning_rate": 2.6956762463685787e-06,
      "loss": 1.65374451,
      "memory(GiB)": 117.38,
      "step": 52810,
      "train_speed(iter/s)": 1.638022
    },
    {
      "acc": 0.66136961,
      "epoch": 1.3398021308980212,
      "grad_norm": 5.25,
      "learning_rate": 2.69474567459375e-06,
      "loss": 1.63458233,
      "memory(GiB)": 117.38,
      "step": 52815,
      "train_speed(iter/s)": 1.638039
    },
    {
      "acc": 0.65879898,
      "epoch": 1.3399289700659565,
      "grad_norm": 6.4375,
      "learning_rate": 2.693815204214299e-06,
      "loss": 1.61339455,
      "memory(GiB)": 117.38,
      "step": 52820,
      "train_speed(iter/s)": 1.638056
    },
    {
      "acc": 0.65663776,
      "epoch": 1.3400558092338914,
      "grad_norm": 5.625,
      "learning_rate": 2.692884835271151e-06,
      "loss": 1.59304152,
      "memory(GiB)": 117.38,
      "step": 52825,
      "train_speed(iter/s)": 1.638074
    },
    {
      "acc": 0.67202492,
      "epoch": 1.3401826484018264,
      "grad_norm": 6.875,
      "learning_rate": 2.6919545678052296e-06,
      "loss": 1.58589973,
      "memory(GiB)": 117.38,
      "step": 52830,
      "train_speed(iter/s)": 1.638091
    },
    {
      "acc": 0.67168484,
      "epoch": 1.3403094875697614,
      "grad_norm": 6.84375,
      "learning_rate": 2.69102440185745e-06,
      "loss": 1.50302792,
      "memory(GiB)": 117.38,
      "step": 52835,
      "train_speed(iter/s)": 1.638108
    },
    {
      "acc": 0.64203653,
      "epoch": 1.3404363267376966,
      "grad_norm": 5.21875,
      "learning_rate": 2.690094337468726e-06,
      "loss": 1.73719349,
      "memory(GiB)": 117.38,
      "step": 52840,
      "train_speed(iter/s)": 1.638125
    },
    {
      "acc": 0.64537287,
      "epoch": 1.3405631659056316,
      "grad_norm": 6.65625,
      "learning_rate": 2.6891643746799643e-06,
      "loss": 1.62930698,
      "memory(GiB)": 117.38,
      "step": 52845,
      "train_speed(iter/s)": 1.638143
    },
    {
      "acc": 0.6239182,
      "epoch": 1.3406900050735668,
      "grad_norm": 5.0625,
      "learning_rate": 2.6882345135320753e-06,
      "loss": 1.65400143,
      "memory(GiB)": 117.38,
      "step": 52850,
      "train_speed(iter/s)": 1.638161
    },
    {
      "acc": 0.65918274,
      "epoch": 1.3408168442415018,
      "grad_norm": 5.375,
      "learning_rate": 2.68730475406595e-06,
      "loss": 1.62808533,
      "memory(GiB)": 117.38,
      "step": 52855,
      "train_speed(iter/s)": 1.638178
    },
    {
      "acc": 0.64438286,
      "epoch": 1.3409436834094368,
      "grad_norm": 5.5625,
      "learning_rate": 2.6863750963224867e-06,
      "loss": 1.66250687,
      "memory(GiB)": 117.38,
      "step": 52860,
      "train_speed(iter/s)": 1.638196
    },
    {
      "acc": 0.63586493,
      "epoch": 1.3410705225773718,
      "grad_norm": 5.84375,
      "learning_rate": 2.685445540342577e-06,
      "loss": 1.65475388,
      "memory(GiB)": 117.38,
      "step": 52865,
      "train_speed(iter/s)": 1.638212
    },
    {
      "acc": 0.64148946,
      "epoch": 1.341197361745307,
      "grad_norm": 6.15625,
      "learning_rate": 2.6845160861671094e-06,
      "loss": 1.70497818,
      "memory(GiB)": 117.38,
      "step": 52870,
      "train_speed(iter/s)": 1.63823
    },
    {
      "acc": 0.65571108,
      "epoch": 1.341324200913242,
      "grad_norm": 5.84375,
      "learning_rate": 2.6835867338369593e-06,
      "loss": 1.54213448,
      "memory(GiB)": 117.38,
      "step": 52875,
      "train_speed(iter/s)": 1.638248
    },
    {
      "acc": 0.656812,
      "epoch": 1.341451040081177,
      "grad_norm": 5.53125,
      "learning_rate": 2.6826574833930053e-06,
      "loss": 1.52983913,
      "memory(GiB)": 117.38,
      "step": 52880,
      "train_speed(iter/s)": 1.638266
    },
    {
      "acc": 0.64465132,
      "epoch": 1.3415778792491122,
      "grad_norm": 6.71875,
      "learning_rate": 2.681728334876123e-06,
      "loss": 1.6092041,
      "memory(GiB)": 117.38,
      "step": 52885,
      "train_speed(iter/s)": 1.638285
    },
    {
      "acc": 0.66320653,
      "epoch": 1.3417047184170472,
      "grad_norm": 6.4375,
      "learning_rate": 2.6807992883271806e-06,
      "loss": 1.51731644,
      "memory(GiB)": 117.38,
      "step": 52890,
      "train_speed(iter/s)": 1.638303
    },
    {
      "acc": 0.65563526,
      "epoch": 1.3418315575849822,
      "grad_norm": 6.46875,
      "learning_rate": 2.6798703437870364e-06,
      "loss": 1.5687912,
      "memory(GiB)": 117.38,
      "step": 52895,
      "train_speed(iter/s)": 1.638061
    },
    {
      "acc": 0.66147747,
      "epoch": 1.3419583967529172,
      "grad_norm": 5.90625,
      "learning_rate": 2.678941501296555e-06,
      "loss": 1.60652409,
      "memory(GiB)": 117.38,
      "step": 52900,
      "train_speed(iter/s)": 1.638081
    },
    {
      "acc": 0.67350202,
      "epoch": 1.3420852359208524,
      "grad_norm": 5.40625,
      "learning_rate": 2.6780127608965896e-06,
      "loss": 1.56545982,
      "memory(GiB)": 117.38,
      "step": 52905,
      "train_speed(iter/s)": 1.638098
    },
    {
      "acc": 0.66206641,
      "epoch": 1.3422120750887874,
      "grad_norm": 5.90625,
      "learning_rate": 2.677084122627991e-06,
      "loss": 1.59041185,
      "memory(GiB)": 117.38,
      "step": 52910,
      "train_speed(iter/s)": 1.637857
    },
    {
      "acc": 0.65431671,
      "epoch": 1.3423389142567226,
      "grad_norm": 4.28125,
      "learning_rate": 2.6761555865316003e-06,
      "loss": 1.66435547,
      "memory(GiB)": 117.38,
      "step": 52915,
      "train_speed(iter/s)": 1.637874
    },
    {
      "acc": 0.67400656,
      "epoch": 1.3424657534246576,
      "grad_norm": 6.59375,
      "learning_rate": 2.6752271526482644e-06,
      "loss": 1.60946808,
      "memory(GiB)": 117.38,
      "step": 52920,
      "train_speed(iter/s)": 1.637892
    },
    {
      "acc": 0.64252248,
      "epoch": 1.3425925925925926,
      "grad_norm": 5.75,
      "learning_rate": 2.674298821018817e-06,
      "loss": 1.65829601,
      "memory(GiB)": 117.38,
      "step": 52925,
      "train_speed(iter/s)": 1.637909
    },
    {
      "acc": 0.65550051,
      "epoch": 1.3427194317605275,
      "grad_norm": 6.8125,
      "learning_rate": 2.673370591684091e-06,
      "loss": 1.66814632,
      "memory(GiB)": 117.38,
      "step": 52930,
      "train_speed(iter/s)": 1.637926
    },
    {
      "acc": 0.64613943,
      "epoch": 1.3428462709284628,
      "grad_norm": 5.34375,
      "learning_rate": 2.672442464684915e-06,
      "loss": 1.68707199,
      "memory(GiB)": 117.38,
      "step": 52935,
      "train_speed(iter/s)": 1.637942
    },
    {
      "acc": 0.65846262,
      "epoch": 1.3429731100963977,
      "grad_norm": 4.78125,
      "learning_rate": 2.671514440062111e-06,
      "loss": 1.57678671,
      "memory(GiB)": 117.38,
      "step": 52940,
      "train_speed(iter/s)": 1.637959
    },
    {
      "acc": 0.66221523,
      "epoch": 1.343099949264333,
      "grad_norm": 5.6875,
      "learning_rate": 2.6705865178564973e-06,
      "loss": 1.5938448,
      "memory(GiB)": 117.38,
      "step": 52945,
      "train_speed(iter/s)": 1.637975
    },
    {
      "acc": 0.64873471,
      "epoch": 1.343226788432268,
      "grad_norm": 5.4375,
      "learning_rate": 2.6696586981088886e-06,
      "loss": 1.65513458,
      "memory(GiB)": 117.38,
      "step": 52950,
      "train_speed(iter/s)": 1.637992
    },
    {
      "acc": 0.64291525,
      "epoch": 1.343353627600203,
      "grad_norm": 6.96875,
      "learning_rate": 2.6687309808600947e-06,
      "loss": 1.65151024,
      "memory(GiB)": 117.38,
      "step": 52955,
      "train_speed(iter/s)": 1.63801
    },
    {
      "acc": 0.6530447,
      "epoch": 1.343480466768138,
      "grad_norm": 6.15625,
      "learning_rate": 2.6678033661509208e-06,
      "loss": 1.62684059,
      "memory(GiB)": 117.38,
      "step": 52960,
      "train_speed(iter/s)": 1.638027
    },
    {
      "acc": 0.64571304,
      "epoch": 1.3436073059360731,
      "grad_norm": 5.78125,
      "learning_rate": 2.6668758540221665e-06,
      "loss": 1.66499176,
      "memory(GiB)": 117.38,
      "step": 52965,
      "train_speed(iter/s)": 1.638044
    },
    {
      "acc": 0.65314083,
      "epoch": 1.3437341451040081,
      "grad_norm": 6.28125,
      "learning_rate": 2.66594844451463e-06,
      "loss": 1.55702152,
      "memory(GiB)": 117.38,
      "step": 52970,
      "train_speed(iter/s)": 1.63806
    },
    {
      "acc": 0.65581408,
      "epoch": 1.3438609842719431,
      "grad_norm": 8.0,
      "learning_rate": 2.6650211376691008e-06,
      "loss": 1.70864468,
      "memory(GiB)": 117.38,
      "step": 52975,
      "train_speed(iter/s)": 1.638077
    },
    {
      "acc": 0.65982456,
      "epoch": 1.3439878234398783,
      "grad_norm": 4.8125,
      "learning_rate": 2.664093933526368e-06,
      "loss": 1.59009027,
      "memory(GiB)": 117.38,
      "step": 52980,
      "train_speed(iter/s)": 1.638093
    },
    {
      "acc": 0.6477129,
      "epoch": 1.3441146626078133,
      "grad_norm": 5.6875,
      "learning_rate": 2.6631668321272097e-06,
      "loss": 1.64277306,
      "memory(GiB)": 117.38,
      "step": 52985,
      "train_speed(iter/s)": 1.638111
    },
    {
      "acc": 0.64682226,
      "epoch": 1.3442415017757483,
      "grad_norm": 6.90625,
      "learning_rate": 2.6622398335124122e-06,
      "loss": 1.62259197,
      "memory(GiB)": 117.38,
      "step": 52990,
      "train_speed(iter/s)": 1.638128
    },
    {
      "acc": 0.6515841,
      "epoch": 1.3443683409436833,
      "grad_norm": 5.4375,
      "learning_rate": 2.661312937722742e-06,
      "loss": 1.62858467,
      "memory(GiB)": 117.38,
      "step": 52995,
      "train_speed(iter/s)": 1.638145
    },
    {
      "acc": 0.65542135,
      "epoch": 1.3444951801116185,
      "grad_norm": 5.0625,
      "learning_rate": 2.6603861447989703e-06,
      "loss": 1.57048187,
      "memory(GiB)": 117.38,
      "step": 53000,
      "train_speed(iter/s)": 1.638163
    },
    {
      "epoch": 1.3444951801116185,
      "eval_acc": 0.6462505507907371,
      "eval_loss": 1.5732170343399048,
      "eval_runtime": 59.1654,
      "eval_samples_per_second": 107.664,
      "eval_steps_per_second": 26.925,
      "step": 53000
    },
    {
      "acc": 0.65749884,
      "epoch": 1.3446220192795535,
      "grad_norm": 5.875,
      "learning_rate": 2.65945945478186e-06,
      "loss": 1.5651413,
      "memory(GiB)": 117.38,
      "step": 53005,
      "train_speed(iter/s)": 1.63497
    },
    {
      "acc": 0.6618412,
      "epoch": 1.3447488584474887,
      "grad_norm": 5.65625,
      "learning_rate": 2.658532867712176e-06,
      "loss": 1.59753542,
      "memory(GiB)": 117.38,
      "step": 53010,
      "train_speed(iter/s)": 1.634986
    },
    {
      "acc": 0.6620615,
      "epoch": 1.3448756976154237,
      "grad_norm": 5.46875,
      "learning_rate": 2.6576063836306687e-06,
      "loss": 1.55156517,
      "memory(GiB)": 117.38,
      "step": 53015,
      "train_speed(iter/s)": 1.635002
    },
    {
      "acc": 0.65507374,
      "epoch": 1.3450025367833587,
      "grad_norm": 7.15625,
      "learning_rate": 2.656680002578088e-06,
      "loss": 1.60954323,
      "memory(GiB)": 117.38,
      "step": 53020,
      "train_speed(iter/s)": 1.63502
    },
    {
      "acc": 0.65546637,
      "epoch": 1.3451293759512937,
      "grad_norm": 5.28125,
      "learning_rate": 2.655753724595186e-06,
      "loss": 1.56377649,
      "memory(GiB)": 117.38,
      "step": 53025,
      "train_speed(iter/s)": 1.635037
    },
    {
      "acc": 0.64627085,
      "epoch": 1.3452562151192289,
      "grad_norm": 5.84375,
      "learning_rate": 2.6548275497227028e-06,
      "loss": 1.58946438,
      "memory(GiB)": 117.38,
      "step": 53030,
      "train_speed(iter/s)": 1.635054
    },
    {
      "acc": 0.66005816,
      "epoch": 1.3453830542871639,
      "grad_norm": 5.09375,
      "learning_rate": 2.6539014780013707e-06,
      "loss": 1.58129702,
      "memory(GiB)": 117.38,
      "step": 53035,
      "train_speed(iter/s)": 1.635071
    },
    {
      "acc": 0.64971399,
      "epoch": 1.3455098934550989,
      "grad_norm": 6.0,
      "learning_rate": 2.6529755094719276e-06,
      "loss": 1.66034966,
      "memory(GiB)": 117.38,
      "step": 53040,
      "train_speed(iter/s)": 1.635088
    },
    {
      "acc": 0.64803171,
      "epoch": 1.345636732623034,
      "grad_norm": 6.21875,
      "learning_rate": 2.652049644175101e-06,
      "loss": 1.63495255,
      "memory(GiB)": 117.38,
      "step": 53045,
      "train_speed(iter/s)": 1.635105
    },
    {
      "acc": 0.65598598,
      "epoch": 1.345763571790969,
      "grad_norm": 4.84375,
      "learning_rate": 2.6511238821516154e-06,
      "loss": 1.5769042,
      "memory(GiB)": 117.38,
      "step": 53050,
      "train_speed(iter/s)": 1.635122
    },
    {
      "acc": 0.65694847,
      "epoch": 1.345890410958904,
      "grad_norm": 5.1875,
      "learning_rate": 2.650198223442185e-06,
      "loss": 1.55485935,
      "memory(GiB)": 117.38,
      "step": 53055,
      "train_speed(iter/s)": 1.635138
    },
    {
      "acc": 0.648349,
      "epoch": 1.346017250126839,
      "grad_norm": 5.71875,
      "learning_rate": 2.6492726680875296e-06,
      "loss": 1.6262619,
      "memory(GiB)": 117.38,
      "step": 53060,
      "train_speed(iter/s)": 1.635156
    },
    {
      "acc": 0.67691779,
      "epoch": 1.3461440892947742,
      "grad_norm": 6.09375,
      "learning_rate": 2.6483472161283576e-06,
      "loss": 1.57359123,
      "memory(GiB)": 117.38,
      "step": 53065,
      "train_speed(iter/s)": 1.635173
    },
    {
      "acc": 0.65168762,
      "epoch": 1.3462709284627092,
      "grad_norm": 6.25,
      "learning_rate": 2.647421867605374e-06,
      "loss": 1.57416496,
      "memory(GiB)": 117.38,
      "step": 53070,
      "train_speed(iter/s)": 1.63519
    },
    {
      "acc": 0.65546989,
      "epoch": 1.3463977676306444,
      "grad_norm": 6.15625,
      "learning_rate": 2.6464966225592804e-06,
      "loss": 1.59917831,
      "memory(GiB)": 117.38,
      "step": 53075,
      "train_speed(iter/s)": 1.635207
    },
    {
      "acc": 0.64507484,
      "epoch": 1.3465246067985794,
      "grad_norm": 6.40625,
      "learning_rate": 2.645571481030773e-06,
      "loss": 1.65069351,
      "memory(GiB)": 117.38,
      "step": 53080,
      "train_speed(iter/s)": 1.635225
    },
    {
      "acc": 0.66880908,
      "epoch": 1.3466514459665144,
      "grad_norm": 7.59375,
      "learning_rate": 2.6446464430605434e-06,
      "loss": 1.58419876,
      "memory(GiB)": 117.38,
      "step": 53085,
      "train_speed(iter/s)": 1.635242
    },
    {
      "acc": 0.65629783,
      "epoch": 1.3467782851344494,
      "grad_norm": 5.1875,
      "learning_rate": 2.6437215086892797e-06,
      "loss": 1.57410336,
      "memory(GiB)": 117.38,
      "step": 53090,
      "train_speed(iter/s)": 1.635259
    },
    {
      "acc": 0.66677265,
      "epoch": 1.3469051243023846,
      "grad_norm": 6.03125,
      "learning_rate": 2.642796677957664e-06,
      "loss": 1.57001553,
      "memory(GiB)": 117.38,
      "step": 53095,
      "train_speed(iter/s)": 1.635277
    },
    {
      "acc": 0.64633703,
      "epoch": 1.3470319634703196,
      "grad_norm": 5.15625,
      "learning_rate": 2.641871950906374e-06,
      "loss": 1.65141735,
      "memory(GiB)": 117.38,
      "step": 53100,
      "train_speed(iter/s)": 1.635296
    },
    {
      "acc": 0.65298882,
      "epoch": 1.3471588026382548,
      "grad_norm": 4.96875,
      "learning_rate": 2.6409473275760843e-06,
      "loss": 1.62078323,
      "memory(GiB)": 117.38,
      "step": 53105,
      "train_speed(iter/s)": 1.635313
    },
    {
      "acc": 0.65995712,
      "epoch": 1.3472856418061898,
      "grad_norm": 5.84375,
      "learning_rate": 2.640022808007463e-06,
      "loss": 1.67465019,
      "memory(GiB)": 117.38,
      "step": 53110,
      "train_speed(iter/s)": 1.63533
    },
    {
      "acc": 0.65148191,
      "epoch": 1.3474124809741248,
      "grad_norm": 6.0,
      "learning_rate": 2.639098392241176e-06,
      "loss": 1.6521965,
      "memory(GiB)": 117.38,
      "step": 53115,
      "train_speed(iter/s)": 1.635347
    },
    {
      "acc": 0.64239984,
      "epoch": 1.3475393201420598,
      "grad_norm": 5.5625,
      "learning_rate": 2.6381740803178826e-06,
      "loss": 1.61989326,
      "memory(GiB)": 117.38,
      "step": 53120,
      "train_speed(iter/s)": 1.635364
    },
    {
      "acc": 0.65592341,
      "epoch": 1.347666159309995,
      "grad_norm": 5.46875,
      "learning_rate": 2.6372498722782346e-06,
      "loss": 1.58394699,
      "memory(GiB)": 117.38,
      "step": 53125,
      "train_speed(iter/s)": 1.635382
    },
    {
      "acc": 0.65077953,
      "epoch": 1.34779299847793,
      "grad_norm": 5.28125,
      "learning_rate": 2.6363257681628907e-06,
      "loss": 1.62168217,
      "memory(GiB)": 117.38,
      "step": 53130,
      "train_speed(iter/s)": 1.635398
    },
    {
      "acc": 0.66489758,
      "epoch": 1.347919837645865,
      "grad_norm": 8.4375,
      "learning_rate": 2.635401768012491e-06,
      "loss": 1.55354099,
      "memory(GiB)": 117.38,
      "step": 53135,
      "train_speed(iter/s)": 1.635415
    },
    {
      "acc": 0.66087213,
      "epoch": 1.3480466768138002,
      "grad_norm": 6.5625,
      "learning_rate": 2.6344778718676783e-06,
      "loss": 1.56101646,
      "memory(GiB)": 117.38,
      "step": 53140,
      "train_speed(iter/s)": 1.63543
    },
    {
      "acc": 0.64854054,
      "epoch": 1.3481735159817352,
      "grad_norm": 6.625,
      "learning_rate": 2.6335540797690886e-06,
      "loss": 1.62264214,
      "memory(GiB)": 117.38,
      "step": 53145,
      "train_speed(iter/s)": 1.635447
    },
    {
      "acc": 0.67462988,
      "epoch": 1.3483003551496702,
      "grad_norm": 5.03125,
      "learning_rate": 2.632630391757359e-06,
      "loss": 1.54717245,
      "memory(GiB)": 117.38,
      "step": 53150,
      "train_speed(iter/s)": 1.635464
    },
    {
      "acc": 0.64381123,
      "epoch": 1.3484271943176052,
      "grad_norm": 5.96875,
      "learning_rate": 2.6317068078731126e-06,
      "loss": 1.554562,
      "memory(GiB)": 117.38,
      "step": 53155,
      "train_speed(iter/s)": 1.63548
    },
    {
      "acc": 0.66206274,
      "epoch": 1.3485540334855404,
      "grad_norm": 5.46875,
      "learning_rate": 2.630783328156973e-06,
      "loss": 1.52031784,
      "memory(GiB)": 117.38,
      "step": 53160,
      "train_speed(iter/s)": 1.635498
    },
    {
      "acc": 0.64649653,
      "epoch": 1.3486808726534754,
      "grad_norm": 5.78125,
      "learning_rate": 2.629859952649562e-06,
      "loss": 1.57510061,
      "memory(GiB)": 117.38,
      "step": 53165,
      "train_speed(iter/s)": 1.635515
    },
    {
      "acc": 0.67456112,
      "epoch": 1.3488077118214106,
      "grad_norm": 6.0,
      "learning_rate": 2.628936681391494e-06,
      "loss": 1.53614311,
      "memory(GiB)": 117.38,
      "step": 53170,
      "train_speed(iter/s)": 1.635533
    },
    {
      "acc": 0.65889921,
      "epoch": 1.3489345509893456,
      "grad_norm": 6.0625,
      "learning_rate": 2.628013514423375e-06,
      "loss": 1.65717278,
      "memory(GiB)": 117.38,
      "step": 53175,
      "train_speed(iter/s)": 1.635548
    },
    {
      "acc": 0.64561205,
      "epoch": 1.3490613901572805,
      "grad_norm": 5.625,
      "learning_rate": 2.6270904517858102e-06,
      "loss": 1.651898,
      "memory(GiB)": 117.38,
      "step": 53180,
      "train_speed(iter/s)": 1.635565
    },
    {
      "acc": 0.65892339,
      "epoch": 1.3491882293252155,
      "grad_norm": 6.90625,
      "learning_rate": 2.6261674935194036e-06,
      "loss": 1.60423622,
      "memory(GiB)": 117.38,
      "step": 53185,
      "train_speed(iter/s)": 1.635582
    },
    {
      "acc": 0.66684794,
      "epoch": 1.3493150684931507,
      "grad_norm": 6.59375,
      "learning_rate": 2.6252446396647503e-06,
      "loss": 1.49408035,
      "memory(GiB)": 117.38,
      "step": 53190,
      "train_speed(iter/s)": 1.635599
    },
    {
      "acc": 0.63280678,
      "epoch": 1.3494419076610857,
      "grad_norm": 5.5,
      "learning_rate": 2.6243218902624367e-06,
      "loss": 1.68768597,
      "memory(GiB)": 117.38,
      "step": 53195,
      "train_speed(iter/s)": 1.635616
    },
    {
      "acc": 0.65492496,
      "epoch": 1.3495687468290207,
      "grad_norm": 5.84375,
      "learning_rate": 2.6233992453530555e-06,
      "loss": 1.58912382,
      "memory(GiB)": 117.38,
      "step": 53200,
      "train_speed(iter/s)": 1.635631
    },
    {
      "acc": 0.67314806,
      "epoch": 1.349695585996956,
      "grad_norm": 5.03125,
      "learning_rate": 2.6224767049771856e-06,
      "loss": 1.52840376,
      "memory(GiB)": 117.38,
      "step": 53205,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.63613391,
      "epoch": 1.349822425164891,
      "grad_norm": 6.875,
      "learning_rate": 2.621554269175405e-06,
      "loss": 1.68029366,
      "memory(GiB)": 117.38,
      "step": 53210,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.63658056,
      "epoch": 1.349949264332826,
      "grad_norm": 6.03125,
      "learning_rate": 2.620631937988287e-06,
      "loss": 1.69095383,
      "memory(GiB)": 117.38,
      "step": 53215,
      "train_speed(iter/s)": 1.635682
    },
    {
      "acc": 0.64621515,
      "epoch": 1.350076103500761,
      "grad_norm": 5.84375,
      "learning_rate": 2.6197097114564e-06,
      "loss": 1.62177086,
      "memory(GiB)": 117.38,
      "step": 53220,
      "train_speed(iter/s)": 1.635698
    },
    {
      "acc": 0.6440197,
      "epoch": 1.3502029426686961,
      "grad_norm": 5.75,
      "learning_rate": 2.618787589620306e-06,
      "loss": 1.63779526,
      "memory(GiB)": 117.38,
      "step": 53225,
      "train_speed(iter/s)": 1.635716
    },
    {
      "acc": 0.66838298,
      "epoch": 1.350329781836631,
      "grad_norm": 6.46875,
      "learning_rate": 2.6178655725205653e-06,
      "loss": 1.54495974,
      "memory(GiB)": 117.38,
      "step": 53230,
      "train_speed(iter/s)": 1.635733
    },
    {
      "acc": 0.67129593,
      "epoch": 1.3504566210045663,
      "grad_norm": 6.46875,
      "learning_rate": 2.6169436601977325e-06,
      "loss": 1.5674818,
      "memory(GiB)": 117.38,
      "step": 53235,
      "train_speed(iter/s)": 1.635749
    },
    {
      "acc": 0.65012026,
      "epoch": 1.3505834601725013,
      "grad_norm": 5.03125,
      "learning_rate": 2.6160218526923576e-06,
      "loss": 1.64780731,
      "memory(GiB)": 117.38,
      "step": 53240,
      "train_speed(iter/s)": 1.635765
    },
    {
      "acc": 0.63448963,
      "epoch": 1.3507102993404363,
      "grad_norm": 7.28125,
      "learning_rate": 2.6151001500449847e-06,
      "loss": 1.67943039,
      "memory(GiB)": 117.38,
      "step": 53245,
      "train_speed(iter/s)": 1.635783
    },
    {
      "acc": 0.65939102,
      "epoch": 1.3508371385083713,
      "grad_norm": 4.96875,
      "learning_rate": 2.614178552296155e-06,
      "loss": 1.62487183,
      "memory(GiB)": 117.38,
      "step": 53250,
      "train_speed(iter/s)": 1.635798
    },
    {
      "acc": 0.6523737,
      "epoch": 1.3509639776763065,
      "grad_norm": 6.03125,
      "learning_rate": 2.6132570594864047e-06,
      "loss": 1.63586292,
      "memory(GiB)": 117.38,
      "step": 53255,
      "train_speed(iter/s)": 1.635815
    },
    {
      "acc": 0.67681699,
      "epoch": 1.3510908168442415,
      "grad_norm": 5.375,
      "learning_rate": 2.612335671656265e-06,
      "loss": 1.61519775,
      "memory(GiB)": 117.38,
      "step": 53260,
      "train_speed(iter/s)": 1.635831
    },
    {
      "acc": 0.68297505,
      "epoch": 1.3512176560121767,
      "grad_norm": 6.03125,
      "learning_rate": 2.6114143888462607e-06,
      "loss": 1.52014666,
      "memory(GiB)": 117.38,
      "step": 53265,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.661343,
      "epoch": 1.3513444951801117,
      "grad_norm": 6.0625,
      "learning_rate": 2.6104932110969195e-06,
      "loss": 1.61415863,
      "memory(GiB)": 117.38,
      "step": 53270,
      "train_speed(iter/s)": 1.635862
    },
    {
      "acc": 0.65879097,
      "epoch": 1.3514713343480467,
      "grad_norm": 5.15625,
      "learning_rate": 2.609572138448753e-06,
      "loss": 1.55318985,
      "memory(GiB)": 117.38,
      "step": 53275,
      "train_speed(iter/s)": 1.635879
    },
    {
      "acc": 0.65854053,
      "epoch": 1.3515981735159817,
      "grad_norm": 5.15625,
      "learning_rate": 2.608651170942277e-06,
      "loss": 1.61781635,
      "memory(GiB)": 117.38,
      "step": 53280,
      "train_speed(iter/s)": 1.635896
    },
    {
      "acc": 0.662182,
      "epoch": 1.3517250126839169,
      "grad_norm": 5.3125,
      "learning_rate": 2.607730308617997e-06,
      "loss": 1.56530571,
      "memory(GiB)": 117.38,
      "step": 53285,
      "train_speed(iter/s)": 1.635914
    },
    {
      "acc": 0.67736173,
      "epoch": 1.3518518518518519,
      "grad_norm": 6.46875,
      "learning_rate": 2.6068095515164226e-06,
      "loss": 1.54855499,
      "memory(GiB)": 117.38,
      "step": 53290,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.64335728,
      "epoch": 1.3519786910197868,
      "grad_norm": 5.71875,
      "learning_rate": 2.605888899678047e-06,
      "loss": 1.63052826,
      "memory(GiB)": 117.38,
      "step": 53295,
      "train_speed(iter/s)": 1.635947
    },
    {
      "acc": 0.65918455,
      "epoch": 1.352105530187722,
      "grad_norm": 7.40625,
      "learning_rate": 2.6049683531433645e-06,
      "loss": 1.61806126,
      "memory(GiB)": 117.38,
      "step": 53300,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.64922285,
      "epoch": 1.352232369355657,
      "grad_norm": 7.6875,
      "learning_rate": 2.6040479119528683e-06,
      "loss": 1.64680138,
      "memory(GiB)": 117.38,
      "step": 53305,
      "train_speed(iter/s)": 1.635981
    },
    {
      "acc": 0.66121531,
      "epoch": 1.352359208523592,
      "grad_norm": 5.75,
      "learning_rate": 2.6031275761470447e-06,
      "loss": 1.63919601,
      "memory(GiB)": 117.38,
      "step": 53310,
      "train_speed(iter/s)": 1.635998
    },
    {
      "acc": 0.63653183,
      "epoch": 1.352486047691527,
      "grad_norm": 6.6875,
      "learning_rate": 2.60220734576637e-06,
      "loss": 1.6381813,
      "memory(GiB)": 117.38,
      "step": 53315,
      "train_speed(iter/s)": 1.636016
    },
    {
      "acc": 0.63646011,
      "epoch": 1.3526128868594622,
      "grad_norm": 6.15625,
      "learning_rate": 2.601287220851319e-06,
      "loss": 1.73433304,
      "memory(GiB)": 117.38,
      "step": 53320,
      "train_speed(iter/s)": 1.636032
    },
    {
      "acc": 0.67886047,
      "epoch": 1.3527397260273972,
      "grad_norm": 9.9375,
      "learning_rate": 2.6003672014423677e-06,
      "loss": 1.60608101,
      "memory(GiB)": 117.38,
      "step": 53325,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.65038939,
      "epoch": 1.3528665651953324,
      "grad_norm": 5.3125,
      "learning_rate": 2.5994472875799827e-06,
      "loss": 1.62295265,
      "memory(GiB)": 117.38,
      "step": 53330,
      "train_speed(iter/s)": 1.636066
    },
    {
      "acc": 0.65853438,
      "epoch": 1.3529934043632674,
      "grad_norm": 4.59375,
      "learning_rate": 2.598527479304619e-06,
      "loss": 1.61953373,
      "memory(GiB)": 117.38,
      "step": 53335,
      "train_speed(iter/s)": 1.636083
    },
    {
      "acc": 0.67081928,
      "epoch": 1.3531202435312024,
      "grad_norm": 5.34375,
      "learning_rate": 2.597607776656741e-06,
      "loss": 1.55799046,
      "memory(GiB)": 117.38,
      "step": 53340,
      "train_speed(iter/s)": 1.636099
    },
    {
      "acc": 0.6609271,
      "epoch": 1.3532470826991374,
      "grad_norm": 6.09375,
      "learning_rate": 2.5966881796767984e-06,
      "loss": 1.5861496,
      "memory(GiB)": 117.38,
      "step": 53345,
      "train_speed(iter/s)": 1.636117
    },
    {
      "acc": 0.64542503,
      "epoch": 1.3533739218670726,
      "grad_norm": 9.0625,
      "learning_rate": 2.5957686884052423e-06,
      "loss": 1.64579029,
      "memory(GiB)": 117.38,
      "step": 53350,
      "train_speed(iter/s)": 1.636134
    },
    {
      "acc": 0.65269003,
      "epoch": 1.3535007610350076,
      "grad_norm": 9.0,
      "learning_rate": 2.5948493028825093e-06,
      "loss": 1.64242401,
      "memory(GiB)": 117.38,
      "step": 53355,
      "train_speed(iter/s)": 1.636152
    },
    {
      "acc": 0.65905704,
      "epoch": 1.3536276002029426,
      "grad_norm": 6.15625,
      "learning_rate": 2.593930023149044e-06,
      "loss": 1.62445793,
      "memory(GiB)": 117.38,
      "step": 53360,
      "train_speed(iter/s)": 1.636168
    },
    {
      "acc": 0.65242438,
      "epoch": 1.3537544393708778,
      "grad_norm": 5.5625,
      "learning_rate": 2.593010849245279e-06,
      "loss": 1.58600121,
      "memory(GiB)": 117.38,
      "step": 53365,
      "train_speed(iter/s)": 1.636186
    },
    {
      "acc": 0.64623499,
      "epoch": 1.3538812785388128,
      "grad_norm": 5.875,
      "learning_rate": 2.592091781211643e-06,
      "loss": 1.60641956,
      "memory(GiB)": 117.38,
      "step": 53370,
      "train_speed(iter/s)": 1.636203
    },
    {
      "acc": 0.68461699,
      "epoch": 1.3540081177067478,
      "grad_norm": 6.15625,
      "learning_rate": 2.591172819088561e-06,
      "loss": 1.46721954,
      "memory(GiB)": 117.38,
      "step": 53375,
      "train_speed(iter/s)": 1.63622
    },
    {
      "acc": 0.64659529,
      "epoch": 1.3541349568746828,
      "grad_norm": 4.59375,
      "learning_rate": 2.590253962916453e-06,
      "loss": 1.62378082,
      "memory(GiB)": 117.38,
      "step": 53380,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.65734396,
      "epoch": 1.354261796042618,
      "grad_norm": 5.34375,
      "learning_rate": 2.5893352127357347e-06,
      "loss": 1.54402256,
      "memory(GiB)": 117.38,
      "step": 53385,
      "train_speed(iter/s)": 1.636254
    },
    {
      "acc": 0.65798941,
      "epoch": 1.354388635210553,
      "grad_norm": 5.625,
      "learning_rate": 2.5884165685868164e-06,
      "loss": 1.58096571,
      "memory(GiB)": 117.38,
      "step": 53390,
      "train_speed(iter/s)": 1.63627
    },
    {
      "acc": 0.64605174,
      "epoch": 1.3545154743784882,
      "grad_norm": 5.90625,
      "learning_rate": 2.5874980305101045e-06,
      "loss": 1.64534111,
      "memory(GiB)": 117.38,
      "step": 53395,
      "train_speed(iter/s)": 1.636288
    },
    {
      "acc": 0.65400658,
      "epoch": 1.3546423135464232,
      "grad_norm": 6.09375,
      "learning_rate": 2.586579598546e-06,
      "loss": 1.65377083,
      "memory(GiB)": 117.38,
      "step": 53400,
      "train_speed(iter/s)": 1.636305
    },
    {
      "acc": 0.67116776,
      "epoch": 1.3547691527143582,
      "grad_norm": 7.59375,
      "learning_rate": 2.5856612727348995e-06,
      "loss": 1.51648579,
      "memory(GiB)": 117.38,
      "step": 53405,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.66574535,
      "epoch": 1.3548959918822931,
      "grad_norm": 5.46875,
      "learning_rate": 2.584743053117196e-06,
      "loss": 1.55454884,
      "memory(GiB)": 117.38,
      "step": 53410,
      "train_speed(iter/s)": 1.636339
    },
    {
      "acc": 0.66346292,
      "epoch": 1.3550228310502284,
      "grad_norm": 5.59375,
      "learning_rate": 2.583824939733277e-06,
      "loss": 1.52395287,
      "memory(GiB)": 117.38,
      "step": 53415,
      "train_speed(iter/s)": 1.636355
    },
    {
      "acc": 0.64766903,
      "epoch": 1.3551496702181633,
      "grad_norm": 5.625,
      "learning_rate": 2.5829069326235234e-06,
      "loss": 1.66314182,
      "memory(GiB)": 117.38,
      "step": 53420,
      "train_speed(iter/s)": 1.636372
    },
    {
      "acc": 0.63730884,
      "epoch": 1.3552765093860986,
      "grad_norm": 5.96875,
      "learning_rate": 2.5819890318283137e-06,
      "loss": 1.60120277,
      "memory(GiB)": 117.38,
      "step": 53425,
      "train_speed(iter/s)": 1.636389
    },
    {
      "acc": 0.65782394,
      "epoch": 1.3554033485540335,
      "grad_norm": 6.1875,
      "learning_rate": 2.5810712373880253e-06,
      "loss": 1.65483608,
      "memory(GiB)": 117.38,
      "step": 53430,
      "train_speed(iter/s)": 1.636407
    },
    {
      "acc": 0.66530867,
      "epoch": 1.3555301877219685,
      "grad_norm": 4.875,
      "learning_rate": 2.5801535493430215e-06,
      "loss": 1.59025354,
      "memory(GiB)": 117.38,
      "step": 53435,
      "train_speed(iter/s)": 1.636424
    },
    {
      "acc": 0.67099171,
      "epoch": 1.3556570268899035,
      "grad_norm": 5.6875,
      "learning_rate": 2.5792359677336685e-06,
      "loss": 1.5561408,
      "memory(GiB)": 117.38,
      "step": 53440,
      "train_speed(iter/s)": 1.636441
    },
    {
      "acc": 0.66039257,
      "epoch": 1.3557838660578387,
      "grad_norm": 5.59375,
      "learning_rate": 2.5783184926003237e-06,
      "loss": 1.6002964,
      "memory(GiB)": 117.38,
      "step": 53445,
      "train_speed(iter/s)": 1.636457
    },
    {
      "acc": 0.64508629,
      "epoch": 1.3559107052257737,
      "grad_norm": 6.3125,
      "learning_rate": 2.5774011239833473e-06,
      "loss": 1.64640236,
      "memory(GiB)": 117.38,
      "step": 53450,
      "train_speed(iter/s)": 1.636475
    },
    {
      "acc": 0.66321273,
      "epoch": 1.3560375443937087,
      "grad_norm": 6.71875,
      "learning_rate": 2.5764838619230843e-06,
      "loss": 1.62081356,
      "memory(GiB)": 117.38,
      "step": 53455,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.67482982,
      "epoch": 1.356164383561644,
      "grad_norm": 7.53125,
      "learning_rate": 2.575566706459879e-06,
      "loss": 1.52221003,
      "memory(GiB)": 117.38,
      "step": 53460,
      "train_speed(iter/s)": 1.636507
    },
    {
      "acc": 0.65256124,
      "epoch": 1.356291222729579,
      "grad_norm": 5.78125,
      "learning_rate": 2.574649657634076e-06,
      "loss": 1.63283176,
      "memory(GiB)": 117.38,
      "step": 53465,
      "train_speed(iter/s)": 1.636524
    },
    {
      "acc": 0.66210742,
      "epoch": 1.356418061897514,
      "grad_norm": 6.53125,
      "learning_rate": 2.5737327154860116e-06,
      "loss": 1.61089935,
      "memory(GiB)": 117.38,
      "step": 53470,
      "train_speed(iter/s)": 1.636542
    },
    {
      "acc": 0.6434392,
      "epoch": 1.356544901065449,
      "grad_norm": 5.8125,
      "learning_rate": 2.572815880056011e-06,
      "loss": 1.67750626,
      "memory(GiB)": 117.38,
      "step": 53475,
      "train_speed(iter/s)": 1.636559
    },
    {
      "acc": 0.66380305,
      "epoch": 1.356671740233384,
      "grad_norm": 5.9375,
      "learning_rate": 2.571899151384406e-06,
      "loss": 1.53025331,
      "memory(GiB)": 117.38,
      "step": 53480,
      "train_speed(iter/s)": 1.636574
    },
    {
      "acc": 0.66234779,
      "epoch": 1.356798579401319,
      "grad_norm": 5.21875,
      "learning_rate": 2.5709825295115178e-06,
      "loss": 1.62158527,
      "memory(GiB)": 117.38,
      "step": 53485,
      "train_speed(iter/s)": 1.636591
    },
    {
      "acc": 0.64412508,
      "epoch": 1.3569254185692543,
      "grad_norm": 5.4375,
      "learning_rate": 2.5700660144776647e-06,
      "loss": 1.6475872,
      "memory(GiB)": 117.38,
      "step": 53490,
      "train_speed(iter/s)": 1.636608
    },
    {
      "acc": 0.65411391,
      "epoch": 1.3570522577371893,
      "grad_norm": 6.46875,
      "learning_rate": 2.5691496063231527e-06,
      "loss": 1.61621971,
      "memory(GiB)": 117.38,
      "step": 53495,
      "train_speed(iter/s)": 1.636625
    },
    {
      "acc": 0.65778389,
      "epoch": 1.3571790969051243,
      "grad_norm": 9.1875,
      "learning_rate": 2.568233305088296e-06,
      "loss": 1.53290215,
      "memory(GiB)": 117.38,
      "step": 53500,
      "train_speed(iter/s)": 1.636641
    },
    {
      "acc": 0.66203542,
      "epoch": 1.3573059360730593,
      "grad_norm": 5.375,
      "learning_rate": 2.5673171108133956e-06,
      "loss": 1.56835804,
      "memory(GiB)": 117.38,
      "step": 53505,
      "train_speed(iter/s)": 1.636658
    },
    {
      "acc": 0.64234896,
      "epoch": 1.3574327752409945,
      "grad_norm": 5.0,
      "learning_rate": 2.5664010235387503e-06,
      "loss": 1.62272148,
      "memory(GiB)": 117.38,
      "step": 53510,
      "train_speed(iter/s)": 1.636676
    },
    {
      "acc": 0.66369486,
      "epoch": 1.3575596144089295,
      "grad_norm": 7.46875,
      "learning_rate": 2.565485043304653e-06,
      "loss": 1.58520908,
      "memory(GiB)": 117.38,
      "step": 53515,
      "train_speed(iter/s)": 1.636692
    },
    {
      "acc": 0.64480968,
      "epoch": 1.3576864535768645,
      "grad_norm": 6.15625,
      "learning_rate": 2.564569170151392e-06,
      "loss": 1.62558746,
      "memory(GiB)": 117.38,
      "step": 53520,
      "train_speed(iter/s)": 1.63671
    },
    {
      "acc": 0.65705748,
      "epoch": 1.3578132927447997,
      "grad_norm": 5.34375,
      "learning_rate": 2.5636534041192534e-06,
      "loss": 1.54227839,
      "memory(GiB)": 117.38,
      "step": 53525,
      "train_speed(iter/s)": 1.636726
    },
    {
      "acc": 0.65346851,
      "epoch": 1.3579401319127347,
      "grad_norm": 5.28125,
      "learning_rate": 2.5627377452485153e-06,
      "loss": 1.59480934,
      "memory(GiB)": 117.38,
      "step": 53530,
      "train_speed(iter/s)": 1.636744
    },
    {
      "acc": 0.66570468,
      "epoch": 1.3580669710806696,
      "grad_norm": 4.90625,
      "learning_rate": 2.561822193579453e-06,
      "loss": 1.52571135,
      "memory(GiB)": 117.38,
      "step": 53535,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.6710207,
      "epoch": 1.3581938102486046,
      "grad_norm": 5.6875,
      "learning_rate": 2.560906749152335e-06,
      "loss": 1.60008297,
      "memory(GiB)": 117.38,
      "step": 53540,
      "train_speed(iter/s)": 1.636778
    },
    {
      "acc": 0.65436282,
      "epoch": 1.3583206494165398,
      "grad_norm": 5.71875,
      "learning_rate": 2.55999141200743e-06,
      "loss": 1.56700296,
      "memory(GiB)": 117.38,
      "step": 53545,
      "train_speed(iter/s)": 1.636795
    },
    {
      "acc": 0.66621675,
      "epoch": 1.3584474885844748,
      "grad_norm": 4.90625,
      "learning_rate": 2.5590761821849954e-06,
      "loss": 1.50343142,
      "memory(GiB)": 117.38,
      "step": 53550,
      "train_speed(iter/s)": 1.636812
    },
    {
      "acc": 0.66937995,
      "epoch": 1.35857432775241,
      "grad_norm": 6.03125,
      "learning_rate": 2.5581610597252883e-06,
      "loss": 1.57336159,
      "memory(GiB)": 117.38,
      "step": 53555,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.65682297,
      "epoch": 1.358701166920345,
      "grad_norm": 5.3125,
      "learning_rate": 2.5572460446685593e-06,
      "loss": 1.53592196,
      "memory(GiB)": 117.38,
      "step": 53560,
      "train_speed(iter/s)": 1.636846
    },
    {
      "acc": 0.65299702,
      "epoch": 1.35882800608828,
      "grad_norm": 5.9375,
      "learning_rate": 2.5563311370550535e-06,
      "loss": 1.61211033,
      "memory(GiB)": 117.38,
      "step": 53565,
      "train_speed(iter/s)": 1.636862
    },
    {
      "acc": 0.64969316,
      "epoch": 1.358954845256215,
      "grad_norm": 6.25,
      "learning_rate": 2.5554163369250194e-06,
      "loss": 1.67772903,
      "memory(GiB)": 117.38,
      "step": 53570,
      "train_speed(iter/s)": 1.636878
    },
    {
      "acc": 0.66940522,
      "epoch": 1.3590816844241502,
      "grad_norm": 6.03125,
      "learning_rate": 2.5545016443186867e-06,
      "loss": 1.53773775,
      "memory(GiB)": 117.38,
      "step": 53575,
      "train_speed(iter/s)": 1.636895
    },
    {
      "acc": 0.65389729,
      "epoch": 1.3592085235920852,
      "grad_norm": 6.0625,
      "learning_rate": 2.55358705927629e-06,
      "loss": 1.58248663,
      "memory(GiB)": 117.38,
      "step": 53580,
      "train_speed(iter/s)": 1.63691
    },
    {
      "acc": 0.64850359,
      "epoch": 1.3593353627600204,
      "grad_norm": 5.09375,
      "learning_rate": 2.552672581838055e-06,
      "loss": 1.62079391,
      "memory(GiB)": 117.38,
      "step": 53585,
      "train_speed(iter/s)": 1.636927
    },
    {
      "acc": 0.68043618,
      "epoch": 1.3594622019279554,
      "grad_norm": 6.25,
      "learning_rate": 2.5517582120442095e-06,
      "loss": 1.54785957,
      "memory(GiB)": 117.38,
      "step": 53590,
      "train_speed(iter/s)": 1.636944
    },
    {
      "acc": 0.64452925,
      "epoch": 1.3595890410958904,
      "grad_norm": 5.6875,
      "learning_rate": 2.5508439499349675e-06,
      "loss": 1.61808472,
      "memory(GiB)": 117.38,
      "step": 53595,
      "train_speed(iter/s)": 1.636961
    },
    {
      "acc": 0.6411911,
      "epoch": 1.3597158802638254,
      "grad_norm": 5.375,
      "learning_rate": 2.549929795550541e-06,
      "loss": 1.62348404,
      "memory(GiB)": 117.38,
      "step": 53600,
      "train_speed(iter/s)": 1.636978
    },
    {
      "acc": 0.66476412,
      "epoch": 1.3598427194317606,
      "grad_norm": 7.28125,
      "learning_rate": 2.549015748931143e-06,
      "loss": 1.56165237,
      "memory(GiB)": 117.38,
      "step": 53605,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.66282101,
      "epoch": 1.3599695585996956,
      "grad_norm": 5.875,
      "learning_rate": 2.5481018101169763e-06,
      "loss": 1.59731636,
      "memory(GiB)": 117.38,
      "step": 53610,
      "train_speed(iter/s)": 1.637011
    },
    {
      "acc": 0.65457664,
      "epoch": 1.3600963977676306,
      "grad_norm": 6.40625,
      "learning_rate": 2.547187979148238e-06,
      "loss": 1.56694136,
      "memory(GiB)": 117.38,
      "step": 53615,
      "train_speed(iter/s)": 1.637029
    },
    {
      "acc": 0.66665554,
      "epoch": 1.3602232369355658,
      "grad_norm": 4.875,
      "learning_rate": 2.546274256065121e-06,
      "loss": 1.59870129,
      "memory(GiB)": 117.38,
      "step": 53620,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.65498114,
      "epoch": 1.3603500761035008,
      "grad_norm": 5.625,
      "learning_rate": 2.545360640907819e-06,
      "loss": 1.59269962,
      "memory(GiB)": 117.38,
      "step": 53625,
      "train_speed(iter/s)": 1.637064
    },
    {
      "acc": 0.64923649,
      "epoch": 1.3604769152714358,
      "grad_norm": 5.8125,
      "learning_rate": 2.544447133716518e-06,
      "loss": 1.56173363,
      "memory(GiB)": 117.38,
      "step": 53630,
      "train_speed(iter/s)": 1.637079
    },
    {
      "acc": 0.637815,
      "epoch": 1.3606037544393708,
      "grad_norm": 5.21875,
      "learning_rate": 2.5435337345313904e-06,
      "loss": 1.64669476,
      "memory(GiB)": 117.38,
      "step": 53635,
      "train_speed(iter/s)": 1.637095
    },
    {
      "acc": 0.64696646,
      "epoch": 1.360730593607306,
      "grad_norm": 6.28125,
      "learning_rate": 2.5426204433926194e-06,
      "loss": 1.5872736,
      "memory(GiB)": 117.38,
      "step": 53640,
      "train_speed(iter/s)": 1.637111
    },
    {
      "acc": 0.64977741,
      "epoch": 1.360857432775241,
      "grad_norm": 5.28125,
      "learning_rate": 2.541707260340372e-06,
      "loss": 1.65675945,
      "memory(GiB)": 117.38,
      "step": 53645,
      "train_speed(iter/s)": 1.637127
    },
    {
      "acc": 0.66039391,
      "epoch": 1.3609842719431762,
      "grad_norm": 5.125,
      "learning_rate": 2.5407941854148156e-06,
      "loss": 1.61841202,
      "memory(GiB)": 117.38,
      "step": 53650,
      "train_speed(iter/s)": 1.637144
    },
    {
      "acc": 0.64172859,
      "epoch": 1.3611111111111112,
      "grad_norm": 5.65625,
      "learning_rate": 2.5398812186561095e-06,
      "loss": 1.67692986,
      "memory(GiB)": 117.38,
      "step": 53655,
      "train_speed(iter/s)": 1.63716
    },
    {
      "acc": 0.64275894,
      "epoch": 1.3612379502790461,
      "grad_norm": 4.875,
      "learning_rate": 2.5389683601044114e-06,
      "loss": 1.61739941,
      "memory(GiB)": 117.38,
      "step": 53660,
      "train_speed(iter/s)": 1.637177
    },
    {
      "acc": 0.65812073,
      "epoch": 1.3613647894469811,
      "grad_norm": 6.46875,
      "learning_rate": 2.538055609799873e-06,
      "loss": 1.59457846,
      "memory(GiB)": 117.38,
      "step": 53665,
      "train_speed(iter/s)": 1.637194
    },
    {
      "acc": 0.65167179,
      "epoch": 1.3614916286149163,
      "grad_norm": 5.625,
      "learning_rate": 2.5371429677826397e-06,
      "loss": 1.59267788,
      "memory(GiB)": 117.38,
      "step": 53670,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.68152027,
      "epoch": 1.3616184677828513,
      "grad_norm": 6.15625,
      "learning_rate": 2.5362304340928556e-06,
      "loss": 1.51827164,
      "memory(GiB)": 117.38,
      "step": 53675,
      "train_speed(iter/s)": 1.637228
    },
    {
      "acc": 0.64996185,
      "epoch": 1.3617453069507863,
      "grad_norm": 6.125,
      "learning_rate": 2.535318008770656e-06,
      "loss": 1.6348875,
      "memory(GiB)": 117.38,
      "step": 53680,
      "train_speed(iter/s)": 1.637244
    },
    {
      "acc": 0.66188283,
      "epoch": 1.3618721461187215,
      "grad_norm": 5.53125,
      "learning_rate": 2.534405691856175e-06,
      "loss": 1.57440948,
      "memory(GiB)": 117.38,
      "step": 53685,
      "train_speed(iter/s)": 1.63726
    },
    {
      "acc": 0.6661376,
      "epoch": 1.3619989852866565,
      "grad_norm": 5.1875,
      "learning_rate": 2.5334934833895396e-06,
      "loss": 1.54691305,
      "memory(GiB)": 117.38,
      "step": 53690,
      "train_speed(iter/s)": 1.637276
    },
    {
      "acc": 0.65001383,
      "epoch": 1.3621258244545915,
      "grad_norm": 5.0,
      "learning_rate": 2.5325813834108724e-06,
      "loss": 1.61841278,
      "memory(GiB)": 117.38,
      "step": 53695,
      "train_speed(iter/s)": 1.637292
    },
    {
      "acc": 0.6714757,
      "epoch": 1.3622526636225265,
      "grad_norm": 5.1875,
      "learning_rate": 2.531669391960293e-06,
      "loss": 1.57934675,
      "memory(GiB)": 117.38,
      "step": 53700,
      "train_speed(iter/s)": 1.637309
    },
    {
      "acc": 0.65481548,
      "epoch": 1.3623795027904617,
      "grad_norm": 5.03125,
      "learning_rate": 2.5307575090779125e-06,
      "loss": 1.65041409,
      "memory(GiB)": 117.38,
      "step": 53705,
      "train_speed(iter/s)": 1.637325
    },
    {
      "acc": 0.65153427,
      "epoch": 1.3625063419583967,
      "grad_norm": 6.15625,
      "learning_rate": 2.529845734803844e-06,
      "loss": 1.59865665,
      "memory(GiB)": 117.38,
      "step": 53710,
      "train_speed(iter/s)": 1.637341
    },
    {
      "acc": 0.66214414,
      "epoch": 1.362633181126332,
      "grad_norm": 6.625,
      "learning_rate": 2.5289340691781872e-06,
      "loss": 1.65031738,
      "memory(GiB)": 117.38,
      "step": 53715,
      "train_speed(iter/s)": 1.637359
    },
    {
      "acc": 0.65672021,
      "epoch": 1.362760020294267,
      "grad_norm": 6.28125,
      "learning_rate": 2.528022512241042e-06,
      "loss": 1.57944927,
      "memory(GiB)": 117.38,
      "step": 53720,
      "train_speed(iter/s)": 1.637376
    },
    {
      "acc": 0.66379671,
      "epoch": 1.362886859462202,
      "grad_norm": 5.53125,
      "learning_rate": 2.5271110640325013e-06,
      "loss": 1.57880936,
      "memory(GiB)": 117.38,
      "step": 53725,
      "train_speed(iter/s)": 1.637392
    },
    {
      "acc": 0.65716343,
      "epoch": 1.3630136986301369,
      "grad_norm": 5.0625,
      "learning_rate": 2.5261997245926612e-06,
      "loss": 1.59805593,
      "memory(GiB)": 117.38,
      "step": 53730,
      "train_speed(iter/s)": 1.637408
    },
    {
      "acc": 0.64645138,
      "epoch": 1.363140537798072,
      "grad_norm": 5.875,
      "learning_rate": 2.5252884939615995e-06,
      "loss": 1.62116013,
      "memory(GiB)": 117.38,
      "step": 53735,
      "train_speed(iter/s)": 1.637424
    },
    {
      "acc": 0.65044165,
      "epoch": 1.363267376966007,
      "grad_norm": 4.6875,
      "learning_rate": 2.5243773721793973e-06,
      "loss": 1.59862518,
      "memory(GiB)": 117.38,
      "step": 53740,
      "train_speed(iter/s)": 1.637441
    },
    {
      "acc": 0.66431432,
      "epoch": 1.3633942161339423,
      "grad_norm": 6.5,
      "learning_rate": 2.5234663592861325e-06,
      "loss": 1.55725327,
      "memory(GiB)": 117.38,
      "step": 53745,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.65912647,
      "epoch": 1.3635210553018773,
      "grad_norm": 5.75,
      "learning_rate": 2.522555455321876e-06,
      "loss": 1.57857447,
      "memory(GiB)": 117.38,
      "step": 53750,
      "train_speed(iter/s)": 1.637473
    },
    {
      "acc": 0.65796304,
      "epoch": 1.3636478944698123,
      "grad_norm": 6.4375,
      "learning_rate": 2.52164466032669e-06,
      "loss": 1.56846848,
      "memory(GiB)": 117.38,
      "step": 53755,
      "train_speed(iter/s)": 1.63749
    },
    {
      "acc": 0.67043819,
      "epoch": 1.3637747336377473,
      "grad_norm": 4.96875,
      "learning_rate": 2.5207339743406344e-06,
      "loss": 1.61434784,
      "memory(GiB)": 117.38,
      "step": 53760,
      "train_speed(iter/s)": 1.637507
    },
    {
      "acc": 0.63644285,
      "epoch": 1.3639015728056825,
      "grad_norm": 4.46875,
      "learning_rate": 2.5198233974037705e-06,
      "loss": 1.60361748,
      "memory(GiB)": 117.38,
      "step": 53765,
      "train_speed(iter/s)": 1.637523
    },
    {
      "acc": 0.64687481,
      "epoch": 1.3640284119736175,
      "grad_norm": 6.5,
      "learning_rate": 2.5189129295561486e-06,
      "loss": 1.64974556,
      "memory(GiB)": 117.38,
      "step": 53770,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.65588765,
      "epoch": 1.3641552511415524,
      "grad_norm": 5.03125,
      "learning_rate": 2.518002570837809e-06,
      "loss": 1.60140057,
      "memory(GiB)": 117.38,
      "step": 53775,
      "train_speed(iter/s)": 1.637557
    },
    {
      "acc": 0.64455504,
      "epoch": 1.3642820903094877,
      "grad_norm": 6.59375,
      "learning_rate": 2.5170923212887997e-06,
      "loss": 1.70492039,
      "memory(GiB)": 117.38,
      "step": 53780,
      "train_speed(iter/s)": 1.637574
    },
    {
      "acc": 0.68089471,
      "epoch": 1.3644089294774226,
      "grad_norm": 5.75,
      "learning_rate": 2.5161821809491554e-06,
      "loss": 1.55858107,
      "memory(GiB)": 117.38,
      "step": 53785,
      "train_speed(iter/s)": 1.637591
    },
    {
      "acc": 0.66361413,
      "epoch": 1.3645357686453576,
      "grad_norm": 5.46875,
      "learning_rate": 2.5152721498589104e-06,
      "loss": 1.57757607,
      "memory(GiB)": 117.38,
      "step": 53790,
      "train_speed(iter/s)": 1.637607
    },
    {
      "acc": 0.64991264,
      "epoch": 1.3646626078132926,
      "grad_norm": 7.875,
      "learning_rate": 2.514362228058086e-06,
      "loss": 1.62747116,
      "memory(GiB)": 117.38,
      "step": 53795,
      "train_speed(iter/s)": 1.637623
    },
    {
      "acc": 0.65238428,
      "epoch": 1.3647894469812278,
      "grad_norm": 6.3125,
      "learning_rate": 2.51345241558671e-06,
      "loss": 1.64885216,
      "memory(GiB)": 117.38,
      "step": 53800,
      "train_speed(iter/s)": 1.63764
    },
    {
      "acc": 0.65161285,
      "epoch": 1.3649162861491628,
      "grad_norm": 5.875,
      "learning_rate": 2.5125427124847985e-06,
      "loss": 1.67882156,
      "memory(GiB)": 117.38,
      "step": 53805,
      "train_speed(iter/s)": 1.637658
    },
    {
      "acc": 0.66031466,
      "epoch": 1.365043125317098,
      "grad_norm": 6.03125,
      "learning_rate": 2.5116331187923645e-06,
      "loss": 1.55103235,
      "memory(GiB)": 117.38,
      "step": 53810,
      "train_speed(iter/s)": 1.637674
    },
    {
      "acc": 0.65867724,
      "epoch": 1.365169964485033,
      "grad_norm": 6.28125,
      "learning_rate": 2.510723634549415e-06,
      "loss": 1.60701294,
      "memory(GiB)": 117.38,
      "step": 53815,
      "train_speed(iter/s)": 1.63769
    },
    {
      "acc": 0.66362724,
      "epoch": 1.365296803652968,
      "grad_norm": 6.84375,
      "learning_rate": 2.509814259795954e-06,
      "loss": 1.59920597,
      "memory(GiB)": 117.38,
      "step": 53820,
      "train_speed(iter/s)": 1.637706
    },
    {
      "acc": 0.66874352,
      "epoch": 1.365423642820903,
      "grad_norm": 6.03125,
      "learning_rate": 2.50890499457198e-06,
      "loss": 1.58843727,
      "memory(GiB)": 117.38,
      "step": 53825,
      "train_speed(iter/s)": 1.637723
    },
    {
      "acc": 0.67149086,
      "epoch": 1.3655504819888382,
      "grad_norm": 5.4375,
      "learning_rate": 2.5079958389174865e-06,
      "loss": 1.51825161,
      "memory(GiB)": 117.38,
      "step": 53830,
      "train_speed(iter/s)": 1.637738
    },
    {
      "acc": 0.64371719,
      "epoch": 1.3656773211567732,
      "grad_norm": 6.5,
      "learning_rate": 2.5070867928724618e-06,
      "loss": 1.6810482,
      "memory(GiB)": 117.38,
      "step": 53835,
      "train_speed(iter/s)": 1.637754
    },
    {
      "acc": 0.65660286,
      "epoch": 1.3658041603247082,
      "grad_norm": 5.625,
      "learning_rate": 2.50617785647689e-06,
      "loss": 1.55161934,
      "memory(GiB)": 117.38,
      "step": 53840,
      "train_speed(iter/s)": 1.637771
    },
    {
      "acc": 0.65005198,
      "epoch": 1.3659309994926434,
      "grad_norm": 7.09375,
      "learning_rate": 2.5052690297707506e-06,
      "loss": 1.58277245,
      "memory(GiB)": 117.38,
      "step": 53845,
      "train_speed(iter/s)": 1.637787
    },
    {
      "acc": 0.66490679,
      "epoch": 1.3660578386605784,
      "grad_norm": 6.4375,
      "learning_rate": 2.5043603127940164e-06,
      "loss": 1.48484783,
      "memory(GiB)": 117.38,
      "step": 53850,
      "train_speed(iter/s)": 1.637804
    },
    {
      "acc": 0.65216084,
      "epoch": 1.3661846778285134,
      "grad_norm": 6.5625,
      "learning_rate": 2.503451705586659e-06,
      "loss": 1.61127319,
      "memory(GiB)": 117.38,
      "step": 53855,
      "train_speed(iter/s)": 1.63782
    },
    {
      "acc": 0.66695738,
      "epoch": 1.3663115169964484,
      "grad_norm": 5.3125,
      "learning_rate": 2.5025432081886412e-06,
      "loss": 1.54849424,
      "memory(GiB)": 117.38,
      "step": 53860,
      "train_speed(iter/s)": 1.637837
    },
    {
      "acc": 0.65645399,
      "epoch": 1.3664383561643836,
      "grad_norm": 6.40625,
      "learning_rate": 2.5016348206399215e-06,
      "loss": 1.57486687,
      "memory(GiB)": 117.38,
      "step": 53865,
      "train_speed(iter/s)": 1.637854
    },
    {
      "acc": 0.66213379,
      "epoch": 1.3665651953323186,
      "grad_norm": 6.6875,
      "learning_rate": 2.500726542980461e-06,
      "loss": 1.6468626,
      "memory(GiB)": 117.38,
      "step": 53870,
      "train_speed(iter/s)": 1.63787
    },
    {
      "acc": 0.65925865,
      "epoch": 1.3666920345002538,
      "grad_norm": 5.5,
      "learning_rate": 2.499818375250204e-06,
      "loss": 1.59423265,
      "memory(GiB)": 117.38,
      "step": 53875,
      "train_speed(iter/s)": 1.637885
    },
    {
      "acc": 0.65358038,
      "epoch": 1.3668188736681888,
      "grad_norm": 7.0,
      "learning_rate": 2.4989103174890946e-06,
      "loss": 1.63917847,
      "memory(GiB)": 117.38,
      "step": 53880,
      "train_speed(iter/s)": 1.637901
    },
    {
      "acc": 0.65485501,
      "epoch": 1.3669457128361238,
      "grad_norm": 4.96875,
      "learning_rate": 2.498002369737078e-06,
      "loss": 1.64345894,
      "memory(GiB)": 117.38,
      "step": 53885,
      "train_speed(iter/s)": 1.637916
    },
    {
      "acc": 0.66022215,
      "epoch": 1.3670725520040587,
      "grad_norm": 7.0625,
      "learning_rate": 2.49709453203409e-06,
      "loss": 1.58046474,
      "memory(GiB)": 117.38,
      "step": 53890,
      "train_speed(iter/s)": 1.637933
    },
    {
      "acc": 0.65218558,
      "epoch": 1.367199391171994,
      "grad_norm": 5.0,
      "learning_rate": 2.496186804420057e-06,
      "loss": 1.62577248,
      "memory(GiB)": 117.38,
      "step": 53895,
      "train_speed(iter/s)": 1.637948
    },
    {
      "acc": 0.65749688,
      "epoch": 1.367326230339929,
      "grad_norm": 4.65625,
      "learning_rate": 2.4952791869349056e-06,
      "loss": 1.61458359,
      "memory(GiB)": 117.38,
      "step": 53900,
      "train_speed(iter/s)": 1.637965
    },
    {
      "acc": 0.66395006,
      "epoch": 1.3674530695078642,
      "grad_norm": 8.4375,
      "learning_rate": 2.4943716796185603e-06,
      "loss": 1.59068127,
      "memory(GiB)": 117.38,
      "step": 53905,
      "train_speed(iter/s)": 1.637981
    },
    {
      "acc": 0.65934453,
      "epoch": 1.3675799086757991,
      "grad_norm": 6.0625,
      "learning_rate": 2.493464282510937e-06,
      "loss": 1.65735378,
      "memory(GiB)": 117.38,
      "step": 53910,
      "train_speed(iter/s)": 1.637999
    },
    {
      "acc": 0.67474036,
      "epoch": 1.3677067478437341,
      "grad_norm": 5.84375,
      "learning_rate": 2.4925569956519414e-06,
      "loss": 1.548738,
      "memory(GiB)": 117.38,
      "step": 53915,
      "train_speed(iter/s)": 1.638014
    },
    {
      "acc": 0.6487464,
      "epoch": 1.3678335870116691,
      "grad_norm": 5.4375,
      "learning_rate": 2.491649819081486e-06,
      "loss": 1.66526947,
      "memory(GiB)": 117.38,
      "step": 53920,
      "train_speed(iter/s)": 1.638031
    },
    {
      "acc": 0.66771212,
      "epoch": 1.3679604261796043,
      "grad_norm": 5.4375,
      "learning_rate": 2.490742752839471e-06,
      "loss": 1.63681526,
      "memory(GiB)": 117.38,
      "step": 53925,
      "train_speed(iter/s)": 1.638048
    },
    {
      "acc": 0.65810323,
      "epoch": 1.3680872653475393,
      "grad_norm": 6.90625,
      "learning_rate": 2.4898357969657943e-06,
      "loss": 1.5605608,
      "memory(GiB)": 117.38,
      "step": 53930,
      "train_speed(iter/s)": 1.638066
    },
    {
      "acc": 0.66883926,
      "epoch": 1.3682141045154743,
      "grad_norm": 5.53125,
      "learning_rate": 2.4889289515003425e-06,
      "loss": 1.58948851,
      "memory(GiB)": 117.38,
      "step": 53935,
      "train_speed(iter/s)": 1.638082
    },
    {
      "acc": 0.67358646,
      "epoch": 1.3683409436834095,
      "grad_norm": 7.1875,
      "learning_rate": 2.4880222164830085e-06,
      "loss": 1.56756659,
      "memory(GiB)": 117.38,
      "step": 53940,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.66199064,
      "epoch": 1.3684677828513445,
      "grad_norm": 4.9375,
      "learning_rate": 2.4871155919536725e-06,
      "loss": 1.59989166,
      "memory(GiB)": 117.38,
      "step": 53945,
      "train_speed(iter/s)": 1.638116
    },
    {
      "acc": 0.6757699,
      "epoch": 1.3685946220192795,
      "grad_norm": 7.96875,
      "learning_rate": 2.486209077952212e-06,
      "loss": 1.54847317,
      "memory(GiB)": 117.38,
      "step": 53950,
      "train_speed(iter/s)": 1.638131
    },
    {
      "acc": 0.66980858,
      "epoch": 1.3687214611872145,
      "grad_norm": 8.5,
      "learning_rate": 2.4853026745185e-06,
      "loss": 1.59327583,
      "memory(GiB)": 117.38,
      "step": 53955,
      "train_speed(iter/s)": 1.638147
    },
    {
      "acc": 0.66415458,
      "epoch": 1.3688483003551497,
      "grad_norm": 5.125,
      "learning_rate": 2.4843963816924035e-06,
      "loss": 1.5571785,
      "memory(GiB)": 117.38,
      "step": 53960,
      "train_speed(iter/s)": 1.638162
    },
    {
      "acc": 0.65693302,
      "epoch": 1.3689751395230847,
      "grad_norm": 5.15625,
      "learning_rate": 2.483490199513785e-06,
      "loss": 1.55262871,
      "memory(GiB)": 117.38,
      "step": 53965,
      "train_speed(iter/s)": 1.638178
    },
    {
      "acc": 0.67094212,
      "epoch": 1.36910197869102,
      "grad_norm": 7.09375,
      "learning_rate": 2.4825841280225033e-06,
      "loss": 1.60100746,
      "memory(GiB)": 117.38,
      "step": 53970,
      "train_speed(iter/s)": 1.638195
    },
    {
      "acc": 0.64782677,
      "epoch": 1.369228817858955,
      "grad_norm": 5.71875,
      "learning_rate": 2.4816781672584107e-06,
      "loss": 1.62448101,
      "memory(GiB)": 117.38,
      "step": 53975,
      "train_speed(iter/s)": 1.638212
    },
    {
      "acc": 0.64191036,
      "epoch": 1.3693556570268899,
      "grad_norm": 5.4375,
      "learning_rate": 2.480772317261356e-06,
      "loss": 1.62860718,
      "memory(GiB)": 117.38,
      "step": 53980,
      "train_speed(iter/s)": 1.638227
    },
    {
      "acc": 0.65724363,
      "epoch": 1.3694824961948249,
      "grad_norm": 5.75,
      "learning_rate": 2.479866578071183e-06,
      "loss": 1.60973969,
      "memory(GiB)": 117.38,
      "step": 53985,
      "train_speed(iter/s)": 1.638244
    },
    {
      "acc": 0.65571737,
      "epoch": 1.36960933536276,
      "grad_norm": 7.96875,
      "learning_rate": 2.4789609497277284e-06,
      "loss": 1.56727686,
      "memory(GiB)": 117.38,
      "step": 53990,
      "train_speed(iter/s)": 1.638261
    },
    {
      "acc": 0.65684423,
      "epoch": 1.369736174530695,
      "grad_norm": 6.28125,
      "learning_rate": 2.478055432270828e-06,
      "loss": 1.60636482,
      "memory(GiB)": 117.38,
      "step": 53995,
      "train_speed(iter/s)": 1.638277
    },
    {
      "acc": 0.69125848,
      "epoch": 1.36986301369863,
      "grad_norm": 4.625,
      "learning_rate": 2.4771500257403086e-06,
      "loss": 1.42107573,
      "memory(GiB)": 117.38,
      "step": 54000,
      "train_speed(iter/s)": 1.638292
    },
    {
      "epoch": 1.36986301369863,
      "eval_acc": 0.646273522157972,
      "eval_loss": 1.5731202363967896,
      "eval_runtime": 58.4127,
      "eval_samples_per_second": 109.052,
      "eval_steps_per_second": 27.271,
      "step": 54000
    },
    {
      "acc": 0.6596487,
      "epoch": 1.3699898528665653,
      "grad_norm": 4.9375,
      "learning_rate": 2.476244730175993e-06,
      "loss": 1.59097767,
      "memory(GiB)": 117.38,
      "step": 54005,
      "train_speed(iter/s)": 1.635195
    },
    {
      "acc": 0.65782223,
      "epoch": 1.3701166920345003,
      "grad_norm": 5.90625,
      "learning_rate": 2.4753395456177056e-06,
      "loss": 1.6168663,
      "memory(GiB)": 117.38,
      "step": 54010,
      "train_speed(iter/s)": 1.635211
    },
    {
      "acc": 0.64816422,
      "epoch": 1.3702435312024352,
      "grad_norm": 5.25,
      "learning_rate": 2.474434472105255e-06,
      "loss": 1.63149796,
      "memory(GiB)": 117.38,
      "step": 54015,
      "train_speed(iter/s)": 1.635228
    },
    {
      "acc": 0.6532033,
      "epoch": 1.3703703703703702,
      "grad_norm": 5.96875,
      "learning_rate": 2.473529509678452e-06,
      "loss": 1.57334414,
      "memory(GiB)": 117.38,
      "step": 54020,
      "train_speed(iter/s)": 1.635245
    },
    {
      "acc": 0.65062523,
      "epoch": 1.3704972095383054,
      "grad_norm": 5.9375,
      "learning_rate": 2.4726246583770996e-06,
      "loss": 1.65277958,
      "memory(GiB)": 117.38,
      "step": 54025,
      "train_speed(iter/s)": 1.635262
    },
    {
      "acc": 0.65357847,
      "epoch": 1.3706240487062404,
      "grad_norm": 5.0,
      "learning_rate": 2.4717199182410025e-06,
      "loss": 1.54226589,
      "memory(GiB)": 117.38,
      "step": 54030,
      "train_speed(iter/s)": 1.635278
    },
    {
      "acc": 0.67414503,
      "epoch": 1.3707508878741756,
      "grad_norm": 7.46875,
      "learning_rate": 2.4708152893099493e-06,
      "loss": 1.57174397,
      "memory(GiB)": 117.38,
      "step": 54035,
      "train_speed(iter/s)": 1.635294
    },
    {
      "acc": 0.66791396,
      "epoch": 1.3708777270421106,
      "grad_norm": 7.3125,
      "learning_rate": 2.4699107716237293e-06,
      "loss": 1.55494394,
      "memory(GiB)": 117.38,
      "step": 54040,
      "train_speed(iter/s)": 1.63531
    },
    {
      "acc": 0.66002736,
      "epoch": 1.3710045662100456,
      "grad_norm": 4.84375,
      "learning_rate": 2.469006365222132e-06,
      "loss": 1.55516348,
      "memory(GiB)": 117.38,
      "step": 54045,
      "train_speed(iter/s)": 1.635325
    },
    {
      "acc": 0.64921136,
      "epoch": 1.3711314053779806,
      "grad_norm": 6.3125,
      "learning_rate": 2.4681020701449365e-06,
      "loss": 1.60622787,
      "memory(GiB)": 117.38,
      "step": 54050,
      "train_speed(iter/s)": 1.635341
    },
    {
      "acc": 0.64892731,
      "epoch": 1.3712582445459158,
      "grad_norm": 5.5625,
      "learning_rate": 2.4671978864319123e-06,
      "loss": 1.60169106,
      "memory(GiB)": 117.38,
      "step": 54055,
      "train_speed(iter/s)": 1.635358
    },
    {
      "acc": 0.65513239,
      "epoch": 1.3713850837138508,
      "grad_norm": 6.40625,
      "learning_rate": 2.466293814122835e-06,
      "loss": 1.64919014,
      "memory(GiB)": 117.38,
      "step": 54060,
      "train_speed(iter/s)": 1.635374
    },
    {
      "acc": 0.67323503,
      "epoch": 1.371511922881786,
      "grad_norm": 6.875,
      "learning_rate": 2.4653898532574684e-06,
      "loss": 1.53550606,
      "memory(GiB)": 117.38,
      "step": 54065,
      "train_speed(iter/s)": 1.635391
    },
    {
      "acc": 0.66631775,
      "epoch": 1.371638762049721,
      "grad_norm": 5.6875,
      "learning_rate": 2.4644860038755737e-06,
      "loss": 1.58006268,
      "memory(GiB)": 117.38,
      "step": 54070,
      "train_speed(iter/s)": 1.635407
    },
    {
      "acc": 0.64985061,
      "epoch": 1.371765601217656,
      "grad_norm": 7.5625,
      "learning_rate": 2.4635822660169007e-06,
      "loss": 1.6397522,
      "memory(GiB)": 117.38,
      "step": 54075,
      "train_speed(iter/s)": 1.635424
    },
    {
      "acc": 0.65005298,
      "epoch": 1.371892440385591,
      "grad_norm": 6.9375,
      "learning_rate": 2.4626786397212065e-06,
      "loss": 1.67704964,
      "memory(GiB)": 117.38,
      "step": 54080,
      "train_speed(iter/s)": 1.63544
    },
    {
      "acc": 0.6678525,
      "epoch": 1.3720192795535262,
      "grad_norm": 7.25,
      "learning_rate": 2.461775125028234e-06,
      "loss": 1.61338387,
      "memory(GiB)": 117.38,
      "step": 54085,
      "train_speed(iter/s)": 1.635456
    },
    {
      "acc": 0.63961186,
      "epoch": 1.3721461187214612,
      "grad_norm": 8.4375,
      "learning_rate": 2.4608717219777236e-06,
      "loss": 1.65973167,
      "memory(GiB)": 117.38,
      "step": 54090,
      "train_speed(iter/s)": 1.635472
    },
    {
      "acc": 0.65916772,
      "epoch": 1.3722729578893962,
      "grad_norm": 5.5625,
      "learning_rate": 2.459968430609411e-06,
      "loss": 1.58463268,
      "memory(GiB)": 117.38,
      "step": 54095,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.65970016,
      "epoch": 1.3723997970573314,
      "grad_norm": 5.78125,
      "learning_rate": 2.459065250963028e-06,
      "loss": 1.54564285,
      "memory(GiB)": 117.38,
      "step": 54100,
      "train_speed(iter/s)": 1.635504
    },
    {
      "acc": 0.6704771,
      "epoch": 1.3725266362252664,
      "grad_norm": 5.84375,
      "learning_rate": 2.458162183078299e-06,
      "loss": 1.56549339,
      "memory(GiB)": 117.38,
      "step": 54105,
      "train_speed(iter/s)": 1.63552
    },
    {
      "acc": 0.67744837,
      "epoch": 1.3726534753932014,
      "grad_norm": 5.875,
      "learning_rate": 2.4572592269949464e-06,
      "loss": 1.47730169,
      "memory(GiB)": 117.38,
      "step": 54110,
      "train_speed(iter/s)": 1.635536
    },
    {
      "acc": 0.65629334,
      "epoch": 1.3727803145611364,
      "grad_norm": 5.90625,
      "learning_rate": 2.4563563827526848e-06,
      "loss": 1.59929123,
      "memory(GiB)": 117.38,
      "step": 54115,
      "train_speed(iter/s)": 1.635552
    },
    {
      "acc": 0.64446516,
      "epoch": 1.3729071537290716,
      "grad_norm": 6.3125,
      "learning_rate": 2.455453650391226e-06,
      "loss": 1.5962491,
      "memory(GiB)": 117.38,
      "step": 54120,
      "train_speed(iter/s)": 1.635569
    },
    {
      "acc": 0.66375628,
      "epoch": 1.3730339928970066,
      "grad_norm": 5.6875,
      "learning_rate": 2.454551029950277e-06,
      "loss": 1.60608788,
      "memory(GiB)": 117.38,
      "step": 54125,
      "train_speed(iter/s)": 1.635586
    },
    {
      "acc": 0.66493359,
      "epoch": 1.3731608320649418,
      "grad_norm": 5.3125,
      "learning_rate": 2.4536485214695377e-06,
      "loss": 1.51064301,
      "memory(GiB)": 117.38,
      "step": 54130,
      "train_speed(iter/s)": 1.635603
    },
    {
      "acc": 0.65660586,
      "epoch": 1.3732876712328768,
      "grad_norm": 4.59375,
      "learning_rate": 2.4527461249887054e-06,
      "loss": 1.5727293,
      "memory(GiB)": 117.38,
      "step": 54135,
      "train_speed(iter/s)": 1.635619
    },
    {
      "acc": 0.65723758,
      "epoch": 1.3734145104008117,
      "grad_norm": 5.34375,
      "learning_rate": 2.451843840547471e-06,
      "loss": 1.57690506,
      "memory(GiB)": 117.38,
      "step": 54140,
      "train_speed(iter/s)": 1.635637
    },
    {
      "acc": 0.65991144,
      "epoch": 1.3735413495687467,
      "grad_norm": 5.6875,
      "learning_rate": 2.4509416681855193e-06,
      "loss": 1.63394127,
      "memory(GiB)": 117.38,
      "step": 54145,
      "train_speed(iter/s)": 1.635649
    },
    {
      "acc": 0.65920591,
      "epoch": 1.373668188736682,
      "grad_norm": 6.375,
      "learning_rate": 2.4500396079425377e-06,
      "loss": 1.62301826,
      "memory(GiB)": 117.38,
      "step": 54150,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.65910325,
      "epoch": 1.373795027904617,
      "grad_norm": 7.15625,
      "learning_rate": 2.4491376598581967e-06,
      "loss": 1.60065994,
      "memory(GiB)": 117.38,
      "step": 54155,
      "train_speed(iter/s)": 1.635681
    },
    {
      "acc": 0.65766726,
      "epoch": 1.373921867072552,
      "grad_norm": 6.96875,
      "learning_rate": 2.4482358239721704e-06,
      "loss": 1.62270088,
      "memory(GiB)": 117.38,
      "step": 54160,
      "train_speed(iter/s)": 1.635696
    },
    {
      "acc": 0.66118236,
      "epoch": 1.3740487062404871,
      "grad_norm": 5.71875,
      "learning_rate": 2.4473341003241234e-06,
      "loss": 1.60639324,
      "memory(GiB)": 117.38,
      "step": 54165,
      "train_speed(iter/s)": 1.635712
    },
    {
      "acc": 0.6666163,
      "epoch": 1.3741755454084221,
      "grad_norm": 6.125,
      "learning_rate": 2.446432488953724e-06,
      "loss": 1.46311569,
      "memory(GiB)": 117.38,
      "step": 54170,
      "train_speed(iter/s)": 1.635729
    },
    {
      "acc": 0.64757438,
      "epoch": 1.3743023845763571,
      "grad_norm": 6.9375,
      "learning_rate": 2.445530989900622e-06,
      "loss": 1.58069305,
      "memory(GiB)": 117.38,
      "step": 54175,
      "train_speed(iter/s)": 1.635744
    },
    {
      "acc": 0.65932388,
      "epoch": 1.374429223744292,
      "grad_norm": 6.375,
      "learning_rate": 2.4446296032044697e-06,
      "loss": 1.6389431,
      "memory(GiB)": 117.38,
      "step": 54180,
      "train_speed(iter/s)": 1.635761
    },
    {
      "acc": 0.65855026,
      "epoch": 1.3745560629122273,
      "grad_norm": 6.0,
      "learning_rate": 2.443728328904919e-06,
      "loss": 1.57429266,
      "memory(GiB)": 117.38,
      "step": 54185,
      "train_speed(iter/s)": 1.635778
    },
    {
      "acc": 0.64706569,
      "epoch": 1.3746829020801623,
      "grad_norm": 6.78125,
      "learning_rate": 2.442827167041611e-06,
      "loss": 1.63452606,
      "memory(GiB)": 117.38,
      "step": 54190,
      "train_speed(iter/s)": 1.635794
    },
    {
      "acc": 0.66797485,
      "epoch": 1.3748097412480975,
      "grad_norm": 6.125,
      "learning_rate": 2.441926117654179e-06,
      "loss": 1.57827721,
      "memory(GiB)": 117.38,
      "step": 54195,
      "train_speed(iter/s)": 1.635811
    },
    {
      "acc": 0.65382614,
      "epoch": 1.3749365804160325,
      "grad_norm": 6.5,
      "learning_rate": 2.4410251807822555e-06,
      "loss": 1.66293869,
      "memory(GiB)": 117.38,
      "step": 54200,
      "train_speed(iter/s)": 1.635828
    },
    {
      "acc": 0.6648035,
      "epoch": 1.3750634195839675,
      "grad_norm": 5.71875,
      "learning_rate": 2.4401243564654713e-06,
      "loss": 1.59469013,
      "memory(GiB)": 117.38,
      "step": 54205,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.66369085,
      "epoch": 1.3751902587519025,
      "grad_norm": 5.5625,
      "learning_rate": 2.4392236447434494e-06,
      "loss": 1.56073608,
      "memory(GiB)": 117.38,
      "step": 54210,
      "train_speed(iter/s)": 1.635862
    },
    {
      "acc": 0.66230984,
      "epoch": 1.3753170979198377,
      "grad_norm": 5.9375,
      "learning_rate": 2.4383230456558005e-06,
      "loss": 1.5442441,
      "memory(GiB)": 117.38,
      "step": 54215,
      "train_speed(iter/s)": 1.635879
    },
    {
      "acc": 0.65729065,
      "epoch": 1.3754439370877727,
      "grad_norm": 6.4375,
      "learning_rate": 2.437422559242143e-06,
      "loss": 1.56194792,
      "memory(GiB)": 117.38,
      "step": 54220,
      "train_speed(iter/s)": 1.635895
    },
    {
      "acc": 0.66322613,
      "epoch": 1.375570776255708,
      "grad_norm": 5.59375,
      "learning_rate": 2.4365221855420822e-06,
      "loss": 1.61534748,
      "memory(GiB)": 117.38,
      "step": 54225,
      "train_speed(iter/s)": 1.635913
    },
    {
      "acc": 0.65688257,
      "epoch": 1.3756976154236429,
      "grad_norm": 5.71875,
      "learning_rate": 2.435621924595221e-06,
      "loss": 1.56321974,
      "memory(GiB)": 117.38,
      "step": 54230,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.6661449,
      "epoch": 1.3758244545915779,
      "grad_norm": 5.6875,
      "learning_rate": 2.4347217764411567e-06,
      "loss": 1.65239201,
      "memory(GiB)": 117.38,
      "step": 54235,
      "train_speed(iter/s)": 1.635946
    },
    {
      "acc": 0.66999512,
      "epoch": 1.3759512937595129,
      "grad_norm": 6.09375,
      "learning_rate": 2.433821741119482e-06,
      "loss": 1.53836136,
      "memory(GiB)": 117.38,
      "step": 54240,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.64185319,
      "epoch": 1.376078132927448,
      "grad_norm": 6.875,
      "learning_rate": 2.432921818669784e-06,
      "loss": 1.63987522,
      "memory(GiB)": 117.38,
      "step": 54245,
      "train_speed(iter/s)": 1.63598
    },
    {
      "acc": 0.67233548,
      "epoch": 1.376204972095383,
      "grad_norm": 5.3125,
      "learning_rate": 2.432022009131646e-06,
      "loss": 1.55459204,
      "memory(GiB)": 117.38,
      "step": 54250,
      "train_speed(iter/s)": 1.635997
    },
    {
      "acc": 0.64748497,
      "epoch": 1.376331811263318,
      "grad_norm": 5.71875,
      "learning_rate": 2.4311223125446447e-06,
      "loss": 1.68569221,
      "memory(GiB)": 117.38,
      "step": 54255,
      "train_speed(iter/s)": 1.636014
    },
    {
      "acc": 0.64551373,
      "epoch": 1.3764586504312533,
      "grad_norm": 5.1875,
      "learning_rate": 2.4302227289483537e-06,
      "loss": 1.67029037,
      "memory(GiB)": 117.38,
      "step": 54260,
      "train_speed(iter/s)": 1.636031
    },
    {
      "acc": 0.65856504,
      "epoch": 1.3765854895991883,
      "grad_norm": 6.5625,
      "learning_rate": 2.42932325838234e-06,
      "loss": 1.62255516,
      "memory(GiB)": 117.38,
      "step": 54265,
      "train_speed(iter/s)": 1.636048
    },
    {
      "acc": 0.66313119,
      "epoch": 1.3767123287671232,
      "grad_norm": 5.34375,
      "learning_rate": 2.4284239008861665e-06,
      "loss": 1.58286352,
      "memory(GiB)": 117.38,
      "step": 54270,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.65616484,
      "epoch": 1.3768391679350582,
      "grad_norm": 6.5625,
      "learning_rate": 2.4275246564993917e-06,
      "loss": 1.62755585,
      "memory(GiB)": 117.38,
      "step": 54275,
      "train_speed(iter/s)": 1.636081
    },
    {
      "acc": 0.65982361,
      "epoch": 1.3769660071029934,
      "grad_norm": 5.6875,
      "learning_rate": 2.426625525261567e-06,
      "loss": 1.55663738,
      "memory(GiB)": 117.38,
      "step": 54280,
      "train_speed(iter/s)": 1.636097
    },
    {
      "acc": 0.6573525,
      "epoch": 1.3770928462709284,
      "grad_norm": 5.0,
      "learning_rate": 2.425726507212242e-06,
      "loss": 1.60071392,
      "memory(GiB)": 117.38,
      "step": 54285,
      "train_speed(iter/s)": 1.636114
    },
    {
      "acc": 0.66005955,
      "epoch": 1.3772196854388636,
      "grad_norm": 5.46875,
      "learning_rate": 2.424827602390958e-06,
      "loss": 1.60382023,
      "memory(GiB)": 117.38,
      "step": 54290,
      "train_speed(iter/s)": 1.636132
    },
    {
      "acc": 0.64934292,
      "epoch": 1.3773465246067986,
      "grad_norm": 4.84375,
      "learning_rate": 2.4239288108372534e-06,
      "loss": 1.59679947,
      "memory(GiB)": 117.38,
      "step": 54295,
      "train_speed(iter/s)": 1.636147
    },
    {
      "acc": 0.66654916,
      "epoch": 1.3774733637747336,
      "grad_norm": 6.65625,
      "learning_rate": 2.4230301325906606e-06,
      "loss": 1.58921251,
      "memory(GiB)": 117.38,
      "step": 54300,
      "train_speed(iter/s)": 1.636165
    },
    {
      "acc": 0.64945903,
      "epoch": 1.3776002029426686,
      "grad_norm": 5.78125,
      "learning_rate": 2.4221315676907066e-06,
      "loss": 1.64934788,
      "memory(GiB)": 117.38,
      "step": 54305,
      "train_speed(iter/s)": 1.636182
    },
    {
      "acc": 0.65386157,
      "epoch": 1.3777270421106038,
      "grad_norm": 6.09375,
      "learning_rate": 2.4212331161769194e-06,
      "loss": 1.6603405,
      "memory(GiB)": 117.38,
      "step": 54310,
      "train_speed(iter/s)": 1.636199
    },
    {
      "acc": 0.6477253,
      "epoch": 1.3778538812785388,
      "grad_norm": 5.6875,
      "learning_rate": 2.420334778088811e-06,
      "loss": 1.60431728,
      "memory(GiB)": 117.38,
      "step": 54315,
      "train_speed(iter/s)": 1.636216
    },
    {
      "acc": 0.66344695,
      "epoch": 1.3779807204464738,
      "grad_norm": 5.15625,
      "learning_rate": 2.4194365534658944e-06,
      "loss": 1.60724869,
      "memory(GiB)": 117.38,
      "step": 54320,
      "train_speed(iter/s)": 1.636233
    },
    {
      "acc": 0.64429431,
      "epoch": 1.378107559614409,
      "grad_norm": 6.125,
      "learning_rate": 2.4185384423476817e-06,
      "loss": 1.67158604,
      "memory(GiB)": 117.38,
      "step": 54325,
      "train_speed(iter/s)": 1.63625
    },
    {
      "acc": 0.64293561,
      "epoch": 1.378234398782344,
      "grad_norm": 6.15625,
      "learning_rate": 2.4176404447736758e-06,
      "loss": 1.66435394,
      "memory(GiB)": 117.38,
      "step": 54330,
      "train_speed(iter/s)": 1.636268
    },
    {
      "acc": 0.64840803,
      "epoch": 1.378361237950279,
      "grad_norm": 6.59375,
      "learning_rate": 2.41674256078337e-06,
      "loss": 1.59952517,
      "memory(GiB)": 117.38,
      "step": 54335,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.64119782,
      "epoch": 1.378488077118214,
      "grad_norm": 5.59375,
      "learning_rate": 2.4158447904162585e-06,
      "loss": 1.68831482,
      "memory(GiB)": 117.38,
      "step": 54340,
      "train_speed(iter/s)": 1.636302
    },
    {
      "acc": 0.65265064,
      "epoch": 1.3786149162861492,
      "grad_norm": 6.3125,
      "learning_rate": 2.414947133711832e-06,
      "loss": 1.60845928,
      "memory(GiB)": 117.38,
      "step": 54345,
      "train_speed(iter/s)": 1.636319
    },
    {
      "acc": 0.64305625,
      "epoch": 1.3787417554540842,
      "grad_norm": 5.375,
      "learning_rate": 2.414049590709574e-06,
      "loss": 1.60407028,
      "memory(GiB)": 117.38,
      "step": 54350,
      "train_speed(iter/s)": 1.636337
    },
    {
      "acc": 0.6422936,
      "epoch": 1.3788685946220194,
      "grad_norm": 5.84375,
      "learning_rate": 2.4131521614489567e-06,
      "loss": 1.6680769,
      "memory(GiB)": 117.38,
      "step": 54355,
      "train_speed(iter/s)": 1.636355
    },
    {
      "acc": 0.66786118,
      "epoch": 1.3789954337899544,
      "grad_norm": 5.0625,
      "learning_rate": 2.412254845969459e-06,
      "loss": 1.52127104,
      "memory(GiB)": 117.38,
      "step": 54360,
      "train_speed(iter/s)": 1.636372
    },
    {
      "acc": 0.66609874,
      "epoch": 1.3791222729578894,
      "grad_norm": 5.0625,
      "learning_rate": 2.4113576443105464e-06,
      "loss": 1.54702654,
      "memory(GiB)": 117.38,
      "step": 54365,
      "train_speed(iter/s)": 1.63639
    },
    {
      "acc": 0.64409099,
      "epoch": 1.3792491121258244,
      "grad_norm": 5.6875,
      "learning_rate": 2.410460556511684e-06,
      "loss": 1.60263042,
      "memory(GiB)": 117.38,
      "step": 54370,
      "train_speed(iter/s)": 1.636407
    },
    {
      "acc": 0.6562439,
      "epoch": 1.3793759512937596,
      "grad_norm": 5.71875,
      "learning_rate": 2.4095635826123235e-06,
      "loss": 1.56911697,
      "memory(GiB)": 117.38,
      "step": 54375,
      "train_speed(iter/s)": 1.636424
    },
    {
      "acc": 0.66322141,
      "epoch": 1.3795027904616946,
      "grad_norm": 5.21875,
      "learning_rate": 2.4086667226519245e-06,
      "loss": 1.57778282,
      "memory(GiB)": 117.38,
      "step": 54380,
      "train_speed(iter/s)": 1.63644
    },
    {
      "acc": 0.65481992,
      "epoch": 1.3796296296296298,
      "grad_norm": 5.875,
      "learning_rate": 2.4077699766699323e-06,
      "loss": 1.64075966,
      "memory(GiB)": 117.38,
      "step": 54385,
      "train_speed(iter/s)": 1.636457
    },
    {
      "acc": 0.66791763,
      "epoch": 1.3797564687975648,
      "grad_norm": 4.96875,
      "learning_rate": 2.4068733447057903e-06,
      "loss": 1.5885067,
      "memory(GiB)": 117.38,
      "step": 54390,
      "train_speed(iter/s)": 1.636475
    },
    {
      "acc": 0.62991519,
      "epoch": 1.3798833079654997,
      "grad_norm": 5.375,
      "learning_rate": 2.405976826798936e-06,
      "loss": 1.69023056,
      "memory(GiB)": 117.38,
      "step": 54395,
      "train_speed(iter/s)": 1.636493
    },
    {
      "acc": 0.66863132,
      "epoch": 1.3800101471334347,
      "grad_norm": 4.8125,
      "learning_rate": 2.405080422988802e-06,
      "loss": 1.57402372,
      "memory(GiB)": 117.38,
      "step": 54400,
      "train_speed(iter/s)": 1.63651
    },
    {
      "acc": 0.65594616,
      "epoch": 1.38013698630137,
      "grad_norm": 5.6875,
      "learning_rate": 2.404184133314817e-06,
      "loss": 1.66446877,
      "memory(GiB)": 117.38,
      "step": 54405,
      "train_speed(iter/s)": 1.636526
    },
    {
      "acc": 0.66494308,
      "epoch": 1.380263825469305,
      "grad_norm": 6.71875,
      "learning_rate": 2.4032879578164027e-06,
      "loss": 1.59335918,
      "memory(GiB)": 117.38,
      "step": 54410,
      "train_speed(iter/s)": 1.636542
    },
    {
      "acc": 0.65673313,
      "epoch": 1.38039066463724,
      "grad_norm": 5.71875,
      "learning_rate": 2.402391896532978e-06,
      "loss": 1.58920755,
      "memory(GiB)": 117.38,
      "step": 54415,
      "train_speed(iter/s)": 1.636559
    },
    {
      "acc": 0.64633751,
      "epoch": 1.3805175038051751,
      "grad_norm": 5.59375,
      "learning_rate": 2.4014959495039548e-06,
      "loss": 1.6979847,
      "memory(GiB)": 117.38,
      "step": 54420,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.65312424,
      "epoch": 1.3806443429731101,
      "grad_norm": 5.75,
      "learning_rate": 2.4006001167687416e-06,
      "loss": 1.63549957,
      "memory(GiB)": 117.38,
      "step": 54425,
      "train_speed(iter/s)": 1.636592
    },
    {
      "acc": 0.66206479,
      "epoch": 1.380771182141045,
      "grad_norm": 5.03125,
      "learning_rate": 2.39970439836674e-06,
      "loss": 1.5907217,
      "memory(GiB)": 117.38,
      "step": 54430,
      "train_speed(iter/s)": 1.636608
    },
    {
      "acc": 0.647054,
      "epoch": 1.38089802130898,
      "grad_norm": 5.25,
      "learning_rate": 2.3988087943373497e-06,
      "loss": 1.59917736,
      "memory(GiB)": 117.38,
      "step": 54435,
      "train_speed(iter/s)": 1.636625
    },
    {
      "acc": 0.67288084,
      "epoch": 1.3810248604769153,
      "grad_norm": 5.1875,
      "learning_rate": 2.397913304719961e-06,
      "loss": 1.51494026,
      "memory(GiB)": 117.38,
      "step": 54440,
      "train_speed(iter/s)": 1.636642
    },
    {
      "acc": 0.65642438,
      "epoch": 1.3811516996448503,
      "grad_norm": 6.1875,
      "learning_rate": 2.397017929553961e-06,
      "loss": 1.56472073,
      "memory(GiB)": 117.38,
      "step": 54445,
      "train_speed(iter/s)": 1.636658
    },
    {
      "acc": 0.66393561,
      "epoch": 1.3812785388127855,
      "grad_norm": 5.71875,
      "learning_rate": 2.396122668878738e-06,
      "loss": 1.5561264,
      "memory(GiB)": 117.38,
      "step": 54450,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.64073334,
      "epoch": 1.3814053779807205,
      "grad_norm": 5.625,
      "learning_rate": 2.3952275227336636e-06,
      "loss": 1.73151112,
      "memory(GiB)": 117.38,
      "step": 54455,
      "train_speed(iter/s)": 1.636692
    },
    {
      "acc": 0.65882759,
      "epoch": 1.3815322171486555,
      "grad_norm": 5.15625,
      "learning_rate": 2.3943324911581117e-06,
      "loss": 1.56787491,
      "memory(GiB)": 117.38,
      "step": 54460,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.65736914,
      "epoch": 1.3816590563165905,
      "grad_norm": 5.0625,
      "learning_rate": 2.393437574191449e-06,
      "loss": 1.57225094,
      "memory(GiB)": 117.38,
      "step": 54465,
      "train_speed(iter/s)": 1.636725
    },
    {
      "acc": 0.65833235,
      "epoch": 1.3817858954845257,
      "grad_norm": 5.21875,
      "learning_rate": 2.3925427718730426e-06,
      "loss": 1.59292936,
      "memory(GiB)": 117.38,
      "step": 54470,
      "train_speed(iter/s)": 1.636742
    },
    {
      "acc": 0.6675139,
      "epoch": 1.3819127346524607,
      "grad_norm": 6.3125,
      "learning_rate": 2.391648084242245e-06,
      "loss": 1.52889223,
      "memory(GiB)": 117.38,
      "step": 54475,
      "train_speed(iter/s)": 1.636757
    },
    {
      "acc": 0.65570593,
      "epoch": 1.3820395738203957,
      "grad_norm": 5.65625,
      "learning_rate": 2.3907535113384084e-06,
      "loss": 1.64451466,
      "memory(GiB)": 117.38,
      "step": 54480,
      "train_speed(iter/s)": 1.636773
    },
    {
      "acc": 0.65295038,
      "epoch": 1.3821664129883309,
      "grad_norm": 6.0,
      "learning_rate": 2.389859053200883e-06,
      "loss": 1.63188324,
      "memory(GiB)": 117.38,
      "step": 54485,
      "train_speed(iter/s)": 1.63679
    },
    {
      "acc": 0.66879358,
      "epoch": 1.3822932521562659,
      "grad_norm": 5.875,
      "learning_rate": 2.3889647098690127e-06,
      "loss": 1.58088703,
      "memory(GiB)": 117.38,
      "step": 54490,
      "train_speed(iter/s)": 1.636805
    },
    {
      "acc": 0.65506191,
      "epoch": 1.3824200913242009,
      "grad_norm": 5.53125,
      "learning_rate": 2.3880704813821275e-06,
      "loss": 1.59755859,
      "memory(GiB)": 117.38,
      "step": 54495,
      "train_speed(iter/s)": 1.636822
    },
    {
      "acc": 0.65546737,
      "epoch": 1.3825469304921358,
      "grad_norm": 4.9375,
      "learning_rate": 2.3871763677795656e-06,
      "loss": 1.62362061,
      "memory(GiB)": 117.38,
      "step": 54500,
      "train_speed(iter/s)": 1.63684
    },
    {
      "acc": 0.65757484,
      "epoch": 1.382673769660071,
      "grad_norm": 6.0625,
      "learning_rate": 2.386282369100653e-06,
      "loss": 1.58688421,
      "memory(GiB)": 117.38,
      "step": 54505,
      "train_speed(iter/s)": 1.636857
    },
    {
      "acc": 0.65478721,
      "epoch": 1.382800608828006,
      "grad_norm": 5.375,
      "learning_rate": 2.385388485384713e-06,
      "loss": 1.65988483,
      "memory(GiB)": 117.38,
      "step": 54510,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.66759429,
      "epoch": 1.3829274479959413,
      "grad_norm": 6.28125,
      "learning_rate": 2.384494716671057e-06,
      "loss": 1.59211254,
      "memory(GiB)": 117.38,
      "step": 54515,
      "train_speed(iter/s)": 1.63689
    },
    {
      "acc": 0.67711535,
      "epoch": 1.3830542871638762,
      "grad_norm": 6.03125,
      "learning_rate": 2.3836010629990027e-06,
      "loss": 1.47437019,
      "memory(GiB)": 117.38,
      "step": 54520,
      "train_speed(iter/s)": 1.636907
    },
    {
      "acc": 0.65576553,
      "epoch": 1.3831811263318112,
      "grad_norm": 7.9375,
      "learning_rate": 2.382707524407855e-06,
      "loss": 1.62105312,
      "memory(GiB)": 117.38,
      "step": 54525,
      "train_speed(iter/s)": 1.636924
    },
    {
      "acc": 0.65163727,
      "epoch": 1.3833079654997462,
      "grad_norm": 5.71875,
      "learning_rate": 2.3818141009369155e-06,
      "loss": 1.61962643,
      "memory(GiB)": 117.38,
      "step": 54530,
      "train_speed(iter/s)": 1.636941
    },
    {
      "acc": 0.65830121,
      "epoch": 1.3834348046676814,
      "grad_norm": 5.4375,
      "learning_rate": 2.3809207926254813e-06,
      "loss": 1.61833763,
      "memory(GiB)": 117.38,
      "step": 54535,
      "train_speed(iter/s)": 1.636959
    },
    {
      "acc": 0.67367992,
      "epoch": 1.3835616438356164,
      "grad_norm": 5.40625,
      "learning_rate": 2.380027599512844e-06,
      "loss": 1.59038191,
      "memory(GiB)": 117.38,
      "step": 54540,
      "train_speed(iter/s)": 1.636976
    },
    {
      "acc": 0.65162201,
      "epoch": 1.3836884830035516,
      "grad_norm": 7.59375,
      "learning_rate": 2.3791345216382906e-06,
      "loss": 1.6196228,
      "memory(GiB)": 117.38,
      "step": 54545,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.65452595,
      "epoch": 1.3838153221714866,
      "grad_norm": 5.78125,
      "learning_rate": 2.378241559041102e-06,
      "loss": 1.63103619,
      "memory(GiB)": 117.38,
      "step": 54550,
      "train_speed(iter/s)": 1.637011
    },
    {
      "acc": 0.6582305,
      "epoch": 1.3839421613394216,
      "grad_norm": 5.875,
      "learning_rate": 2.377348711760555e-06,
      "loss": 1.6536974,
      "memory(GiB)": 117.38,
      "step": 54555,
      "train_speed(iter/s)": 1.637027
    },
    {
      "acc": 0.64812927,
      "epoch": 1.3840690005073566,
      "grad_norm": 6.53125,
      "learning_rate": 2.3764559798359204e-06,
      "loss": 1.6553772,
      "memory(GiB)": 117.38,
      "step": 54560,
      "train_speed(iter/s)": 1.637046
    },
    {
      "acc": 0.6654295,
      "epoch": 1.3841958396752918,
      "grad_norm": 5.125,
      "learning_rate": 2.3755633633064658e-06,
      "loss": 1.51865454,
      "memory(GiB)": 117.38,
      "step": 54565,
      "train_speed(iter/s)": 1.637062
    },
    {
      "acc": 0.62914252,
      "epoch": 1.3843226788432268,
      "grad_norm": 4.71875,
      "learning_rate": 2.374670862211451e-06,
      "loss": 1.70866776,
      "memory(GiB)": 117.38,
      "step": 54570,
      "train_speed(iter/s)": 1.63708
    },
    {
      "acc": 0.66669559,
      "epoch": 1.3844495180111618,
      "grad_norm": 6.5625,
      "learning_rate": 2.373778476590134e-06,
      "loss": 1.56118927,
      "memory(GiB)": 117.38,
      "step": 54575,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.66218114,
      "epoch": 1.384576357179097,
      "grad_norm": 5.4375,
      "learning_rate": 2.372886206481764e-06,
      "loss": 1.58207045,
      "memory(GiB)": 117.38,
      "step": 54580,
      "train_speed(iter/s)": 1.637114
    },
    {
      "acc": 0.64545379,
      "epoch": 1.384703196347032,
      "grad_norm": 4.5625,
      "learning_rate": 2.3719940519255864e-06,
      "loss": 1.64367085,
      "memory(GiB)": 117.38,
      "step": 54585,
      "train_speed(iter/s)": 1.637131
    },
    {
      "acc": 0.64167919,
      "epoch": 1.384830035514967,
      "grad_norm": 6.21875,
      "learning_rate": 2.371102012960847e-06,
      "loss": 1.66074333,
      "memory(GiB)": 117.38,
      "step": 54590,
      "train_speed(iter/s)": 1.637147
    },
    {
      "acc": 0.65991077,
      "epoch": 1.384956874682902,
      "grad_norm": 7.5625,
      "learning_rate": 2.3702100896267767e-06,
      "loss": 1.58650208,
      "memory(GiB)": 117.38,
      "step": 54595,
      "train_speed(iter/s)": 1.637164
    },
    {
      "acc": 0.66061969,
      "epoch": 1.3850837138508372,
      "grad_norm": 4.8125,
      "learning_rate": 2.3693182819626077e-06,
      "loss": 1.59709663,
      "memory(GiB)": 117.38,
      "step": 54600,
      "train_speed(iter/s)": 1.637179
    },
    {
      "acc": 0.64905834,
      "epoch": 1.3852105530187722,
      "grad_norm": 6.1875,
      "learning_rate": 2.3684265900075637e-06,
      "loss": 1.61571255,
      "memory(GiB)": 117.38,
      "step": 54605,
      "train_speed(iter/s)": 1.637196
    },
    {
      "acc": 0.662532,
      "epoch": 1.3853373921867074,
      "grad_norm": 5.21875,
      "learning_rate": 2.3675350138008714e-06,
      "loss": 1.61051311,
      "memory(GiB)": 117.38,
      "step": 54610,
      "train_speed(iter/s)": 1.637212
    },
    {
      "acc": 0.66010008,
      "epoch": 1.3854642313546424,
      "grad_norm": 5.53125,
      "learning_rate": 2.3666435533817406e-06,
      "loss": 1.6194458,
      "memory(GiB)": 117.38,
      "step": 54615,
      "train_speed(iter/s)": 1.637227
    },
    {
      "acc": 0.65904284,
      "epoch": 1.3855910705225774,
      "grad_norm": 7.125,
      "learning_rate": 2.3657522087893806e-06,
      "loss": 1.61277561,
      "memory(GiB)": 117.38,
      "step": 54620,
      "train_speed(iter/s)": 1.637244
    },
    {
      "acc": 0.66919174,
      "epoch": 1.3857179096905123,
      "grad_norm": 5.34375,
      "learning_rate": 2.3648609800630022e-06,
      "loss": 1.57579651,
      "memory(GiB)": 117.38,
      "step": 54625,
      "train_speed(iter/s)": 1.63726
    },
    {
      "acc": 0.64660482,
      "epoch": 1.3858447488584476,
      "grad_norm": 5.3125,
      "learning_rate": 2.363969867241805e-06,
      "loss": 1.59215279,
      "memory(GiB)": 117.38,
      "step": 54630,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.64593916,
      "epoch": 1.3859715880263825,
      "grad_norm": 6.09375,
      "learning_rate": 2.36307887036498e-06,
      "loss": 1.60736256,
      "memory(GiB)": 117.38,
      "step": 54635,
      "train_speed(iter/s)": 1.637294
    },
    {
      "acc": 0.64959455,
      "epoch": 1.3860984271943175,
      "grad_norm": 5.4375,
      "learning_rate": 2.3621879894717177e-06,
      "loss": 1.57939568,
      "memory(GiB)": 117.38,
      "step": 54640,
      "train_speed(iter/s)": 1.637311
    },
    {
      "acc": 0.67929225,
      "epoch": 1.3862252663622527,
      "grad_norm": 6.40625,
      "learning_rate": 2.361297224601206e-06,
      "loss": 1.55587587,
      "memory(GiB)": 117.38,
      "step": 54645,
      "train_speed(iter/s)": 1.637327
    },
    {
      "acc": 0.67475038,
      "epoch": 1.3863521055301877,
      "grad_norm": 5.625,
      "learning_rate": 2.360406575792625e-06,
      "loss": 1.52268801,
      "memory(GiB)": 117.38,
      "step": 54650,
      "train_speed(iter/s)": 1.637342
    },
    {
      "acc": 0.67204671,
      "epoch": 1.3864789446981227,
      "grad_norm": 5.34375,
      "learning_rate": 2.3595160430851445e-06,
      "loss": 1.49425068,
      "memory(GiB)": 117.38,
      "step": 54655,
      "train_speed(iter/s)": 1.637359
    },
    {
      "acc": 0.6472033,
      "epoch": 1.3866057838660577,
      "grad_norm": 7.28125,
      "learning_rate": 2.3586256265179392e-06,
      "loss": 1.62397461,
      "memory(GiB)": 117.38,
      "step": 54660,
      "train_speed(iter/s)": 1.637375
    },
    {
      "acc": 0.66439342,
      "epoch": 1.386732623033993,
      "grad_norm": 6.09375,
      "learning_rate": 2.3577353261301715e-06,
      "loss": 1.52258282,
      "memory(GiB)": 117.38,
      "step": 54665,
      "train_speed(iter/s)": 1.637391
    },
    {
      "acc": 0.65613689,
      "epoch": 1.386859462201928,
      "grad_norm": 6.40625,
      "learning_rate": 2.356845141961001e-06,
      "loss": 1.63419609,
      "memory(GiB)": 117.38,
      "step": 54670,
      "train_speed(iter/s)": 1.637406
    },
    {
      "acc": 0.64701138,
      "epoch": 1.3869863013698631,
      "grad_norm": 5.65625,
      "learning_rate": 2.355955074049582e-06,
      "loss": 1.600383,
      "memory(GiB)": 117.38,
      "step": 54675,
      "train_speed(iter/s)": 1.637423
    },
    {
      "acc": 0.66493602,
      "epoch": 1.387113140537798,
      "grad_norm": 6.65625,
      "learning_rate": 2.355065122435064e-06,
      "loss": 1.57711973,
      "memory(GiB)": 117.38,
      "step": 54680,
      "train_speed(iter/s)": 1.63744
    },
    {
      "acc": 0.65375957,
      "epoch": 1.387239979705733,
      "grad_norm": 6.40625,
      "learning_rate": 2.3541752871565902e-06,
      "loss": 1.63986969,
      "memory(GiB)": 117.38,
      "step": 54685,
      "train_speed(iter/s)": 1.637456
    },
    {
      "acc": 0.65510831,
      "epoch": 1.387366818873668,
      "grad_norm": 4.78125,
      "learning_rate": 2.3532855682533003e-06,
      "loss": 1.59486332,
      "memory(GiB)": 117.38,
      "step": 54690,
      "train_speed(iter/s)": 1.637472
    },
    {
      "acc": 0.66585188,
      "epoch": 1.3874936580416033,
      "grad_norm": 6.34375,
      "learning_rate": 2.352395965764328e-06,
      "loss": 1.59426537,
      "memory(GiB)": 117.38,
      "step": 54695,
      "train_speed(iter/s)": 1.637488
    },
    {
      "acc": 0.65251646,
      "epoch": 1.3876204972095383,
      "grad_norm": 4.96875,
      "learning_rate": 2.3515064797288013e-06,
      "loss": 1.5913372,
      "memory(GiB)": 117.38,
      "step": 54700,
      "train_speed(iter/s)": 1.637504
    },
    {
      "acc": 0.65431547,
      "epoch": 1.3877473363774735,
      "grad_norm": 5.125,
      "learning_rate": 2.350617110185845e-06,
      "loss": 1.65106812,
      "memory(GiB)": 117.38,
      "step": 54705,
      "train_speed(iter/s)": 1.63752
    },
    {
      "acc": 0.65923738,
      "epoch": 1.3878741755454085,
      "grad_norm": 5.21875,
      "learning_rate": 2.3497278571745763e-06,
      "loss": 1.59178038,
      "memory(GiB)": 117.38,
      "step": 54710,
      "train_speed(iter/s)": 1.637537
    },
    {
      "acc": 0.66219721,
      "epoch": 1.3880010147133435,
      "grad_norm": 5.875,
      "learning_rate": 2.348838720734109e-06,
      "loss": 1.55047865,
      "memory(GiB)": 117.38,
      "step": 54715,
      "train_speed(iter/s)": 1.637554
    },
    {
      "acc": 0.66923361,
      "epoch": 1.3881278538812785,
      "grad_norm": 6.125,
      "learning_rate": 2.347949700903552e-06,
      "loss": 1.62030277,
      "memory(GiB)": 117.38,
      "step": 54720,
      "train_speed(iter/s)": 1.63757
    },
    {
      "acc": 0.65253315,
      "epoch": 1.3882546930492137,
      "grad_norm": 6.65625,
      "learning_rate": 2.3470607977220066e-06,
      "loss": 1.57142239,
      "memory(GiB)": 117.38,
      "step": 54725,
      "train_speed(iter/s)": 1.637586
    },
    {
      "acc": 0.65744362,
      "epoch": 1.3883815322171487,
      "grad_norm": 5.34375,
      "learning_rate": 2.346172011228573e-06,
      "loss": 1.62530308,
      "memory(GiB)": 117.38,
      "step": 54730,
      "train_speed(iter/s)": 1.637604
    },
    {
      "acc": 0.6665256,
      "epoch": 1.3885083713850837,
      "grad_norm": 5.5625,
      "learning_rate": 2.345283341462342e-06,
      "loss": 1.56241436,
      "memory(GiB)": 117.38,
      "step": 54735,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.65214968,
      "epoch": 1.3886352105530189,
      "grad_norm": 5.46875,
      "learning_rate": 2.3443947884624026e-06,
      "loss": 1.6450182,
      "memory(GiB)": 117.38,
      "step": 54740,
      "train_speed(iter/s)": 1.637636
    },
    {
      "acc": 0.66708302,
      "epoch": 1.3887620497209539,
      "grad_norm": 5.625,
      "learning_rate": 2.3435063522678346e-06,
      "loss": 1.54241266,
      "memory(GiB)": 117.38,
      "step": 54745,
      "train_speed(iter/s)": 1.637652
    },
    {
      "acc": 0.65314879,
      "epoch": 1.3888888888888888,
      "grad_norm": 4.875,
      "learning_rate": 2.3426180329177217e-06,
      "loss": 1.61339149,
      "memory(GiB)": 117.38,
      "step": 54750,
      "train_speed(iter/s)": 1.637668
    },
    {
      "acc": 0.6494813,
      "epoch": 1.3890157280568238,
      "grad_norm": 5.03125,
      "learning_rate": 2.3417298304511297e-06,
      "loss": 1.5931282,
      "memory(GiB)": 117.38,
      "step": 54755,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.63992729,
      "epoch": 1.389142567224759,
      "grad_norm": 5.53125,
      "learning_rate": 2.340841744907127e-06,
      "loss": 1.70688591,
      "memory(GiB)": 117.38,
      "step": 54760,
      "train_speed(iter/s)": 1.637701
    },
    {
      "acc": 0.66486187,
      "epoch": 1.389269406392694,
      "grad_norm": 6.5,
      "learning_rate": 2.3399537763247783e-06,
      "loss": 1.53978243,
      "memory(GiB)": 117.38,
      "step": 54765,
      "train_speed(iter/s)": 1.637718
    },
    {
      "acc": 0.64669728,
      "epoch": 1.3893962455606292,
      "grad_norm": 5.375,
      "learning_rate": 2.3390659247431404e-06,
      "loss": 1.60624332,
      "memory(GiB)": 117.38,
      "step": 54770,
      "train_speed(iter/s)": 1.637734
    },
    {
      "acc": 0.66217871,
      "epoch": 1.3895230847285642,
      "grad_norm": 6.34375,
      "learning_rate": 2.338178190201261e-06,
      "loss": 1.56973162,
      "memory(GiB)": 117.38,
      "step": 54775,
      "train_speed(iter/s)": 1.637751
    },
    {
      "acc": 0.65562439,
      "epoch": 1.3896499238964992,
      "grad_norm": 5.8125,
      "learning_rate": 2.3372905727381877e-06,
      "loss": 1.60259323,
      "memory(GiB)": 117.38,
      "step": 54780,
      "train_speed(iter/s)": 1.637767
    },
    {
      "acc": 0.67119951,
      "epoch": 1.3897767630644342,
      "grad_norm": 5.8125,
      "learning_rate": 2.3364030723929647e-06,
      "loss": 1.54498358,
      "memory(GiB)": 117.38,
      "step": 54785,
      "train_speed(iter/s)": 1.637783
    },
    {
      "acc": 0.66917076,
      "epoch": 1.3899036022323694,
      "grad_norm": 5.375,
      "learning_rate": 2.335515689204629e-06,
      "loss": 1.55101547,
      "memory(GiB)": 117.38,
      "step": 54790,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.65899162,
      "epoch": 1.3900304414003044,
      "grad_norm": 6.25,
      "learning_rate": 2.334628423212206e-06,
      "loss": 1.59170866,
      "memory(GiB)": 117.38,
      "step": 54795,
      "train_speed(iter/s)": 1.637816
    },
    {
      "acc": 0.65975137,
      "epoch": 1.3901572805682394,
      "grad_norm": 6.40625,
      "learning_rate": 2.3337412744547256e-06,
      "loss": 1.58847504,
      "memory(GiB)": 117.38,
      "step": 54800,
      "train_speed(iter/s)": 1.637832
    },
    {
      "acc": 0.66842957,
      "epoch": 1.3902841197361746,
      "grad_norm": 5.59375,
      "learning_rate": 2.332854242971209e-06,
      "loss": 1.56257715,
      "memory(GiB)": 117.38,
      "step": 54805,
      "train_speed(iter/s)": 1.637849
    },
    {
      "acc": 0.65224628,
      "epoch": 1.3904109589041096,
      "grad_norm": 8.0625,
      "learning_rate": 2.331967328800672e-06,
      "loss": 1.58485088,
      "memory(GiB)": 117.38,
      "step": 54810,
      "train_speed(iter/s)": 1.637866
    },
    {
      "acc": 0.65946445,
      "epoch": 1.3905377980720446,
      "grad_norm": 6.1875,
      "learning_rate": 2.33108053198212e-06,
      "loss": 1.53212433,
      "memory(GiB)": 117.38,
      "step": 54815,
      "train_speed(iter/s)": 1.637882
    },
    {
      "acc": 0.65844302,
      "epoch": 1.3906646372399796,
      "grad_norm": 5.4375,
      "learning_rate": 2.330193852554564e-06,
      "loss": 1.57762346,
      "memory(GiB)": 117.38,
      "step": 54820,
      "train_speed(iter/s)": 1.637899
    },
    {
      "acc": 0.66593008,
      "epoch": 1.3907914764079148,
      "grad_norm": 6.90625,
      "learning_rate": 2.3293072905570024e-06,
      "loss": 1.60221062,
      "memory(GiB)": 117.38,
      "step": 54825,
      "train_speed(iter/s)": 1.637914
    },
    {
      "acc": 0.67084436,
      "epoch": 1.3909183155758498,
      "grad_norm": 4.78125,
      "learning_rate": 2.3284208460284303e-06,
      "loss": 1.56453915,
      "memory(GiB)": 117.38,
      "step": 54830,
      "train_speed(iter/s)": 1.63793
    },
    {
      "acc": 0.66992269,
      "epoch": 1.391045154743785,
      "grad_norm": 6.46875,
      "learning_rate": 2.3275345190078364e-06,
      "loss": 1.60624962,
      "memory(GiB)": 117.38,
      "step": 54835,
      "train_speed(iter/s)": 1.637947
    },
    {
      "acc": 0.64309888,
      "epoch": 1.39117199391172,
      "grad_norm": 5.40625,
      "learning_rate": 2.3266483095342064e-06,
      "loss": 1.62993565,
      "memory(GiB)": 117.38,
      "step": 54840,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.66009889,
      "epoch": 1.391298833079655,
      "grad_norm": 5.6875,
      "learning_rate": 2.3257622176465194e-06,
      "loss": 1.54410744,
      "memory(GiB)": 117.38,
      "step": 54845,
      "train_speed(iter/s)": 1.63798
    },
    {
      "acc": 0.65634217,
      "epoch": 1.39142567224759,
      "grad_norm": 6.34375,
      "learning_rate": 2.3248762433837494e-06,
      "loss": 1.63350487,
      "memory(GiB)": 117.38,
      "step": 54850,
      "train_speed(iter/s)": 1.637997
    },
    {
      "acc": 0.64866471,
      "epoch": 1.3915525114155252,
      "grad_norm": 5.28125,
      "learning_rate": 2.323990386784867e-06,
      "loss": 1.58710537,
      "memory(GiB)": 117.38,
      "step": 54855,
      "train_speed(iter/s)": 1.638014
    },
    {
      "acc": 0.66248913,
      "epoch": 1.3916793505834602,
      "grad_norm": 8.5,
      "learning_rate": 2.3231046478888335e-06,
      "loss": 1.60312004,
      "memory(GiB)": 117.38,
      "step": 54860,
      "train_speed(iter/s)": 1.638031
    },
    {
      "acc": 0.65088711,
      "epoch": 1.3918061897513954,
      "grad_norm": 5.75,
      "learning_rate": 2.3222190267346094e-06,
      "loss": 1.60557976,
      "memory(GiB)": 117.38,
      "step": 54865,
      "train_speed(iter/s)": 1.638049
    },
    {
      "acc": 0.65654144,
      "epoch": 1.3919330289193304,
      "grad_norm": 5.78125,
      "learning_rate": 2.3213335233611484e-06,
      "loss": 1.60163422,
      "memory(GiB)": 117.38,
      "step": 54870,
      "train_speed(iter/s)": 1.638066
    },
    {
      "acc": 0.65225067,
      "epoch": 1.3920598680872653,
      "grad_norm": 6.96875,
      "learning_rate": 2.320448137807398e-06,
      "loss": 1.59785175,
      "memory(GiB)": 117.38,
      "step": 54875,
      "train_speed(iter/s)": 1.638083
    },
    {
      "acc": 0.65564923,
      "epoch": 1.3921867072552003,
      "grad_norm": 5.8125,
      "learning_rate": 2.3195628701123017e-06,
      "loss": 1.6506506,
      "memory(GiB)": 117.38,
      "step": 54880,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.66144891,
      "epoch": 1.3923135464231355,
      "grad_norm": 5.25,
      "learning_rate": 2.3186777203147964e-06,
      "loss": 1.57962303,
      "memory(GiB)": 117.38,
      "step": 54885,
      "train_speed(iter/s)": 1.638117
    },
    {
      "acc": 0.66502342,
      "epoch": 1.3924403855910705,
      "grad_norm": 6.09375,
      "learning_rate": 2.3177926884538193e-06,
      "loss": 1.63195229,
      "memory(GiB)": 117.38,
      "step": 54890,
      "train_speed(iter/s)": 1.638135
    },
    {
      "acc": 0.66079597,
      "epoch": 1.3925672247590055,
      "grad_norm": 6.15625,
      "learning_rate": 2.3169077745682933e-06,
      "loss": 1.575457,
      "memory(GiB)": 117.38,
      "step": 54895,
      "train_speed(iter/s)": 1.638153
    },
    {
      "acc": 0.65259972,
      "epoch": 1.3926940639269407,
      "grad_norm": 6.0,
      "learning_rate": 2.316022978697143e-06,
      "loss": 1.58997059,
      "memory(GiB)": 117.38,
      "step": 54900,
      "train_speed(iter/s)": 1.63817
    },
    {
      "acc": 0.65334277,
      "epoch": 1.3928209030948757,
      "grad_norm": 5.15625,
      "learning_rate": 2.3151383008792826e-06,
      "loss": 1.59839573,
      "memory(GiB)": 117.38,
      "step": 54905,
      "train_speed(iter/s)": 1.638187
    },
    {
      "acc": 0.65705514,
      "epoch": 1.3929477422628107,
      "grad_norm": 5.8125,
      "learning_rate": 2.314253741153631e-06,
      "loss": 1.58769064,
      "memory(GiB)": 117.38,
      "step": 54910,
      "train_speed(iter/s)": 1.638203
    },
    {
      "acc": 0.65232139,
      "epoch": 1.3930745814307457,
      "grad_norm": 6.34375,
      "learning_rate": 2.313369299559088e-06,
      "loss": 1.55695667,
      "memory(GiB)": 117.38,
      "step": 54915,
      "train_speed(iter/s)": 1.63822
    },
    {
      "acc": 0.6594377,
      "epoch": 1.393201420598681,
      "grad_norm": 4.6875,
      "learning_rate": 2.3124849761345576e-06,
      "loss": 1.50405865,
      "memory(GiB)": 117.38,
      "step": 54920,
      "train_speed(iter/s)": 1.638236
    },
    {
      "acc": 0.66456151,
      "epoch": 1.393328259766616,
      "grad_norm": 4.71875,
      "learning_rate": 2.311600770918938e-06,
      "loss": 1.59530659,
      "memory(GiB)": 117.38,
      "step": 54925,
      "train_speed(iter/s)": 1.638254
    },
    {
      "acc": 0.66930633,
      "epoch": 1.393455098934551,
      "grad_norm": 5.15625,
      "learning_rate": 2.310716683951122e-06,
      "loss": 1.57571974,
      "memory(GiB)": 117.38,
      "step": 54930,
      "train_speed(iter/s)": 1.63827
    },
    {
      "acc": 0.6582674,
      "epoch": 1.393581938102486,
      "grad_norm": 7.53125,
      "learning_rate": 2.3098327152699884e-06,
      "loss": 1.65666256,
      "memory(GiB)": 117.38,
      "step": 54935,
      "train_speed(iter/s)": 1.638288
    },
    {
      "acc": 0.6635499,
      "epoch": 1.393708777270421,
      "grad_norm": 5.84375,
      "learning_rate": 2.308948864914425e-06,
      "loss": 1.60049973,
      "memory(GiB)": 117.38,
      "step": 54940,
      "train_speed(iter/s)": 1.638304
    },
    {
      "acc": 0.65124331,
      "epoch": 1.393835616438356,
      "grad_norm": 5.84375,
      "learning_rate": 2.308065132923305e-06,
      "loss": 1.63863602,
      "memory(GiB)": 117.38,
      "step": 54945,
      "train_speed(iter/s)": 1.638321
    },
    {
      "acc": 0.6608758,
      "epoch": 1.3939624556062913,
      "grad_norm": 5.4375,
      "learning_rate": 2.3071815193355005e-06,
      "loss": 1.55960388,
      "memory(GiB)": 117.38,
      "step": 54950,
      "train_speed(iter/s)": 1.638339
    },
    {
      "acc": 0.66309366,
      "epoch": 1.3940892947742263,
      "grad_norm": 5.625,
      "learning_rate": 2.3062980241898725e-06,
      "loss": 1.5562542,
      "memory(GiB)": 117.38,
      "step": 54955,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.64930382,
      "epoch": 1.3942161339421613,
      "grad_norm": 6.625,
      "learning_rate": 2.3054146475252852e-06,
      "loss": 1.63108883,
      "memory(GiB)": 117.38,
      "step": 54960,
      "train_speed(iter/s)": 1.638374
    },
    {
      "acc": 0.67725663,
      "epoch": 1.3943429731100965,
      "grad_norm": 6.125,
      "learning_rate": 2.3045313893805926e-06,
      "loss": 1.45865288,
      "memory(GiB)": 117.38,
      "step": 54965,
      "train_speed(iter/s)": 1.638393
    },
    {
      "acc": 0.6610034,
      "epoch": 1.3944698122780315,
      "grad_norm": 7.59375,
      "learning_rate": 2.303648249794644e-06,
      "loss": 1.62752399,
      "memory(GiB)": 117.38,
      "step": 54970,
      "train_speed(iter/s)": 1.63841
    },
    {
      "acc": 0.65557089,
      "epoch": 1.3945966514459665,
      "grad_norm": 5.25,
      "learning_rate": 2.302765228806283e-06,
      "loss": 1.57916603,
      "memory(GiB)": 117.38,
      "step": 54975,
      "train_speed(iter/s)": 1.638428
    },
    {
      "acc": 0.65172729,
      "epoch": 1.3947234906139014,
      "grad_norm": 6.4375,
      "learning_rate": 2.30188232645435e-06,
      "loss": 1.6548996,
      "memory(GiB)": 117.38,
      "step": 54980,
      "train_speed(iter/s)": 1.638444
    },
    {
      "acc": 0.64850979,
      "epoch": 1.3948503297818367,
      "grad_norm": 5.84375,
      "learning_rate": 2.300999542777678e-06,
      "loss": 1.67980804,
      "memory(GiB)": 117.38,
      "step": 54985,
      "train_speed(iter/s)": 1.638461
    },
    {
      "acc": 0.6555099,
      "epoch": 1.3949771689497716,
      "grad_norm": 5.6875,
      "learning_rate": 2.300116877815097e-06,
      "loss": 1.53745718,
      "memory(GiB)": 117.38,
      "step": 54990,
      "train_speed(iter/s)": 1.638478
    },
    {
      "acc": 0.64655685,
      "epoch": 1.3951040081177069,
      "grad_norm": 5.78125,
      "learning_rate": 2.2992343316054296e-06,
      "loss": 1.63145065,
      "memory(GiB)": 117.38,
      "step": 54995,
      "train_speed(iter/s)": 1.638496
    },
    {
      "acc": 0.65918884,
      "epoch": 1.3952308472856418,
      "grad_norm": 5.1875,
      "learning_rate": 2.298351904187494e-06,
      "loss": 1.57708597,
      "memory(GiB)": 117.38,
      "step": 55000,
      "train_speed(iter/s)": 1.638513
    },
    {
      "epoch": 1.3952308472856418,
      "eval_acc": 0.6461933312032611,
      "eval_loss": 1.5732694864273071,
      "eval_runtime": 58.4638,
      "eval_samples_per_second": 108.956,
      "eval_steps_per_second": 27.248,
      "step": 55000
    },
    {
      "acc": 0.65558567,
      "epoch": 1.3953576864535768,
      "grad_norm": 5.59375,
      "learning_rate": 2.297469595600104e-06,
      "loss": 1.62075157,
      "memory(GiB)": 117.38,
      "step": 55005,
      "train_speed(iter/s)": 1.635469
    },
    {
      "acc": 0.65070286,
      "epoch": 1.3954845256215118,
      "grad_norm": 5.28125,
      "learning_rate": 2.2965874058820668e-06,
      "loss": 1.64342937,
      "memory(GiB)": 117.38,
      "step": 55010,
      "train_speed(iter/s)": 1.635485
    },
    {
      "acc": 0.65479927,
      "epoch": 1.395611364789447,
      "grad_norm": 7.34375,
      "learning_rate": 2.2957053350721857e-06,
      "loss": 1.60056801,
      "memory(GiB)": 117.38,
      "step": 55015,
      "train_speed(iter/s)": 1.635501
    },
    {
      "acc": 0.65444026,
      "epoch": 1.395738203957382,
      "grad_norm": 6.4375,
      "learning_rate": 2.294823383209258e-06,
      "loss": 1.620327,
      "memory(GiB)": 117.38,
      "step": 55020,
      "train_speed(iter/s)": 1.635516
    },
    {
      "acc": 0.65791359,
      "epoch": 1.3958650431253172,
      "grad_norm": 5.6875,
      "learning_rate": 2.2939415503320733e-06,
      "loss": 1.57491636,
      "memory(GiB)": 117.38,
      "step": 55025,
      "train_speed(iter/s)": 1.635529
    },
    {
      "acc": 0.63341837,
      "epoch": 1.3959918822932522,
      "grad_norm": 5.15625,
      "learning_rate": 2.293059836479425e-06,
      "loss": 1.65767384,
      "memory(GiB)": 117.38,
      "step": 55030,
      "train_speed(iter/s)": 1.635546
    },
    {
      "acc": 0.66097469,
      "epoch": 1.3961187214611872,
      "grad_norm": 5.46875,
      "learning_rate": 2.2921782416900883e-06,
      "loss": 1.55248327,
      "memory(GiB)": 117.38,
      "step": 55035,
      "train_speed(iter/s)": 1.635562
    },
    {
      "acc": 0.66823359,
      "epoch": 1.3962455606291222,
      "grad_norm": 6.59375,
      "learning_rate": 2.2912967660028425e-06,
      "loss": 1.57226105,
      "memory(GiB)": 117.38,
      "step": 55040,
      "train_speed(iter/s)": 1.635578
    },
    {
      "acc": 0.66065459,
      "epoch": 1.3963723997970574,
      "grad_norm": 4.28125,
      "learning_rate": 2.2904154094564568e-06,
      "loss": 1.56496401,
      "memory(GiB)": 117.38,
      "step": 55045,
      "train_speed(iter/s)": 1.635594
    },
    {
      "acc": 0.65135527,
      "epoch": 1.3964992389649924,
      "grad_norm": 6.21875,
      "learning_rate": 2.2895341720897018e-06,
      "loss": 1.6189209,
      "memory(GiB)": 117.38,
      "step": 55050,
      "train_speed(iter/s)": 1.63561
    },
    {
      "acc": 0.64938097,
      "epoch": 1.3966260781329274,
      "grad_norm": 5.90625,
      "learning_rate": 2.2886530539413336e-06,
      "loss": 1.66383324,
      "memory(GiB)": 117.38,
      "step": 55055,
      "train_speed(iter/s)": 1.635627
    },
    {
      "acc": 0.66291885,
      "epoch": 1.3967529173008626,
      "grad_norm": 5.53125,
      "learning_rate": 2.2877720550501082e-06,
      "loss": 1.59094162,
      "memory(GiB)": 117.38,
      "step": 55060,
      "train_speed(iter/s)": 1.635644
    },
    {
      "acc": 0.68015699,
      "epoch": 1.3968797564687976,
      "grad_norm": 6.40625,
      "learning_rate": 2.2868911754547783e-06,
      "loss": 1.5454298,
      "memory(GiB)": 117.38,
      "step": 55065,
      "train_speed(iter/s)": 1.63566
    },
    {
      "acc": 0.65581703,
      "epoch": 1.3970065956367326,
      "grad_norm": 5.5,
      "learning_rate": 2.28601041519409e-06,
      "loss": 1.55169697,
      "memory(GiB)": 117.38,
      "step": 55070,
      "train_speed(iter/s)": 1.635676
    },
    {
      "acc": 0.6483285,
      "epoch": 1.3971334348046676,
      "grad_norm": 6.3125,
      "learning_rate": 2.2851297743067786e-06,
      "loss": 1.6016367,
      "memory(GiB)": 117.38,
      "step": 55075,
      "train_speed(iter/s)": 1.635692
    },
    {
      "acc": 0.66045046,
      "epoch": 1.3972602739726028,
      "grad_norm": 6.0,
      "learning_rate": 2.2842492528315784e-06,
      "loss": 1.52488384,
      "memory(GiB)": 117.38,
      "step": 55080,
      "train_speed(iter/s)": 1.635709
    },
    {
      "acc": 0.66070461,
      "epoch": 1.3973871131405378,
      "grad_norm": 5.8125,
      "learning_rate": 2.283368850807223e-06,
      "loss": 1.57361317,
      "memory(GiB)": 117.38,
      "step": 55085,
      "train_speed(iter/s)": 1.635726
    },
    {
      "acc": 0.66746345,
      "epoch": 1.397513952308473,
      "grad_norm": 5.78125,
      "learning_rate": 2.282488568272437e-06,
      "loss": 1.58817005,
      "memory(GiB)": 117.38,
      "step": 55090,
      "train_speed(iter/s)": 1.635737
    },
    {
      "acc": 0.64651456,
      "epoch": 1.397640791476408,
      "grad_norm": 5.65625,
      "learning_rate": 2.2816084052659326e-06,
      "loss": 1.62980385,
      "memory(GiB)": 117.38,
      "step": 55095,
      "train_speed(iter/s)": 1.635752
    },
    {
      "acc": 0.65791364,
      "epoch": 1.397767630644343,
      "grad_norm": 5.375,
      "learning_rate": 2.2807283618264288e-06,
      "loss": 1.62679749,
      "memory(GiB)": 117.38,
      "step": 55100,
      "train_speed(iter/s)": 1.635769
    },
    {
      "acc": 0.65377564,
      "epoch": 1.397894469812278,
      "grad_norm": 6.28125,
      "learning_rate": 2.2798484379926324e-06,
      "loss": 1.57475939,
      "memory(GiB)": 117.38,
      "step": 55105,
      "train_speed(iter/s)": 1.635786
    },
    {
      "acc": 0.66424837,
      "epoch": 1.3980213089802132,
      "grad_norm": 5.78125,
      "learning_rate": 2.278968633803246e-06,
      "loss": 1.52081919,
      "memory(GiB)": 117.38,
      "step": 55110,
      "train_speed(iter/s)": 1.635802
    },
    {
      "acc": 0.65621176,
      "epoch": 1.3981481481481481,
      "grad_norm": 5.625,
      "learning_rate": 2.2780889492969684e-06,
      "loss": 1.56401024,
      "memory(GiB)": 117.38,
      "step": 55115,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.6706502,
      "epoch": 1.3982749873160831,
      "grad_norm": 6.21875,
      "learning_rate": 2.277209384512491e-06,
      "loss": 1.53983889,
      "memory(GiB)": 117.38,
      "step": 55120,
      "train_speed(iter/s)": 1.635836
    },
    {
      "acc": 0.64606833,
      "epoch": 1.3984018264840183,
      "grad_norm": 8.25,
      "learning_rate": 2.2763299394885013e-06,
      "loss": 1.61826744,
      "memory(GiB)": 117.38,
      "step": 55125,
      "train_speed(iter/s)": 1.635853
    },
    {
      "acc": 0.65655813,
      "epoch": 1.3985286656519533,
      "grad_norm": 5.0,
      "learning_rate": 2.2754506142636808e-06,
      "loss": 1.62069283,
      "memory(GiB)": 117.38,
      "step": 55130,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.66974339,
      "epoch": 1.3986555048198883,
      "grad_norm": 5.0625,
      "learning_rate": 2.274571408876707e-06,
      "loss": 1.5615572,
      "memory(GiB)": 117.38,
      "step": 55135,
      "train_speed(iter/s)": 1.635886
    },
    {
      "acc": 0.66698098,
      "epoch": 1.3987823439878233,
      "grad_norm": 6.53125,
      "learning_rate": 2.2736923233662504e-06,
      "loss": 1.58785343,
      "memory(GiB)": 117.38,
      "step": 55140,
      "train_speed(iter/s)": 1.635903
    },
    {
      "acc": 0.65583334,
      "epoch": 1.3989091831557585,
      "grad_norm": 5.0,
      "learning_rate": 2.2728133577709776e-06,
      "loss": 1.65006905,
      "memory(GiB)": 117.38,
      "step": 55145,
      "train_speed(iter/s)": 1.63592
    },
    {
      "acc": 0.65484171,
      "epoch": 1.3990360223236935,
      "grad_norm": 6.78125,
      "learning_rate": 2.2719345121295495e-06,
      "loss": 1.61099815,
      "memory(GiB)": 117.38,
      "step": 55150,
      "train_speed(iter/s)": 1.635937
    },
    {
      "acc": 0.64947262,
      "epoch": 1.3991628614916287,
      "grad_norm": 5.3125,
      "learning_rate": 2.2710557864806214e-06,
      "loss": 1.67447109,
      "memory(GiB)": 117.38,
      "step": 55155,
      "train_speed(iter/s)": 1.635953
    },
    {
      "acc": 0.6540925,
      "epoch": 1.3992897006595637,
      "grad_norm": 5.59375,
      "learning_rate": 2.2701771808628438e-06,
      "loss": 1.61505032,
      "memory(GiB)": 117.38,
      "step": 55160,
      "train_speed(iter/s)": 1.63597
    },
    {
      "acc": 0.65101833,
      "epoch": 1.3994165398274987,
      "grad_norm": 5.78125,
      "learning_rate": 2.269298695314861e-06,
      "loss": 1.64927082,
      "memory(GiB)": 117.38,
      "step": 55165,
      "train_speed(iter/s)": 1.635986
    },
    {
      "acc": 0.65803304,
      "epoch": 1.3995433789954337,
      "grad_norm": 5.90625,
      "learning_rate": 2.268420329875314e-06,
      "loss": 1.61427803,
      "memory(GiB)": 117.38,
      "step": 55170,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.65885649,
      "epoch": 1.399670218163369,
      "grad_norm": 5.40625,
      "learning_rate": 2.2675420845828363e-06,
      "loss": 1.52747374,
      "memory(GiB)": 117.38,
      "step": 55175,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.65031261,
      "epoch": 1.3997970573313039,
      "grad_norm": 7.46875,
      "learning_rate": 2.266663959476057e-06,
      "loss": 1.58954802,
      "memory(GiB)": 117.38,
      "step": 55180,
      "train_speed(iter/s)": 1.636036
    },
    {
      "acc": 0.66151996,
      "epoch": 1.399923896499239,
      "grad_norm": 6.6875,
      "learning_rate": 2.265785954593598e-06,
      "loss": 1.5725173,
      "memory(GiB)": 117.38,
      "step": 55185,
      "train_speed(iter/s)": 1.636054
    },
    {
      "acc": 0.64516759,
      "epoch": 1.400050735667174,
      "grad_norm": 6.0,
      "learning_rate": 2.264908069974085e-06,
      "loss": 1.59116058,
      "memory(GiB)": 117.38,
      "step": 55190,
      "train_speed(iter/s)": 1.636071
    },
    {
      "acc": 0.66232872,
      "epoch": 1.400177574835109,
      "grad_norm": 6.4375,
      "learning_rate": 2.2640303056561236e-06,
      "loss": 1.54893026,
      "memory(GiB)": 117.38,
      "step": 55195,
      "train_speed(iter/s)": 1.636086
    },
    {
      "acc": 0.64688368,
      "epoch": 1.400304414003044,
      "grad_norm": 7.625,
      "learning_rate": 2.2631526616783234e-06,
      "loss": 1.59646835,
      "memory(GiB)": 117.38,
      "step": 55200,
      "train_speed(iter/s)": 1.636104
    },
    {
      "acc": 0.66893501,
      "epoch": 1.4004312531709793,
      "grad_norm": 5.6875,
      "learning_rate": 2.2622751380792896e-06,
      "loss": 1.55401516,
      "memory(GiB)": 117.38,
      "step": 55205,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.64609971,
      "epoch": 1.4005580923389143,
      "grad_norm": 5.75,
      "learning_rate": 2.26139773489762e-06,
      "loss": 1.64581604,
      "memory(GiB)": 117.38,
      "step": 55210,
      "train_speed(iter/s)": 1.636138
    },
    {
      "acc": 0.66110115,
      "epoch": 1.4006849315068493,
      "grad_norm": 4.84375,
      "learning_rate": 2.260520452171904e-06,
      "loss": 1.56986027,
      "memory(GiB)": 117.38,
      "step": 55215,
      "train_speed(iter/s)": 1.636155
    },
    {
      "acc": 0.65141363,
      "epoch": 1.4008117706747845,
      "grad_norm": 5.4375,
      "learning_rate": 2.259643289940727e-06,
      "loss": 1.57192612,
      "memory(GiB)": 117.38,
      "step": 55220,
      "train_speed(iter/s)": 1.636171
    },
    {
      "acc": 0.66034131,
      "epoch": 1.4009386098427195,
      "grad_norm": 5.59375,
      "learning_rate": 2.2587662482426748e-06,
      "loss": 1.59899263,
      "memory(GiB)": 117.38,
      "step": 55225,
      "train_speed(iter/s)": 1.636188
    },
    {
      "acc": 0.65564446,
      "epoch": 1.4010654490106544,
      "grad_norm": 7.125,
      "learning_rate": 2.2578893271163234e-06,
      "loss": 1.65590668,
      "memory(GiB)": 117.38,
      "step": 55230,
      "train_speed(iter/s)": 1.636204
    },
    {
      "acc": 0.6525218,
      "epoch": 1.4011922881785894,
      "grad_norm": 5.6875,
      "learning_rate": 2.2570125266002385e-06,
      "loss": 1.5934391,
      "memory(GiB)": 117.38,
      "step": 55235,
      "train_speed(iter/s)": 1.636221
    },
    {
      "acc": 0.66293912,
      "epoch": 1.4013191273465246,
      "grad_norm": 6.71875,
      "learning_rate": 2.2561358467329907e-06,
      "loss": 1.59095764,
      "memory(GiB)": 117.38,
      "step": 55240,
      "train_speed(iter/s)": 1.636238
    },
    {
      "acc": 0.66244698,
      "epoch": 1.4014459665144596,
      "grad_norm": 6.375,
      "learning_rate": 2.2552592875531397e-06,
      "loss": 1.59582348,
      "memory(GiB)": 117.38,
      "step": 55245,
      "train_speed(iter/s)": 1.636254
    },
    {
      "acc": 0.66017885,
      "epoch": 1.4015728056823948,
      "grad_norm": 5.5,
      "learning_rate": 2.254382849099241e-06,
      "loss": 1.57824068,
      "memory(GiB)": 117.38,
      "step": 55250,
      "train_speed(iter/s)": 1.63627
    },
    {
      "acc": 0.66618204,
      "epoch": 1.4016996448503298,
      "grad_norm": 6.1875,
      "learning_rate": 2.253506531409839e-06,
      "loss": 1.54128647,
      "memory(GiB)": 117.38,
      "step": 55255,
      "train_speed(iter/s)": 1.636287
    },
    {
      "acc": 0.6507659,
      "epoch": 1.4018264840182648,
      "grad_norm": 6.71875,
      "learning_rate": 2.252630334523484e-06,
      "loss": 1.64957542,
      "memory(GiB)": 117.38,
      "step": 55260,
      "train_speed(iter/s)": 1.636303
    },
    {
      "acc": 0.64942713,
      "epoch": 1.4019533231861998,
      "grad_norm": 6.5,
      "learning_rate": 2.2517542584787134e-06,
      "loss": 1.66868362,
      "memory(GiB)": 117.38,
      "step": 55265,
      "train_speed(iter/s)": 1.636319
    },
    {
      "acc": 0.66312199,
      "epoch": 1.402080162354135,
      "grad_norm": 5.9375,
      "learning_rate": 2.2508783033140596e-06,
      "loss": 1.56169405,
      "memory(GiB)": 117.38,
      "step": 55270,
      "train_speed(iter/s)": 1.636335
    },
    {
      "acc": 0.63883991,
      "epoch": 1.40220700152207,
      "grad_norm": 7.15625,
      "learning_rate": 2.2500024690680528e-06,
      "loss": 1.68168755,
      "memory(GiB)": 117.38,
      "step": 55275,
      "train_speed(iter/s)": 1.636353
    },
    {
      "acc": 0.66981516,
      "epoch": 1.402333840690005,
      "grad_norm": 5.40625,
      "learning_rate": 2.249126755779215e-06,
      "loss": 1.51545362,
      "memory(GiB)": 117.38,
      "step": 55280,
      "train_speed(iter/s)": 1.636369
    },
    {
      "acc": 0.65273981,
      "epoch": 1.4024606798579402,
      "grad_norm": 5.40625,
      "learning_rate": 2.2482511634860645e-06,
      "loss": 1.5833827,
      "memory(GiB)": 117.38,
      "step": 55285,
      "train_speed(iter/s)": 1.636385
    },
    {
      "acc": 0.64515276,
      "epoch": 1.4025875190258752,
      "grad_norm": 6.75,
      "learning_rate": 2.247375692227113e-06,
      "loss": 1.6318943,
      "memory(GiB)": 117.38,
      "step": 55290,
      "train_speed(iter/s)": 1.636402
    },
    {
      "acc": 0.6652565,
      "epoch": 1.4027143581938102,
      "grad_norm": 6.375,
      "learning_rate": 2.2465003420408683e-06,
      "loss": 1.60801868,
      "memory(GiB)": 117.38,
      "step": 55295,
      "train_speed(iter/s)": 1.636419
    },
    {
      "acc": 0.65893621,
      "epoch": 1.4028411973617452,
      "grad_norm": 5.65625,
      "learning_rate": 2.2456251129658325e-06,
      "loss": 1.60439358,
      "memory(GiB)": 117.38,
      "step": 55300,
      "train_speed(iter/s)": 1.636434
    },
    {
      "acc": 0.64047923,
      "epoch": 1.4029680365296804,
      "grad_norm": 5.78125,
      "learning_rate": 2.2447500050405008e-06,
      "loss": 1.64773216,
      "memory(GiB)": 117.38,
      "step": 55305,
      "train_speed(iter/s)": 1.63645
    },
    {
      "acc": 0.66279316,
      "epoch": 1.4030948756976154,
      "grad_norm": 9.9375,
      "learning_rate": 2.2438750183033657e-06,
      "loss": 1.60474739,
      "memory(GiB)": 117.38,
      "step": 55310,
      "train_speed(iter/s)": 1.636468
    },
    {
      "acc": 0.66279058,
      "epoch": 1.4032217148655506,
      "grad_norm": 5.5,
      "learning_rate": 2.2430001527929123e-06,
      "loss": 1.59033442,
      "memory(GiB)": 117.38,
      "step": 55315,
      "train_speed(iter/s)": 1.636486
    },
    {
      "acc": 0.65323105,
      "epoch": 1.4033485540334856,
      "grad_norm": 7.0,
      "learning_rate": 2.242125408547622e-06,
      "loss": 1.65653687,
      "memory(GiB)": 117.38,
      "step": 55320,
      "train_speed(iter/s)": 1.636503
    },
    {
      "acc": 0.65559368,
      "epoch": 1.4034753932014206,
      "grad_norm": 7.5,
      "learning_rate": 2.2412507856059667e-06,
      "loss": 1.64404964,
      "memory(GiB)": 117.38,
      "step": 55325,
      "train_speed(iter/s)": 1.636521
    },
    {
      "acc": 0.65261588,
      "epoch": 1.4036022323693556,
      "grad_norm": 5.65625,
      "learning_rate": 2.2403762840064223e-06,
      "loss": 1.58226433,
      "memory(GiB)": 117.38,
      "step": 55330,
      "train_speed(iter/s)": 1.636538
    },
    {
      "acc": 0.67162457,
      "epoch": 1.4037290715372908,
      "grad_norm": 4.8125,
      "learning_rate": 2.239501903787448e-06,
      "loss": 1.57336235,
      "memory(GiB)": 117.38,
      "step": 55335,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.65162492,
      "epoch": 1.4038559107052258,
      "grad_norm": 7.375,
      "learning_rate": 2.2386276449875057e-06,
      "loss": 1.62444992,
      "memory(GiB)": 117.38,
      "step": 55340,
      "train_speed(iter/s)": 1.63657
    },
    {
      "acc": 0.64949026,
      "epoch": 1.403982749873161,
      "grad_norm": 6.78125,
      "learning_rate": 2.2377535076450452e-06,
      "loss": 1.61994133,
      "memory(GiB)": 117.38,
      "step": 55345,
      "train_speed(iter/s)": 1.636586
    },
    {
      "acc": 0.65781751,
      "epoch": 1.404109589041096,
      "grad_norm": 6.0,
      "learning_rate": 2.236879491798522e-06,
      "loss": 1.5575325,
      "memory(GiB)": 117.38,
      "step": 55350,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.66194191,
      "epoch": 1.404236428209031,
      "grad_norm": 6.375,
      "learning_rate": 2.2360055974863737e-06,
      "loss": 1.61743317,
      "memory(GiB)": 117.38,
      "step": 55355,
      "train_speed(iter/s)": 1.63662
    },
    {
      "acc": 0.66451869,
      "epoch": 1.404363267376966,
      "grad_norm": 5.28125,
      "learning_rate": 2.2351318247470376e-06,
      "loss": 1.56740646,
      "memory(GiB)": 117.38,
      "step": 55360,
      "train_speed(iter/s)": 1.636636
    },
    {
      "acc": 0.64429374,
      "epoch": 1.4044901065449011,
      "grad_norm": 5.0625,
      "learning_rate": 2.2342581736189496e-06,
      "loss": 1.62785683,
      "memory(GiB)": 117.38,
      "step": 55365,
      "train_speed(iter/s)": 1.636651
    },
    {
      "acc": 0.65563703,
      "epoch": 1.4046169457128361,
      "grad_norm": 5.125,
      "learning_rate": 2.233384644140537e-06,
      "loss": 1.59156523,
      "memory(GiB)": 117.38,
      "step": 55370,
      "train_speed(iter/s)": 1.636667
    },
    {
      "acc": 0.67236867,
      "epoch": 1.4047437848807711,
      "grad_norm": 7.1875,
      "learning_rate": 2.2325112363502167e-06,
      "loss": 1.51769981,
      "memory(GiB)": 117.38,
      "step": 55375,
      "train_speed(iter/s)": 1.636683
    },
    {
      "acc": 0.66594553,
      "epoch": 1.4048706240487063,
      "grad_norm": 5.78125,
      "learning_rate": 2.23163795028641e-06,
      "loss": 1.54175014,
      "memory(GiB)": 117.38,
      "step": 55380,
      "train_speed(iter/s)": 1.636697
    },
    {
      "acc": 0.65792103,
      "epoch": 1.4049974632166413,
      "grad_norm": 5.21875,
      "learning_rate": 2.230764785987526e-06,
      "loss": 1.56255751,
      "memory(GiB)": 117.38,
      "step": 55385,
      "train_speed(iter/s)": 1.636714
    },
    {
      "acc": 0.65201612,
      "epoch": 1.4051243023845763,
      "grad_norm": 6.90625,
      "learning_rate": 2.229891743491972e-06,
      "loss": 1.54268131,
      "memory(GiB)": 117.38,
      "step": 55390,
      "train_speed(iter/s)": 1.63673
    },
    {
      "acc": 0.65024819,
      "epoch": 1.4052511415525113,
      "grad_norm": 5.28125,
      "learning_rate": 2.2290188228381434e-06,
      "loss": 1.58718824,
      "memory(GiB)": 117.38,
      "step": 55395,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.66040421,
      "epoch": 1.4053779807204465,
      "grad_norm": 5.0625,
      "learning_rate": 2.2281460240644397e-06,
      "loss": 1.59405689,
      "memory(GiB)": 117.38,
      "step": 55400,
      "train_speed(iter/s)": 1.636761
    },
    {
      "acc": 0.64602785,
      "epoch": 1.4055048198883815,
      "grad_norm": 5.6875,
      "learning_rate": 2.22727334720925e-06,
      "loss": 1.61734486,
      "memory(GiB)": 117.38,
      "step": 55405,
      "train_speed(iter/s)": 1.636778
    },
    {
      "acc": 0.66305494,
      "epoch": 1.4056316590563167,
      "grad_norm": 5.25,
      "learning_rate": 2.2264007923109575e-06,
      "loss": 1.57885513,
      "memory(GiB)": 117.38,
      "step": 55410,
      "train_speed(iter/s)": 1.636792
    },
    {
      "acc": 0.66346731,
      "epoch": 1.4057584982242517,
      "grad_norm": 6.8125,
      "learning_rate": 2.225528359407942e-06,
      "loss": 1.60510044,
      "memory(GiB)": 117.38,
      "step": 55415,
      "train_speed(iter/s)": 1.636808
    },
    {
      "acc": 0.65615096,
      "epoch": 1.4058853373921867,
      "grad_norm": 5.96875,
      "learning_rate": 2.2246560485385756e-06,
      "loss": 1.56561718,
      "memory(GiB)": 117.38,
      "step": 55420,
      "train_speed(iter/s)": 1.636824
    },
    {
      "acc": 0.65666571,
      "epoch": 1.4060121765601217,
      "grad_norm": 5.28125,
      "learning_rate": 2.2237838597412277e-06,
      "loss": 1.59757919,
      "memory(GiB)": 117.38,
      "step": 55425,
      "train_speed(iter/s)": 1.636841
    },
    {
      "acc": 0.66810236,
      "epoch": 1.4061390157280569,
      "grad_norm": 5.40625,
      "learning_rate": 2.22291179305426e-06,
      "loss": 1.51167374,
      "memory(GiB)": 117.38,
      "step": 55430,
      "train_speed(iter/s)": 1.636857
    },
    {
      "acc": 0.66811314,
      "epoch": 1.4062658548959919,
      "grad_norm": 5.6875,
      "learning_rate": 2.222039848516031e-06,
      "loss": 1.56607914,
      "memory(GiB)": 117.38,
      "step": 55435,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.65201788,
      "epoch": 1.4063926940639269,
      "grad_norm": 6.03125,
      "learning_rate": 2.2211680261648918e-06,
      "loss": 1.60019817,
      "memory(GiB)": 117.38,
      "step": 55440,
      "train_speed(iter/s)": 1.63689
    },
    {
      "acc": 0.65698485,
      "epoch": 1.406519533231862,
      "grad_norm": 5.40625,
      "learning_rate": 2.22029632603919e-06,
      "loss": 1.59546318,
      "memory(GiB)": 117.38,
      "step": 55445,
      "train_speed(iter/s)": 1.636906
    },
    {
      "acc": 0.65754786,
      "epoch": 1.406646372399797,
      "grad_norm": 6.125,
      "learning_rate": 2.2194247481772652e-06,
      "loss": 1.59910812,
      "memory(GiB)": 117.38,
      "step": 55450,
      "train_speed(iter/s)": 1.636924
    },
    {
      "acc": 0.65745711,
      "epoch": 1.406773211567732,
      "grad_norm": 5.5625,
      "learning_rate": 2.218553292617455e-06,
      "loss": 1.6074646,
      "memory(GiB)": 117.38,
      "step": 55455,
      "train_speed(iter/s)": 1.636941
    },
    {
      "acc": 0.65728054,
      "epoch": 1.406900050735667,
      "grad_norm": 5.125,
      "learning_rate": 2.2176819593980892e-06,
      "loss": 1.55498753,
      "memory(GiB)": 117.38,
      "step": 55460,
      "train_speed(iter/s)": 1.636957
    },
    {
      "acc": 0.65538864,
      "epoch": 1.4070268899036023,
      "grad_norm": 7.125,
      "learning_rate": 2.2168107485574914e-06,
      "loss": 1.68885326,
      "memory(GiB)": 117.38,
      "step": 55465,
      "train_speed(iter/s)": 1.636973
    },
    {
      "acc": 0.64546318,
      "epoch": 1.4071537290715372,
      "grad_norm": 5.875,
      "learning_rate": 2.215939660133986e-06,
      "loss": 1.5962822,
      "memory(GiB)": 117.38,
      "step": 55470,
      "train_speed(iter/s)": 1.63699
    },
    {
      "acc": 0.66275277,
      "epoch": 1.4072805682394725,
      "grad_norm": 5.6875,
      "learning_rate": 2.215068694165883e-06,
      "loss": 1.60228329,
      "memory(GiB)": 117.38,
      "step": 55475,
      "train_speed(iter/s)": 1.637007
    },
    {
      "acc": 0.66169815,
      "epoch": 1.4074074074074074,
      "grad_norm": 5.15625,
      "learning_rate": 2.2141978506914922e-06,
      "loss": 1.50000324,
      "memory(GiB)": 117.38,
      "step": 55480,
      "train_speed(iter/s)": 1.637024
    },
    {
      "acc": 0.63967075,
      "epoch": 1.4075342465753424,
      "grad_norm": 5.3125,
      "learning_rate": 2.2133271297491165e-06,
      "loss": 1.69420967,
      "memory(GiB)": 117.38,
      "step": 55485,
      "train_speed(iter/s)": 1.637041
    },
    {
      "acc": 0.66241055,
      "epoch": 1.4076610857432774,
      "grad_norm": 5.625,
      "learning_rate": 2.2124565313770584e-06,
      "loss": 1.56081228,
      "memory(GiB)": 117.38,
      "step": 55490,
      "train_speed(iter/s)": 1.637056
    },
    {
      "acc": 0.66528988,
      "epoch": 1.4077879249112126,
      "grad_norm": 5.09375,
      "learning_rate": 2.211586055613606e-06,
      "loss": 1.57082357,
      "memory(GiB)": 117.38,
      "step": 55495,
      "train_speed(iter/s)": 1.637072
    },
    {
      "acc": 0.65307875,
      "epoch": 1.4079147640791476,
      "grad_norm": 5.09375,
      "learning_rate": 2.210715702497046e-06,
      "loss": 1.58034525,
      "memory(GiB)": 117.38,
      "step": 55500,
      "train_speed(iter/s)": 1.63709
    },
    {
      "acc": 0.66315298,
      "epoch": 1.4080416032470828,
      "grad_norm": 6.0,
      "learning_rate": 2.2098454720656647e-06,
      "loss": 1.6256485,
      "memory(GiB)": 117.38,
      "step": 55505,
      "train_speed(iter/s)": 1.637106
    },
    {
      "acc": 0.66349649,
      "epoch": 1.4081684424150178,
      "grad_norm": 7.0,
      "learning_rate": 2.2089753643577384e-06,
      "loss": 1.55409184,
      "memory(GiB)": 117.38,
      "step": 55510,
      "train_speed(iter/s)": 1.637123
    },
    {
      "acc": 0.66498456,
      "epoch": 1.4082952815829528,
      "grad_norm": 5.3125,
      "learning_rate": 2.208105379411535e-06,
      "loss": 1.56644669,
      "memory(GiB)": 117.38,
      "step": 55515,
      "train_speed(iter/s)": 1.637141
    },
    {
      "acc": 0.65405598,
      "epoch": 1.4084221207508878,
      "grad_norm": 5.78125,
      "learning_rate": 2.2072355172653197e-06,
      "loss": 1.56064854,
      "memory(GiB)": 117.38,
      "step": 55520,
      "train_speed(iter/s)": 1.637158
    },
    {
      "acc": 0.65392909,
      "epoch": 1.408548959918823,
      "grad_norm": 6.03125,
      "learning_rate": 2.2063657779573573e-06,
      "loss": 1.56219311,
      "memory(GiB)": 117.38,
      "step": 55525,
      "train_speed(iter/s)": 1.637175
    },
    {
      "acc": 0.66338797,
      "epoch": 1.408675799086758,
      "grad_norm": 5.59375,
      "learning_rate": 2.2054961615259023e-06,
      "loss": 1.57238331,
      "memory(GiB)": 117.38,
      "step": 55530,
      "train_speed(iter/s)": 1.637192
    },
    {
      "acc": 0.67478223,
      "epoch": 1.408802638254693,
      "grad_norm": 4.71875,
      "learning_rate": 2.2046266680091994e-06,
      "loss": 1.52098751,
      "memory(GiB)": 117.38,
      "step": 55535,
      "train_speed(iter/s)": 1.63721
    },
    {
      "acc": 0.64974861,
      "epoch": 1.4089294774226282,
      "grad_norm": 5.46875,
      "learning_rate": 2.2037572974454974e-06,
      "loss": 1.69125538,
      "memory(GiB)": 117.38,
      "step": 55540,
      "train_speed(iter/s)": 1.637227
    },
    {
      "acc": 0.66572843,
      "epoch": 1.4090563165905632,
      "grad_norm": 6.8125,
      "learning_rate": 2.202888049873034e-06,
      "loss": 1.58676138,
      "memory(GiB)": 117.38,
      "step": 55545,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.65373125,
      "epoch": 1.4091831557584982,
      "grad_norm": 6.0,
      "learning_rate": 2.2020189253300428e-06,
      "loss": 1.71913567,
      "memory(GiB)": 117.38,
      "step": 55550,
      "train_speed(iter/s)": 1.63726
    },
    {
      "acc": 0.6474061,
      "epoch": 1.4093099949264332,
      "grad_norm": 5.46875,
      "learning_rate": 2.2011499238547506e-06,
      "loss": 1.66344872,
      "memory(GiB)": 117.38,
      "step": 55555,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.64668779,
      "epoch": 1.4094368340943684,
      "grad_norm": 7.125,
      "learning_rate": 2.2002810454853813e-06,
      "loss": 1.646521,
      "memory(GiB)": 117.38,
      "step": 55560,
      "train_speed(iter/s)": 1.637296
    },
    {
      "acc": 0.65319386,
      "epoch": 1.4095636732623034,
      "grad_norm": 6.46875,
      "learning_rate": 2.1994122902601513e-06,
      "loss": 1.62287388,
      "memory(GiB)": 117.38,
      "step": 55565,
      "train_speed(iter/s)": 1.637313
    },
    {
      "acc": 0.65166717,
      "epoch": 1.4096905124302386,
      "grad_norm": 5.65625,
      "learning_rate": 2.1985436582172724e-06,
      "loss": 1.58232927,
      "memory(GiB)": 117.38,
      "step": 55570,
      "train_speed(iter/s)": 1.637329
    },
    {
      "acc": 0.65493631,
      "epoch": 1.4098173515981736,
      "grad_norm": 5.65625,
      "learning_rate": 2.1976751493949512e-06,
      "loss": 1.64372635,
      "memory(GiB)": 117.38,
      "step": 55575,
      "train_speed(iter/s)": 1.637345
    },
    {
      "acc": 0.65079679,
      "epoch": 1.4099441907661086,
      "grad_norm": 5.9375,
      "learning_rate": 2.196806763831388e-06,
      "loss": 1.59450188,
      "memory(GiB)": 117.38,
      "step": 55580,
      "train_speed(iter/s)": 1.637362
    },
    {
      "acc": 0.66014271,
      "epoch": 1.4100710299340435,
      "grad_norm": 5.875,
      "learning_rate": 2.1959385015647794e-06,
      "loss": 1.65788307,
      "memory(GiB)": 117.38,
      "step": 55585,
      "train_speed(iter/s)": 1.637379
    },
    {
      "acc": 0.65923615,
      "epoch": 1.4101978691019788,
      "grad_norm": 5.1875,
      "learning_rate": 2.195070362633314e-06,
      "loss": 1.6290554,
      "memory(GiB)": 117.38,
      "step": 55590,
      "train_speed(iter/s)": 1.637396
    },
    {
      "acc": 0.65366058,
      "epoch": 1.4103247082699137,
      "grad_norm": 7.75,
      "learning_rate": 2.194202347075178e-06,
      "loss": 1.62625923,
      "memory(GiB)": 117.38,
      "step": 55595,
      "train_speed(iter/s)": 1.637414
    },
    {
      "acc": 0.6517252,
      "epoch": 1.4104515474378487,
      "grad_norm": 5.75,
      "learning_rate": 2.1933344549285485e-06,
      "loss": 1.64228897,
      "memory(GiB)": 117.38,
      "step": 55600,
      "train_speed(iter/s)": 1.63743
    },
    {
      "acc": 0.65299754,
      "epoch": 1.410578386605784,
      "grad_norm": 9.0,
      "learning_rate": 2.1924666862316015e-06,
      "loss": 1.62476692,
      "memory(GiB)": 117.38,
      "step": 55605,
      "train_speed(iter/s)": 1.637448
    },
    {
      "acc": 0.66371651,
      "epoch": 1.410705225773719,
      "grad_norm": 5.9375,
      "learning_rate": 2.191599041022504e-06,
      "loss": 1.564042,
      "memory(GiB)": 117.38,
      "step": 55610,
      "train_speed(iter/s)": 1.637465
    },
    {
      "acc": 0.66063395,
      "epoch": 1.410832064941654,
      "grad_norm": 5.1875,
      "learning_rate": 2.19073151933942e-06,
      "loss": 1.53701601,
      "memory(GiB)": 117.38,
      "step": 55615,
      "train_speed(iter/s)": 1.637481
    },
    {
      "acc": 0.65441561,
      "epoch": 1.410958904109589,
      "grad_norm": 5.625,
      "learning_rate": 2.1898641212205053e-06,
      "loss": 1.59882078,
      "memory(GiB)": 117.38,
      "step": 55620,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.64533687,
      "epoch": 1.4110857432775241,
      "grad_norm": 5.28125,
      "learning_rate": 2.1889968467039114e-06,
      "loss": 1.60900536,
      "memory(GiB)": 117.38,
      "step": 55625,
      "train_speed(iter/s)": 1.637516
    },
    {
      "acc": 0.65276604,
      "epoch": 1.411212582445459,
      "grad_norm": 5.53125,
      "learning_rate": 2.1881296958277897e-06,
      "loss": 1.59919558,
      "memory(GiB)": 117.38,
      "step": 55630,
      "train_speed(iter/s)": 1.637532
    },
    {
      "acc": 0.67059474,
      "epoch": 1.4113394216133943,
      "grad_norm": 6.21875,
      "learning_rate": 2.1872626686302767e-06,
      "loss": 1.54859467,
      "memory(GiB)": 117.38,
      "step": 55635,
      "train_speed(iter/s)": 1.637549
    },
    {
      "acc": 0.6617178,
      "epoch": 1.4114662607813293,
      "grad_norm": 5.9375,
      "learning_rate": 2.186395765149508e-06,
      "loss": 1.50149059,
      "memory(GiB)": 117.38,
      "step": 55640,
      "train_speed(iter/s)": 1.637566
    },
    {
      "acc": 0.6456203,
      "epoch": 1.4115930999492643,
      "grad_norm": 5.3125,
      "learning_rate": 2.1855289854236165e-06,
      "loss": 1.66820869,
      "memory(GiB)": 117.38,
      "step": 55645,
      "train_speed(iter/s)": 1.637582
    },
    {
      "acc": 0.65637083,
      "epoch": 1.4117199391171993,
      "grad_norm": 5.0625,
      "learning_rate": 2.184662329490728e-06,
      "loss": 1.58305645,
      "memory(GiB)": 117.38,
      "step": 55650,
      "train_speed(iter/s)": 1.637598
    },
    {
      "acc": 0.65677662,
      "epoch": 1.4118467782851345,
      "grad_norm": 6.75,
      "learning_rate": 2.1837957973889584e-06,
      "loss": 1.60450344,
      "memory(GiB)": 117.38,
      "step": 55655,
      "train_speed(iter/s)": 1.637614
    },
    {
      "acc": 0.63853054,
      "epoch": 1.4119736174530695,
      "grad_norm": 6.4375,
      "learning_rate": 2.1829293891564212e-06,
      "loss": 1.64831276,
      "memory(GiB)": 117.38,
      "step": 55660,
      "train_speed(iter/s)": 1.637631
    },
    {
      "acc": 0.6536479,
      "epoch": 1.4121004566210047,
      "grad_norm": 5.4375,
      "learning_rate": 2.182063104831228e-06,
      "loss": 1.60060005,
      "memory(GiB)": 117.38,
      "step": 55665,
      "train_speed(iter/s)": 1.637648
    },
    {
      "acc": 0.65629382,
      "epoch": 1.4122272957889397,
      "grad_norm": 5.21875,
      "learning_rate": 2.181196944451483e-06,
      "loss": 1.59252434,
      "memory(GiB)": 117.38,
      "step": 55670,
      "train_speed(iter/s)": 1.637664
    },
    {
      "acc": 0.65572109,
      "epoch": 1.4123541349568747,
      "grad_norm": 6.40625,
      "learning_rate": 2.180330908055278e-06,
      "loss": 1.65411358,
      "memory(GiB)": 117.38,
      "step": 55675,
      "train_speed(iter/s)": 1.63768
    },
    {
      "acc": 0.64828358,
      "epoch": 1.4124809741248097,
      "grad_norm": 6.3125,
      "learning_rate": 2.1794649956807094e-06,
      "loss": 1.61419334,
      "memory(GiB)": 117.38,
      "step": 55680,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.66647234,
      "epoch": 1.4126078132927449,
      "grad_norm": 5.65625,
      "learning_rate": 2.178599207365864e-06,
      "loss": 1.50949287,
      "memory(GiB)": 117.38,
      "step": 55685,
      "train_speed(iter/s)": 1.637714
    },
    {
      "acc": 0.66886826,
      "epoch": 1.4127346524606799,
      "grad_norm": 5.1875,
      "learning_rate": 2.1777335431488234e-06,
      "loss": 1.55648022,
      "memory(GiB)": 117.38,
      "step": 55690,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.66577911,
      "epoch": 1.4128614916286149,
      "grad_norm": 5.875,
      "learning_rate": 2.176868003067659e-06,
      "loss": 1.57494259,
      "memory(GiB)": 117.38,
      "step": 55695,
      "train_speed(iter/s)": 1.637748
    },
    {
      "acc": 0.65630693,
      "epoch": 1.41298833079655,
      "grad_norm": 5.96875,
      "learning_rate": 2.1760025871604456e-06,
      "loss": 1.59197769,
      "memory(GiB)": 117.38,
      "step": 55700,
      "train_speed(iter/s)": 1.637766
    },
    {
      "acc": 0.66285625,
      "epoch": 1.413115169964485,
      "grad_norm": 6.90625,
      "learning_rate": 2.175137295465247e-06,
      "loss": 1.60119648,
      "memory(GiB)": 117.38,
      "step": 55705,
      "train_speed(iter/s)": 1.637782
    },
    {
      "acc": 0.65775256,
      "epoch": 1.41324200913242,
      "grad_norm": 4.71875,
      "learning_rate": 2.174272128020122e-06,
      "loss": 1.62330437,
      "memory(GiB)": 117.38,
      "step": 55710,
      "train_speed(iter/s)": 1.637799
    },
    {
      "acc": 0.6739552,
      "epoch": 1.413368848300355,
      "grad_norm": 6.0625,
      "learning_rate": 2.1734070848631245e-06,
      "loss": 1.56818304,
      "memory(GiB)": 117.38,
      "step": 55715,
      "train_speed(iter/s)": 1.637815
    },
    {
      "acc": 0.66088099,
      "epoch": 1.4134956874682902,
      "grad_norm": 10.4375,
      "learning_rate": 2.172542166032303e-06,
      "loss": 1.54413128,
      "memory(GiB)": 117.38,
      "step": 55720,
      "train_speed(iter/s)": 1.637833
    },
    {
      "acc": 0.65214715,
      "epoch": 1.4136225266362252,
      "grad_norm": 5.3125,
      "learning_rate": 2.171677371565701e-06,
      "loss": 1.6402565,
      "memory(GiB)": 117.38,
      "step": 55725,
      "train_speed(iter/s)": 1.637848
    },
    {
      "acc": 0.66133156,
      "epoch": 1.4137493658041604,
      "grad_norm": 6.3125,
      "learning_rate": 2.1708127015013565e-06,
      "loss": 1.61832027,
      "memory(GiB)": 117.38,
      "step": 55730,
      "train_speed(iter/s)": 1.637864
    },
    {
      "acc": 0.64802055,
      "epoch": 1.4138762049720954,
      "grad_norm": 5.84375,
      "learning_rate": 2.1699481558773e-06,
      "loss": 1.58707027,
      "memory(GiB)": 117.38,
      "step": 55735,
      "train_speed(iter/s)": 1.63788
    },
    {
      "acc": 0.6499661,
      "epoch": 1.4140030441400304,
      "grad_norm": 4.4375,
      "learning_rate": 2.1690837347315597e-06,
      "loss": 1.6137928,
      "memory(GiB)": 117.38,
      "step": 55740,
      "train_speed(iter/s)": 1.637897
    },
    {
      "acc": 0.66404982,
      "epoch": 1.4141298833079654,
      "grad_norm": 6.34375,
      "learning_rate": 2.168219438102155e-06,
      "loss": 1.57214193,
      "memory(GiB)": 117.38,
      "step": 55745,
      "train_speed(iter/s)": 1.637913
    },
    {
      "acc": 0.65018663,
      "epoch": 1.4142567224759006,
      "grad_norm": 6.84375,
      "learning_rate": 2.167355266027103e-06,
      "loss": 1.60913467,
      "memory(GiB)": 117.38,
      "step": 55750,
      "train_speed(iter/s)": 1.63793
    },
    {
      "acc": 0.66313162,
      "epoch": 1.4143835616438356,
      "grad_norm": 6.125,
      "learning_rate": 2.1664912185444127e-06,
      "loss": 1.5276576,
      "memory(GiB)": 117.38,
      "step": 55755,
      "train_speed(iter/s)": 1.637948
    },
    {
      "acc": 0.65130501,
      "epoch": 1.4145104008117706,
      "grad_norm": 5.65625,
      "learning_rate": 2.16562729569209e-06,
      "loss": 1.69024944,
      "memory(GiB)": 117.38,
      "step": 55760,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.66500769,
      "epoch": 1.4146372399797058,
      "grad_norm": 5.84375,
      "learning_rate": 2.164763497508131e-06,
      "loss": 1.54899521,
      "memory(GiB)": 117.38,
      "step": 55765,
      "train_speed(iter/s)": 1.63798
    },
    {
      "acc": 0.67111855,
      "epoch": 1.4147640791476408,
      "grad_norm": 5.625,
      "learning_rate": 2.1638998240305355e-06,
      "loss": 1.57638149,
      "memory(GiB)": 117.38,
      "step": 55770,
      "train_speed(iter/s)": 1.637998
    },
    {
      "acc": 0.6395658,
      "epoch": 1.4148909183155758,
      "grad_norm": 6.125,
      "learning_rate": 2.163036275297286e-06,
      "loss": 1.65650291,
      "memory(GiB)": 117.38,
      "step": 55775,
      "train_speed(iter/s)": 1.638014
    },
    {
      "acc": 0.65556026,
      "epoch": 1.4150177574835108,
      "grad_norm": 5.53125,
      "learning_rate": 2.162172851346368e-06,
      "loss": 1.57244339,
      "memory(GiB)": 117.38,
      "step": 55780,
      "train_speed(iter/s)": 1.63803
    },
    {
      "acc": 0.66797991,
      "epoch": 1.415144596651446,
      "grad_norm": 6.28125,
      "learning_rate": 2.1613095522157557e-06,
      "loss": 1.57551298,
      "memory(GiB)": 117.38,
      "step": 55785,
      "train_speed(iter/s)": 1.638047
    },
    {
      "acc": 0.65960522,
      "epoch": 1.415271435819381,
      "grad_norm": 5.40625,
      "learning_rate": 2.1604463779434267e-06,
      "loss": 1.56033268,
      "memory(GiB)": 117.38,
      "step": 55790,
      "train_speed(iter/s)": 1.638063
    },
    {
      "acc": 0.65391316,
      "epoch": 1.4153982749873162,
      "grad_norm": 5.78125,
      "learning_rate": 2.159583328567342e-06,
      "loss": 1.63294563,
      "memory(GiB)": 117.38,
      "step": 55795,
      "train_speed(iter/s)": 1.63808
    },
    {
      "acc": 0.66140633,
      "epoch": 1.4155251141552512,
      "grad_norm": 5.65625,
      "learning_rate": 2.158720404125462e-06,
      "loss": 1.55224876,
      "memory(GiB)": 117.38,
      "step": 55800,
      "train_speed(iter/s)": 1.638096
    },
    {
      "acc": 0.66071339,
      "epoch": 1.4156519533231862,
      "grad_norm": 5.4375,
      "learning_rate": 2.1578576046557463e-06,
      "loss": 1.57883329,
      "memory(GiB)": 117.38,
      "step": 55805,
      "train_speed(iter/s)": 1.638113
    },
    {
      "acc": 0.64482632,
      "epoch": 1.4157787924911212,
      "grad_norm": 5.8125,
      "learning_rate": 2.156994930196144e-06,
      "loss": 1.6230217,
      "memory(GiB)": 117.38,
      "step": 55810,
      "train_speed(iter/s)": 1.63813
    },
    {
      "acc": 0.66490197,
      "epoch": 1.4159056316590564,
      "grad_norm": 5.5625,
      "learning_rate": 2.156132380784594e-06,
      "loss": 1.53020048,
      "memory(GiB)": 117.38,
      "step": 55815,
      "train_speed(iter/s)": 1.638146
    },
    {
      "acc": 0.64963198,
      "epoch": 1.4160324708269914,
      "grad_norm": 7.40625,
      "learning_rate": 2.155269956459041e-06,
      "loss": 1.61162949,
      "memory(GiB)": 117.38,
      "step": 55820,
      "train_speed(iter/s)": 1.638163
    },
    {
      "acc": 0.67074494,
      "epoch": 1.4161593099949266,
      "grad_norm": 5.78125,
      "learning_rate": 2.1544076572574156e-06,
      "loss": 1.57712784,
      "memory(GiB)": 117.38,
      "step": 55825,
      "train_speed(iter/s)": 1.638179
    },
    {
      "acc": 0.65239029,
      "epoch": 1.4162861491628616,
      "grad_norm": 5.15625,
      "learning_rate": 2.1535454832176482e-06,
      "loss": 1.59608822,
      "memory(GiB)": 117.38,
      "step": 55830,
      "train_speed(iter/s)": 1.638195
    },
    {
      "acc": 0.64519591,
      "epoch": 1.4164129883307965,
      "grad_norm": 6.03125,
      "learning_rate": 2.1526834343776556e-06,
      "loss": 1.67440529,
      "memory(GiB)": 117.38,
      "step": 55835,
      "train_speed(iter/s)": 1.638213
    },
    {
      "acc": 0.6316041,
      "epoch": 1.4165398274987315,
      "grad_norm": 7.4375,
      "learning_rate": 2.1518215107753593e-06,
      "loss": 1.67723885,
      "memory(GiB)": 117.38,
      "step": 55840,
      "train_speed(iter/s)": 1.63823
    },
    {
      "acc": 0.63944454,
      "epoch": 1.4166666666666667,
      "grad_norm": 5.46875,
      "learning_rate": 2.1509597124486693e-06,
      "loss": 1.62081108,
      "memory(GiB)": 117.38,
      "step": 55845,
      "train_speed(iter/s)": 1.638247
    },
    {
      "acc": 0.66900139,
      "epoch": 1.4167935058346017,
      "grad_norm": 5.0625,
      "learning_rate": 2.1500980394354907e-06,
      "loss": 1.5674037,
      "memory(GiB)": 117.38,
      "step": 55850,
      "train_speed(iter/s)": 1.638263
    },
    {
      "acc": 0.6554409,
      "epoch": 1.4169203450025367,
      "grad_norm": 5.78125,
      "learning_rate": 2.1492364917737252e-06,
      "loss": 1.60124626,
      "memory(GiB)": 117.38,
      "step": 55855,
      "train_speed(iter/s)": 1.63828
    },
    {
      "acc": 0.66146145,
      "epoch": 1.417047184170472,
      "grad_norm": 5.71875,
      "learning_rate": 2.148375069501266e-06,
      "loss": 1.56895208,
      "memory(GiB)": 117.38,
      "step": 55860,
      "train_speed(iter/s)": 1.638296
    },
    {
      "acc": 0.65574379,
      "epoch": 1.417174023338407,
      "grad_norm": 4.46875,
      "learning_rate": 2.147513772656003e-06,
      "loss": 1.57744589,
      "memory(GiB)": 117.38,
      "step": 55865,
      "train_speed(iter/s)": 1.638313
    },
    {
      "acc": 0.64747858,
      "epoch": 1.417300862506342,
      "grad_norm": 7.5625,
      "learning_rate": 2.1466526012758194e-06,
      "loss": 1.59985256,
      "memory(GiB)": 117.38,
      "step": 55870,
      "train_speed(iter/s)": 1.63833
    },
    {
      "acc": 0.66212597,
      "epoch": 1.417427701674277,
      "grad_norm": 5.84375,
      "learning_rate": 2.145791555398594e-06,
      "loss": 1.54324026,
      "memory(GiB)": 117.38,
      "step": 55875,
      "train_speed(iter/s)": 1.638346
    },
    {
      "acc": 0.66027417,
      "epoch": 1.417554540842212,
      "grad_norm": 6.46875,
      "learning_rate": 2.144930635062199e-06,
      "loss": 1.51939335,
      "memory(GiB)": 117.38,
      "step": 55880,
      "train_speed(iter/s)": 1.638363
    },
    {
      "acc": 0.63951788,
      "epoch": 1.417681380010147,
      "grad_norm": 6.15625,
      "learning_rate": 2.144069840304502e-06,
      "loss": 1.6772913,
      "memory(GiB)": 117.38,
      "step": 55885,
      "train_speed(iter/s)": 1.63838
    },
    {
      "acc": 0.66648498,
      "epoch": 1.4178082191780823,
      "grad_norm": 8.8125,
      "learning_rate": 2.1432091711633634e-06,
      "loss": 1.62115002,
      "memory(GiB)": 117.38,
      "step": 55890,
      "train_speed(iter/s)": 1.638396
    },
    {
      "acc": 0.65990505,
      "epoch": 1.4179350583460173,
      "grad_norm": 5.8125,
      "learning_rate": 2.142348627676641e-06,
      "loss": 1.58954868,
      "memory(GiB)": 117.38,
      "step": 55895,
      "train_speed(iter/s)": 1.638412
    },
    {
      "acc": 0.65212393,
      "epoch": 1.4180618975139523,
      "grad_norm": 5.375,
      "learning_rate": 2.1414882098821836e-06,
      "loss": 1.59152317,
      "memory(GiB)": 117.38,
      "step": 55900,
      "train_speed(iter/s)": 1.638429
    },
    {
      "acc": 0.66284485,
      "epoch": 1.4181887366818873,
      "grad_norm": 6.03125,
      "learning_rate": 2.1406279178178355e-06,
      "loss": 1.61916275,
      "memory(GiB)": 117.38,
      "step": 55905,
      "train_speed(iter/s)": 1.638445
    },
    {
      "acc": 0.65922441,
      "epoch": 1.4183155758498225,
      "grad_norm": 5.875,
      "learning_rate": 2.1397677515214422e-06,
      "loss": 1.60093956,
      "memory(GiB)": 117.38,
      "step": 55910,
      "train_speed(iter/s)": 1.638462
    },
    {
      "acc": 0.65031242,
      "epoch": 1.4184424150177575,
      "grad_norm": 7.53125,
      "learning_rate": 2.1389077110308304e-06,
      "loss": 1.63191299,
      "memory(GiB)": 117.38,
      "step": 55915,
      "train_speed(iter/s)": 1.638478
    },
    {
      "acc": 0.64999566,
      "epoch": 1.4185692541856925,
      "grad_norm": 6.40625,
      "learning_rate": 2.138047796383832e-06,
      "loss": 1.66177025,
      "memory(GiB)": 117.38,
      "step": 55920,
      "train_speed(iter/s)": 1.638496
    },
    {
      "acc": 0.65035777,
      "epoch": 1.4186960933536277,
      "grad_norm": 9.3125,
      "learning_rate": 2.1371880076182666e-06,
      "loss": 1.66402454,
      "memory(GiB)": 117.38,
      "step": 55925,
      "train_speed(iter/s)": 1.638512
    },
    {
      "acc": 0.64330549,
      "epoch": 1.4188229325215627,
      "grad_norm": 6.8125,
      "learning_rate": 2.1363283447719584e-06,
      "loss": 1.67426949,
      "memory(GiB)": 117.38,
      "step": 55930,
      "train_speed(iter/s)": 1.638529
    },
    {
      "acc": 0.6642303,
      "epoch": 1.4189497716894977,
      "grad_norm": 7.21875,
      "learning_rate": 2.135468807882713e-06,
      "loss": 1.55817451,
      "memory(GiB)": 117.38,
      "step": 55935,
      "train_speed(iter/s)": 1.638547
    },
    {
      "acc": 0.64840069,
      "epoch": 1.4190766108574326,
      "grad_norm": 6.5,
      "learning_rate": 2.1346093969883367e-06,
      "loss": 1.63250809,
      "memory(GiB)": 117.38,
      "step": 55940,
      "train_speed(iter/s)": 1.638564
    },
    {
      "acc": 0.65852861,
      "epoch": 1.4192034500253679,
      "grad_norm": 5.5625,
      "learning_rate": 2.1337501121266345e-06,
      "loss": 1.56873293,
      "memory(GiB)": 117.38,
      "step": 55945,
      "train_speed(iter/s)": 1.63858
    },
    {
      "acc": 0.64395843,
      "epoch": 1.4193302891933028,
      "grad_norm": 6.90625,
      "learning_rate": 2.132890953335401e-06,
      "loss": 1.67318535,
      "memory(GiB)": 117.38,
      "step": 55950,
      "train_speed(iter/s)": 1.638599
    },
    {
      "acc": 0.65122981,
      "epoch": 1.419457128361238,
      "grad_norm": 5.75,
      "learning_rate": 2.1320319206524215e-06,
      "loss": 1.69549026,
      "memory(GiB)": 117.38,
      "step": 55955,
      "train_speed(iter/s)": 1.638616
    },
    {
      "acc": 0.65861716,
      "epoch": 1.419583967529173,
      "grad_norm": 5.0625,
      "learning_rate": 2.1311730141154813e-06,
      "loss": 1.60972843,
      "memory(GiB)": 117.38,
      "step": 55960,
      "train_speed(iter/s)": 1.638632
    },
    {
      "acc": 0.6550437,
      "epoch": 1.419710806697108,
      "grad_norm": 5.15625,
      "learning_rate": 2.1303142337623623e-06,
      "loss": 1.60616798,
      "memory(GiB)": 117.38,
      "step": 55965,
      "train_speed(iter/s)": 1.638649
    },
    {
      "acc": 0.65619144,
      "epoch": 1.419837645865043,
      "grad_norm": 6.34375,
      "learning_rate": 2.1294555796308375e-06,
      "loss": 1.57014132,
      "memory(GiB)": 117.38,
      "step": 55970,
      "train_speed(iter/s)": 1.638665
    },
    {
      "acc": 0.65816698,
      "epoch": 1.4199644850329782,
      "grad_norm": 5.90625,
      "learning_rate": 2.1285970517586686e-06,
      "loss": 1.6872406,
      "memory(GiB)": 117.38,
      "step": 55975,
      "train_speed(iter/s)": 1.638683
    },
    {
      "acc": 0.64084349,
      "epoch": 1.4200913242009132,
      "grad_norm": 6.8125,
      "learning_rate": 2.127738650183623e-06,
      "loss": 1.60930195,
      "memory(GiB)": 117.38,
      "step": 55980,
      "train_speed(iter/s)": 1.638699
    },
    {
      "acc": 0.65255833,
      "epoch": 1.4202181633688484,
      "grad_norm": 5.46875,
      "learning_rate": 2.1268803749434546e-06,
      "loss": 1.6115757,
      "memory(GiB)": 117.38,
      "step": 55985,
      "train_speed(iter/s)": 1.638715
    },
    {
      "acc": 0.64217582,
      "epoch": 1.4203450025367834,
      "grad_norm": 8.25,
      "learning_rate": 2.1260222260759158e-06,
      "loss": 1.5963171,
      "memory(GiB)": 117.38,
      "step": 55990,
      "train_speed(iter/s)": 1.638732
    },
    {
      "acc": 0.66537237,
      "epoch": 1.4204718417047184,
      "grad_norm": 5.46875,
      "learning_rate": 2.1251642036187502e-06,
      "loss": 1.60161285,
      "memory(GiB)": 117.38,
      "step": 55995,
      "train_speed(iter/s)": 1.638748
    },
    {
      "acc": 0.66448884,
      "epoch": 1.4205986808726534,
      "grad_norm": 5.46875,
      "learning_rate": 2.124306307609699e-06,
      "loss": 1.59291735,
      "memory(GiB)": 117.38,
      "step": 56000,
      "train_speed(iter/s)": 1.638764
    },
    {
      "epoch": 1.4205986808726534,
      "eval_acc": 0.6462279970847247,
      "eval_loss": 1.573153018951416,
      "eval_runtime": 58.4646,
      "eval_samples_per_second": 108.955,
      "eval_steps_per_second": 27.247,
      "step": 56000
    },
    {
      "acc": 0.63351436,
      "epoch": 1.4207255200405886,
      "grad_norm": 5.84375,
      "learning_rate": 2.123448538086495e-06,
      "loss": 1.70207481,
      "memory(GiB)": 117.38,
      "step": 56005,
      "train_speed(iter/s)": 1.635773
    },
    {
      "acc": 0.64949937,
      "epoch": 1.4208523592085236,
      "grad_norm": 6.65625,
      "learning_rate": 2.122590895086867e-06,
      "loss": 1.66971092,
      "memory(GiB)": 117.38,
      "step": 56010,
      "train_speed(iter/s)": 1.635787
    },
    {
      "acc": 0.65008979,
      "epoch": 1.4209791983764586,
      "grad_norm": 6.28125,
      "learning_rate": 2.1217333786485385e-06,
      "loss": 1.58441048,
      "memory(GiB)": 117.38,
      "step": 56015,
      "train_speed(iter/s)": 1.635804
    },
    {
      "acc": 0.65072289,
      "epoch": 1.4211060375443938,
      "grad_norm": 5.96875,
      "learning_rate": 2.120875988809226e-06,
      "loss": 1.63161125,
      "memory(GiB)": 117.38,
      "step": 56020,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.65938396,
      "epoch": 1.4212328767123288,
      "grad_norm": 6.0,
      "learning_rate": 2.1200187256066425e-06,
      "loss": 1.57050209,
      "memory(GiB)": 117.38,
      "step": 56025,
      "train_speed(iter/s)": 1.635835
    },
    {
      "acc": 0.66699615,
      "epoch": 1.4213597158802638,
      "grad_norm": 6.53125,
      "learning_rate": 2.119161589078493e-06,
      "loss": 1.59796228,
      "memory(GiB)": 117.38,
      "step": 56030,
      "train_speed(iter/s)": 1.63585
    },
    {
      "acc": 0.6388308,
      "epoch": 1.4214865550481988,
      "grad_norm": 5.90625,
      "learning_rate": 2.118304579262479e-06,
      "loss": 1.70197716,
      "memory(GiB)": 117.38,
      "step": 56035,
      "train_speed(iter/s)": 1.635865
    },
    {
      "acc": 0.67194672,
      "epoch": 1.421613394216134,
      "grad_norm": 6.4375,
      "learning_rate": 2.1174476961962957e-06,
      "loss": 1.58991671,
      "memory(GiB)": 117.38,
      "step": 56040,
      "train_speed(iter/s)": 1.635881
    },
    {
      "acc": 0.64458737,
      "epoch": 1.421740233384069,
      "grad_norm": 6.125,
      "learning_rate": 2.1165909399176328e-06,
      "loss": 1.6444334,
      "memory(GiB)": 117.38,
      "step": 56045,
      "train_speed(iter/s)": 1.635898
    },
    {
      "acc": 0.64549866,
      "epoch": 1.4218670725520042,
      "grad_norm": 5.5625,
      "learning_rate": 2.1157343104641733e-06,
      "loss": 1.65981789,
      "memory(GiB)": 117.38,
      "step": 56050,
      "train_speed(iter/s)": 1.635914
    },
    {
      "acc": 0.65684814,
      "epoch": 1.4219939117199392,
      "grad_norm": 6.25,
      "learning_rate": 2.114877807873596e-06,
      "loss": 1.55561247,
      "memory(GiB)": 117.38,
      "step": 56055,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.64241743,
      "epoch": 1.4221207508878742,
      "grad_norm": 5.28125,
      "learning_rate": 2.114021432183574e-06,
      "loss": 1.6173317,
      "memory(GiB)": 117.38,
      "step": 56060,
      "train_speed(iter/s)": 1.635945
    },
    {
      "acc": 0.64835644,
      "epoch": 1.4222475900558091,
      "grad_norm": 5.71875,
      "learning_rate": 2.113165183431773e-06,
      "loss": 1.61581402,
      "memory(GiB)": 117.38,
      "step": 56065,
      "train_speed(iter/s)": 1.635961
    },
    {
      "acc": 0.67609644,
      "epoch": 1.4223744292237444,
      "grad_norm": 6.8125,
      "learning_rate": 2.112309061655859e-06,
      "loss": 1.50941238,
      "memory(GiB)": 117.38,
      "step": 56070,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.68436561,
      "epoch": 1.4225012683916793,
      "grad_norm": 6.15625,
      "learning_rate": 2.1114530668934836e-06,
      "loss": 1.51605415,
      "memory(GiB)": 117.38,
      "step": 56075,
      "train_speed(iter/s)": 1.635991
    },
    {
      "acc": 0.64157991,
      "epoch": 1.4226281075596143,
      "grad_norm": 5.59375,
      "learning_rate": 2.1105971991822966e-06,
      "loss": 1.61358604,
      "memory(GiB)": 117.38,
      "step": 56080,
      "train_speed(iter/s)": 1.636007
    },
    {
      "acc": 0.64268394,
      "epoch": 1.4227549467275495,
      "grad_norm": 7.59375,
      "learning_rate": 2.1097414585599474e-06,
      "loss": 1.65658455,
      "memory(GiB)": 117.38,
      "step": 56085,
      "train_speed(iter/s)": 1.636024
    },
    {
      "acc": 0.66413655,
      "epoch": 1.4228817858954845,
      "grad_norm": 6.71875,
      "learning_rate": 2.1088858450640743e-06,
      "loss": 1.63587303,
      "memory(GiB)": 117.38,
      "step": 56090,
      "train_speed(iter/s)": 1.636039
    },
    {
      "acc": 0.66147404,
      "epoch": 1.4230086250634195,
      "grad_norm": 7.3125,
      "learning_rate": 2.108030358732308e-06,
      "loss": 1.57082043,
      "memory(GiB)": 117.38,
      "step": 56095,
      "train_speed(iter/s)": 1.636056
    },
    {
      "acc": 0.65377502,
      "epoch": 1.4231354642313545,
      "grad_norm": 5.09375,
      "learning_rate": 2.107174999602277e-06,
      "loss": 1.62279377,
      "memory(GiB)": 117.38,
      "step": 56100,
      "train_speed(iter/s)": 1.636071
    },
    {
      "acc": 0.66465039,
      "epoch": 1.4232623033992897,
      "grad_norm": 6.03125,
      "learning_rate": 2.106319767711606e-06,
      "loss": 1.561306,
      "memory(GiB)": 117.38,
      "step": 56105,
      "train_speed(iter/s)": 1.636087
    },
    {
      "acc": 0.64482765,
      "epoch": 1.4233891425672247,
      "grad_norm": 6.34375,
      "learning_rate": 2.105464663097913e-06,
      "loss": 1.64756794,
      "memory(GiB)": 117.38,
      "step": 56110,
      "train_speed(iter/s)": 1.636103
    },
    {
      "acc": 0.66628647,
      "epoch": 1.42351598173516,
      "grad_norm": 6.03125,
      "learning_rate": 2.1046096857988033e-06,
      "loss": 1.55195751,
      "memory(GiB)": 117.38,
      "step": 56115,
      "train_speed(iter/s)": 1.636117
    },
    {
      "acc": 0.6538125,
      "epoch": 1.423642820903095,
      "grad_norm": 5.0625,
      "learning_rate": 2.103754835851889e-06,
      "loss": 1.5825346,
      "memory(GiB)": 117.38,
      "step": 56120,
      "train_speed(iter/s)": 1.636133
    },
    {
      "acc": 0.66755147,
      "epoch": 1.42376966007103,
      "grad_norm": 5.40625,
      "learning_rate": 2.102900113294768e-06,
      "loss": 1.54982643,
      "memory(GiB)": 117.38,
      "step": 56125,
      "train_speed(iter/s)": 1.636148
    },
    {
      "acc": 0.647686,
      "epoch": 1.4238964992389649,
      "grad_norm": 5.90625,
      "learning_rate": 2.1020455181650356e-06,
      "loss": 1.58786545,
      "memory(GiB)": 117.38,
      "step": 56130,
      "train_speed(iter/s)": 1.636164
    },
    {
      "acc": 0.66273246,
      "epoch": 1.4240233384069,
      "grad_norm": 6.0625,
      "learning_rate": 2.101191050500277e-06,
      "loss": 1.59326544,
      "memory(GiB)": 117.38,
      "step": 56135,
      "train_speed(iter/s)": 1.63618
    },
    {
      "acc": 0.64656487,
      "epoch": 1.424150177574835,
      "grad_norm": 6.03125,
      "learning_rate": 2.1003367103380797e-06,
      "loss": 1.61582146,
      "memory(GiB)": 117.38,
      "step": 56140,
      "train_speed(iter/s)": 1.636196
    },
    {
      "acc": 0.65650616,
      "epoch": 1.4242770167427703,
      "grad_norm": 6.46875,
      "learning_rate": 2.09948249771602e-06,
      "loss": 1.66964779,
      "memory(GiB)": 117.38,
      "step": 56145,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.65303869,
      "epoch": 1.4244038559107053,
      "grad_norm": 6.96875,
      "learning_rate": 2.09862841267167e-06,
      "loss": 1.52922688,
      "memory(GiB)": 117.38,
      "step": 56150,
      "train_speed(iter/s)": 1.636226
    },
    {
      "acc": 0.65279756,
      "epoch": 1.4245306950786403,
      "grad_norm": 7.125,
      "learning_rate": 2.097774455242596e-06,
      "loss": 1.62442284,
      "memory(GiB)": 117.38,
      "step": 56155,
      "train_speed(iter/s)": 1.636242
    },
    {
      "acc": 0.65691862,
      "epoch": 1.4246575342465753,
      "grad_norm": 5.53125,
      "learning_rate": 2.096920625466359e-06,
      "loss": 1.58108883,
      "memory(GiB)": 117.38,
      "step": 56160,
      "train_speed(iter/s)": 1.636257
    },
    {
      "acc": 0.64106445,
      "epoch": 1.4247843734145105,
      "grad_norm": 5.4375,
      "learning_rate": 2.0960669233805136e-06,
      "loss": 1.66551342,
      "memory(GiB)": 117.38,
      "step": 56165,
      "train_speed(iter/s)": 1.636269
    },
    {
      "acc": 0.65888739,
      "epoch": 1.4249112125824455,
      "grad_norm": 4.78125,
      "learning_rate": 2.0952133490226106e-06,
      "loss": 1.66334324,
      "memory(GiB)": 117.38,
      "step": 56170,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.649435,
      "epoch": 1.4250380517503805,
      "grad_norm": 6.28125,
      "learning_rate": 2.0943599024301935e-06,
      "loss": 1.64008312,
      "memory(GiB)": 117.38,
      "step": 56175,
      "train_speed(iter/s)": 1.636301
    },
    {
      "acc": 0.66099072,
      "epoch": 1.4251648909183157,
      "grad_norm": 6.21875,
      "learning_rate": 2.093506583640801e-06,
      "loss": 1.55307102,
      "memory(GiB)": 117.38,
      "step": 56180,
      "train_speed(iter/s)": 1.636317
    },
    {
      "acc": 0.65816002,
      "epoch": 1.4252917300862507,
      "grad_norm": 6.0625,
      "learning_rate": 2.092653392691965e-06,
      "loss": 1.56229925,
      "memory(GiB)": 117.38,
      "step": 56185,
      "train_speed(iter/s)": 1.636333
    },
    {
      "acc": 0.65581341,
      "epoch": 1.4254185692541856,
      "grad_norm": 6.1875,
      "learning_rate": 2.091800329621213e-06,
      "loss": 1.59925108,
      "memory(GiB)": 117.38,
      "step": 56190,
      "train_speed(iter/s)": 1.636349
    },
    {
      "acc": 0.67187867,
      "epoch": 1.4255454084221206,
      "grad_norm": 5.0625,
      "learning_rate": 2.0909473944660667e-06,
      "loss": 1.5537281,
      "memory(GiB)": 117.38,
      "step": 56195,
      "train_speed(iter/s)": 1.636365
    },
    {
      "acc": 0.65392485,
      "epoch": 1.4256722475900558,
      "grad_norm": 5.875,
      "learning_rate": 2.0900945872640427e-06,
      "loss": 1.64563847,
      "memory(GiB)": 117.38,
      "step": 56200,
      "train_speed(iter/s)": 1.636382
    },
    {
      "acc": 0.66792488,
      "epoch": 1.4257990867579908,
      "grad_norm": 6.59375,
      "learning_rate": 2.0892419080526484e-06,
      "loss": 1.5796175,
      "memory(GiB)": 117.38,
      "step": 56205,
      "train_speed(iter/s)": 1.636397
    },
    {
      "acc": 0.66316986,
      "epoch": 1.425925925925926,
      "grad_norm": 5.875,
      "learning_rate": 2.0883893568693935e-06,
      "loss": 1.56624556,
      "memory(GiB)": 117.38,
      "step": 56210,
      "train_speed(iter/s)": 1.636413
    },
    {
      "acc": 0.66564879,
      "epoch": 1.426052765093861,
      "grad_norm": 5.25,
      "learning_rate": 2.087536933751773e-06,
      "loss": 1.59430733,
      "memory(GiB)": 117.38,
      "step": 56215,
      "train_speed(iter/s)": 1.63643
    },
    {
      "acc": 0.65701547,
      "epoch": 1.426179604261796,
      "grad_norm": 7.78125,
      "learning_rate": 2.0866846387372814e-06,
      "loss": 1.61675491,
      "memory(GiB)": 117.38,
      "step": 56220,
      "train_speed(iter/s)": 1.636445
    },
    {
      "acc": 0.64128389,
      "epoch": 1.426306443429731,
      "grad_norm": 5.75,
      "learning_rate": 2.085832471863404e-06,
      "loss": 1.69387493,
      "memory(GiB)": 117.38,
      "step": 56225,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.64727774,
      "epoch": 1.4264332825976662,
      "grad_norm": 6.0625,
      "learning_rate": 2.08498043316763e-06,
      "loss": 1.65560341,
      "memory(GiB)": 117.38,
      "step": 56230,
      "train_speed(iter/s)": 1.636479
    },
    {
      "acc": 0.6692924,
      "epoch": 1.4265601217656012,
      "grad_norm": 6.15625,
      "learning_rate": 2.0841285226874282e-06,
      "loss": 1.59545631,
      "memory(GiB)": 117.38,
      "step": 56235,
      "train_speed(iter/s)": 1.636494
    },
    {
      "acc": 0.6468823,
      "epoch": 1.4266869609335362,
      "grad_norm": 5.53125,
      "learning_rate": 2.083276740460271e-06,
      "loss": 1.63130455,
      "memory(GiB)": 117.38,
      "step": 56240,
      "train_speed(iter/s)": 1.636511
    },
    {
      "acc": 0.66154289,
      "epoch": 1.4268138001014714,
      "grad_norm": 5.25,
      "learning_rate": 2.082425086523627e-06,
      "loss": 1.52195644,
      "memory(GiB)": 117.38,
      "step": 56245,
      "train_speed(iter/s)": 1.636526
    },
    {
      "acc": 0.65864873,
      "epoch": 1.4269406392694064,
      "grad_norm": 5.875,
      "learning_rate": 2.0815735609149556e-06,
      "loss": 1.57371788,
      "memory(GiB)": 117.38,
      "step": 56250,
      "train_speed(iter/s)": 1.636543
    },
    {
      "acc": 0.67760506,
      "epoch": 1.4270674784373414,
      "grad_norm": 5.46875,
      "learning_rate": 2.080722163671705e-06,
      "loss": 1.54128227,
      "memory(GiB)": 117.38,
      "step": 56255,
      "train_speed(iter/s)": 1.63656
    },
    {
      "acc": 0.66518574,
      "epoch": 1.4271943176052764,
      "grad_norm": 5.40625,
      "learning_rate": 2.079870894831329e-06,
      "loss": 1.60316563,
      "memory(GiB)": 117.38,
      "step": 56260,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.65389404,
      "epoch": 1.4273211567732116,
      "grad_norm": 5.4375,
      "learning_rate": 2.0790197544312683e-06,
      "loss": 1.64111633,
      "memory(GiB)": 117.38,
      "step": 56265,
      "train_speed(iter/s)": 1.636591
    },
    {
      "acc": 0.6464653,
      "epoch": 1.4274479959411466,
      "grad_norm": 5.375,
      "learning_rate": 2.0781687425089624e-06,
      "loss": 1.64246216,
      "memory(GiB)": 117.38,
      "step": 56270,
      "train_speed(iter/s)": 1.636606
    },
    {
      "acc": 0.65611019,
      "epoch": 1.4275748351090818,
      "grad_norm": 6.90625,
      "learning_rate": 2.0773178591018363e-06,
      "loss": 1.67025375,
      "memory(GiB)": 117.38,
      "step": 56275,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.67640247,
      "epoch": 1.4277016742770168,
      "grad_norm": 6.21875,
      "learning_rate": 2.076467104247322e-06,
      "loss": 1.51072865,
      "memory(GiB)": 117.38,
      "step": 56280,
      "train_speed(iter/s)": 1.636638
    },
    {
      "acc": 0.67425675,
      "epoch": 1.4278285134449518,
      "grad_norm": 5.28125,
      "learning_rate": 2.0756164779828365e-06,
      "loss": 1.51532183,
      "memory(GiB)": 117.38,
      "step": 56285,
      "train_speed(iter/s)": 1.636655
    },
    {
      "acc": 0.65487924,
      "epoch": 1.4279553526128868,
      "grad_norm": 4.90625,
      "learning_rate": 2.0747659803457943e-06,
      "loss": 1.58731823,
      "memory(GiB)": 117.38,
      "step": 56290,
      "train_speed(iter/s)": 1.636671
    },
    {
      "acc": 0.67112637,
      "epoch": 1.428082191780822,
      "grad_norm": 5.25,
      "learning_rate": 2.073915611373605e-06,
      "loss": 1.5472847,
      "memory(GiB)": 117.38,
      "step": 56295,
      "train_speed(iter/s)": 1.636687
    },
    {
      "acc": 0.64999266,
      "epoch": 1.428209030948757,
      "grad_norm": 4.9375,
      "learning_rate": 2.0730653711036713e-06,
      "loss": 1.59760695,
      "memory(GiB)": 117.38,
      "step": 56300,
      "train_speed(iter/s)": 1.636703
    },
    {
      "acc": 0.66855268,
      "epoch": 1.4283358701166922,
      "grad_norm": 6.25,
      "learning_rate": 2.0722152595733903e-06,
      "loss": 1.51363525,
      "memory(GiB)": 117.38,
      "step": 56305,
      "train_speed(iter/s)": 1.636719
    },
    {
      "acc": 0.66257396,
      "epoch": 1.4284627092846272,
      "grad_norm": 6.0,
      "learning_rate": 2.0713652768201536e-06,
      "loss": 1.56442022,
      "memory(GiB)": 117.38,
      "step": 56310,
      "train_speed(iter/s)": 1.636737
    },
    {
      "acc": 0.65923796,
      "epoch": 1.4285895484525621,
      "grad_norm": 7.0625,
      "learning_rate": 2.0705154228813477e-06,
      "loss": 1.60552979,
      "memory(GiB)": 117.38,
      "step": 56315,
      "train_speed(iter/s)": 1.636754
    },
    {
      "acc": 0.66712313,
      "epoch": 1.4287163876204971,
      "grad_norm": 5.03125,
      "learning_rate": 2.0696656977943524e-06,
      "loss": 1.52747192,
      "memory(GiB)": 117.38,
      "step": 56320,
      "train_speed(iter/s)": 1.636769
    },
    {
      "acc": 0.64965863,
      "epoch": 1.4288432267884323,
      "grad_norm": 5.78125,
      "learning_rate": 2.068816101596543e-06,
      "loss": 1.63623161,
      "memory(GiB)": 117.38,
      "step": 56325,
      "train_speed(iter/s)": 1.636784
    },
    {
      "acc": 0.6621418,
      "epoch": 1.4289700659563673,
      "grad_norm": 5.0625,
      "learning_rate": 2.067966634325288e-06,
      "loss": 1.5722641,
      "memory(GiB)": 117.38,
      "step": 56330,
      "train_speed(iter/s)": 1.6368
    },
    {
      "acc": 0.66067924,
      "epoch": 1.4290969051243023,
      "grad_norm": 6.78125,
      "learning_rate": 2.0671172960179513e-06,
      "loss": 1.57224522,
      "memory(GiB)": 117.38,
      "step": 56335,
      "train_speed(iter/s)": 1.636816
    },
    {
      "acc": 0.66659017,
      "epoch": 1.4292237442922375,
      "grad_norm": 5.65625,
      "learning_rate": 2.06626808671189e-06,
      "loss": 1.5469346,
      "memory(GiB)": 117.38,
      "step": 56340,
      "train_speed(iter/s)": 1.636832
    },
    {
      "acc": 0.65840988,
      "epoch": 1.4293505834601725,
      "grad_norm": 6.1875,
      "learning_rate": 2.065419006444455e-06,
      "loss": 1.60647335,
      "memory(GiB)": 117.38,
      "step": 56345,
      "train_speed(iter/s)": 1.636848
    },
    {
      "acc": 0.66085382,
      "epoch": 1.4294774226281075,
      "grad_norm": 5.8125,
      "learning_rate": 2.0645700552529973e-06,
      "loss": 1.60675125,
      "memory(GiB)": 117.38,
      "step": 56350,
      "train_speed(iter/s)": 1.636865
    },
    {
      "acc": 0.63569336,
      "epoch": 1.4296042617960425,
      "grad_norm": 5.40625,
      "learning_rate": 2.063721233174853e-06,
      "loss": 1.6966711,
      "memory(GiB)": 117.38,
      "step": 56355,
      "train_speed(iter/s)": 1.63688
    },
    {
      "acc": 0.66826468,
      "epoch": 1.4297311009639777,
      "grad_norm": 6.6875,
      "learning_rate": 2.0628725402473574e-06,
      "loss": 1.50753717,
      "memory(GiB)": 117.38,
      "step": 56360,
      "train_speed(iter/s)": 1.636896
    },
    {
      "acc": 0.66465473,
      "epoch": 1.4298579401319127,
      "grad_norm": 7.0,
      "learning_rate": 2.0620239765078404e-06,
      "loss": 1.55679455,
      "memory(GiB)": 117.38,
      "step": 56365,
      "train_speed(iter/s)": 1.636912
    },
    {
      "acc": 0.65117722,
      "epoch": 1.429984779299848,
      "grad_norm": 5.1875,
      "learning_rate": 2.0611755419936285e-06,
      "loss": 1.57975626,
      "memory(GiB)": 117.38,
      "step": 56370,
      "train_speed(iter/s)": 1.636928
    },
    {
      "acc": 0.66004658,
      "epoch": 1.430111618467783,
      "grad_norm": 5.3125,
      "learning_rate": 2.0603272367420357e-06,
      "loss": 1.58404722,
      "memory(GiB)": 117.38,
      "step": 56375,
      "train_speed(iter/s)": 1.636945
    },
    {
      "acc": 0.66127939,
      "epoch": 1.4302384576357179,
      "grad_norm": 6.0,
      "learning_rate": 2.0594790607903743e-06,
      "loss": 1.55673847,
      "memory(GiB)": 117.38,
      "step": 56380,
      "train_speed(iter/s)": 1.636961
    },
    {
      "acc": 0.66097116,
      "epoch": 1.4303652968036529,
      "grad_norm": 5.375,
      "learning_rate": 2.0586310141759534e-06,
      "loss": 1.58874025,
      "memory(GiB)": 117.38,
      "step": 56385,
      "train_speed(iter/s)": 1.636977
    },
    {
      "acc": 0.64952879,
      "epoch": 1.430492135971588,
      "grad_norm": 5.71875,
      "learning_rate": 2.0577830969360753e-06,
      "loss": 1.61046181,
      "memory(GiB)": 117.38,
      "step": 56390,
      "train_speed(iter/s)": 1.636995
    },
    {
      "acc": 0.66503525,
      "epoch": 1.430618975139523,
      "grad_norm": 6.84375,
      "learning_rate": 2.0569353091080304e-06,
      "loss": 1.56640615,
      "memory(GiB)": 117.38,
      "step": 56395,
      "train_speed(iter/s)": 1.63701
    },
    {
      "acc": 0.66329212,
      "epoch": 1.430745814307458,
      "grad_norm": 5.6875,
      "learning_rate": 2.056087650729109e-06,
      "loss": 1.52314224,
      "memory(GiB)": 117.38,
      "step": 56400,
      "train_speed(iter/s)": 1.637026
    },
    {
      "acc": 0.6475297,
      "epoch": 1.4308726534753933,
      "grad_norm": 5.4375,
      "learning_rate": 2.0552401218365975e-06,
      "loss": 1.63864059,
      "memory(GiB)": 117.38,
      "step": 56405,
      "train_speed(iter/s)": 1.637041
    },
    {
      "acc": 0.65295458,
      "epoch": 1.4309994926433283,
      "grad_norm": 4.90625,
      "learning_rate": 2.054392722467775e-06,
      "loss": 1.56698761,
      "memory(GiB)": 117.38,
      "step": 56410,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.63664794,
      "epoch": 1.4311263318112633,
      "grad_norm": 5.46875,
      "learning_rate": 2.0535454526599086e-06,
      "loss": 1.70207424,
      "memory(GiB)": 117.38,
      "step": 56415,
      "train_speed(iter/s)": 1.637073
    },
    {
      "acc": 0.65941172,
      "epoch": 1.4312531709791982,
      "grad_norm": 6.46875,
      "learning_rate": 2.0526983124502692e-06,
      "loss": 1.55500393,
      "memory(GiB)": 117.38,
      "step": 56420,
      "train_speed(iter/s)": 1.637087
    },
    {
      "acc": 0.64197426,
      "epoch": 1.4313800101471335,
      "grad_norm": 6.46875,
      "learning_rate": 2.051851301876117e-06,
      "loss": 1.64885674,
      "memory(GiB)": 117.38,
      "step": 56425,
      "train_speed(iter/s)": 1.637103
    },
    {
      "acc": 0.65074062,
      "epoch": 1.4315068493150684,
      "grad_norm": 6.59375,
      "learning_rate": 2.0510044209747078e-06,
      "loss": 1.64282455,
      "memory(GiB)": 117.38,
      "step": 56430,
      "train_speed(iter/s)": 1.637119
    },
    {
      "acc": 0.65405674,
      "epoch": 1.4316336884830037,
      "grad_norm": 5.84375,
      "learning_rate": 2.05015766978329e-06,
      "loss": 1.60566139,
      "memory(GiB)": 117.38,
      "step": 56435,
      "train_speed(iter/s)": 1.637135
    },
    {
      "acc": 0.64742441,
      "epoch": 1.4317605276509386,
      "grad_norm": 5.25,
      "learning_rate": 2.049311048339108e-06,
      "loss": 1.67665367,
      "memory(GiB)": 117.38,
      "step": 56440,
      "train_speed(iter/s)": 1.637151
    },
    {
      "acc": 0.65694876,
      "epoch": 1.4318873668188736,
      "grad_norm": 5.3125,
      "learning_rate": 2.0484645566793993e-06,
      "loss": 1.62892342,
      "memory(GiB)": 117.38,
      "step": 56445,
      "train_speed(iter/s)": 1.637166
    },
    {
      "acc": 0.64785614,
      "epoch": 1.4320142059868086,
      "grad_norm": 6.03125,
      "learning_rate": 2.0476181948413975e-06,
      "loss": 1.67124596,
      "memory(GiB)": 117.38,
      "step": 56450,
      "train_speed(iter/s)": 1.637183
    },
    {
      "acc": 0.65103188,
      "epoch": 1.4321410451547438,
      "grad_norm": 6.40625,
      "learning_rate": 2.0467719628623293e-06,
      "loss": 1.59926062,
      "memory(GiB)": 117.38,
      "step": 56455,
      "train_speed(iter/s)": 1.637199
    },
    {
      "acc": 0.66634035,
      "epoch": 1.4322678843226788,
      "grad_norm": 5.53125,
      "learning_rate": 2.045925860779415e-06,
      "loss": 1.52695847,
      "memory(GiB)": 117.38,
      "step": 56460,
      "train_speed(iter/s)": 1.637215
    },
    {
      "acc": 0.64607854,
      "epoch": 1.432394723490614,
      "grad_norm": 5.96875,
      "learning_rate": 2.0450798886298707e-06,
      "loss": 1.60617828,
      "memory(GiB)": 117.38,
      "step": 56465,
      "train_speed(iter/s)": 1.63723
    },
    {
      "acc": 0.65271149,
      "epoch": 1.432521562658549,
      "grad_norm": 5.5,
      "learning_rate": 2.044234046450905e-06,
      "loss": 1.57414618,
      "memory(GiB)": 117.38,
      "step": 56470,
      "train_speed(iter/s)": 1.637245
    },
    {
      "acc": 0.66269073,
      "epoch": 1.432648401826484,
      "grad_norm": 5.5,
      "learning_rate": 2.0433883342797233e-06,
      "loss": 1.54614553,
      "memory(GiB)": 117.38,
      "step": 56475,
      "train_speed(iter/s)": 1.637261
    },
    {
      "acc": 0.66632228,
      "epoch": 1.432775240994419,
      "grad_norm": 5.5,
      "learning_rate": 2.042542752153522e-06,
      "loss": 1.56111469,
      "memory(GiB)": 117.38,
      "step": 56480,
      "train_speed(iter/s)": 1.637277
    },
    {
      "acc": 0.65908566,
      "epoch": 1.4329020801623542,
      "grad_norm": 6.8125,
      "learning_rate": 2.0416973001094953e-06,
      "loss": 1.60543747,
      "memory(GiB)": 117.38,
      "step": 56485,
      "train_speed(iter/s)": 1.637293
    },
    {
      "acc": 0.66335478,
      "epoch": 1.4330289193302892,
      "grad_norm": 6.0625,
      "learning_rate": 2.0408519781848292e-06,
      "loss": 1.60566158,
      "memory(GiB)": 117.38,
      "step": 56490,
      "train_speed(iter/s)": 1.637309
    },
    {
      "acc": 0.673488,
      "epoch": 1.4331557584982242,
      "grad_norm": 4.9375,
      "learning_rate": 2.0400067864167044e-06,
      "loss": 1.52056847,
      "memory(GiB)": 117.38,
      "step": 56495,
      "train_speed(iter/s)": 1.637324
    },
    {
      "acc": 0.66895247,
      "epoch": 1.4332825976661594,
      "grad_norm": 5.625,
      "learning_rate": 2.0391617248422967e-06,
      "loss": 1.54592705,
      "memory(GiB)": 117.38,
      "step": 56500,
      "train_speed(iter/s)": 1.63734
    },
    {
      "acc": 0.65170913,
      "epoch": 1.4334094368340944,
      "grad_norm": 6.28125,
      "learning_rate": 2.038316793498774e-06,
      "loss": 1.66394424,
      "memory(GiB)": 117.38,
      "step": 56505,
      "train_speed(iter/s)": 1.637355
    },
    {
      "acc": 0.65481281,
      "epoch": 1.4335362760020294,
      "grad_norm": 7.40625,
      "learning_rate": 2.037471992423305e-06,
      "loss": 1.59310789,
      "memory(GiB)": 117.38,
      "step": 56510,
      "train_speed(iter/s)": 1.637371
    },
    {
      "acc": 0.63660326,
      "epoch": 1.4336631151699644,
      "grad_norm": 6.34375,
      "learning_rate": 2.036627321653043e-06,
      "loss": 1.66875992,
      "memory(GiB)": 117.38,
      "step": 56515,
      "train_speed(iter/s)": 1.637387
    },
    {
      "acc": 0.64011612,
      "epoch": 1.4337899543378996,
      "grad_norm": 7.0,
      "learning_rate": 2.0357827812251403e-06,
      "loss": 1.67292538,
      "memory(GiB)": 117.38,
      "step": 56520,
      "train_speed(iter/s)": 1.637403
    },
    {
      "acc": 0.65720253,
      "epoch": 1.4339167935058346,
      "grad_norm": 7.0,
      "learning_rate": 2.0349383711767463e-06,
      "loss": 1.62096252,
      "memory(GiB)": 117.38,
      "step": 56525,
      "train_speed(iter/s)": 1.63742
    },
    {
      "acc": 0.64625545,
      "epoch": 1.4340436326737698,
      "grad_norm": 5.78125,
      "learning_rate": 2.0340940915450026e-06,
      "loss": 1.59795609,
      "memory(GiB)": 117.38,
      "step": 56530,
      "train_speed(iter/s)": 1.637435
    },
    {
      "acc": 0.64653025,
      "epoch": 1.4341704718417048,
      "grad_norm": 6.03125,
      "learning_rate": 2.033249942367041e-06,
      "loss": 1.62711067,
      "memory(GiB)": 117.38,
      "step": 56535,
      "train_speed(iter/s)": 1.637451
    },
    {
      "acc": 0.66112103,
      "epoch": 1.4342973110096398,
      "grad_norm": 6.03125,
      "learning_rate": 2.032405923679991e-06,
      "loss": 1.54778318,
      "memory(GiB)": 117.38,
      "step": 56540,
      "train_speed(iter/s)": 1.637466
    },
    {
      "acc": 0.65065727,
      "epoch": 1.4344241501775747,
      "grad_norm": 4.96875,
      "learning_rate": 2.0315620355209792e-06,
      "loss": 1.64242725,
      "memory(GiB)": 117.38,
      "step": 56545,
      "train_speed(iter/s)": 1.637482
    },
    {
      "acc": 0.6565176,
      "epoch": 1.43455098934551,
      "grad_norm": 8.625,
      "learning_rate": 2.030718277927124e-06,
      "loss": 1.61386261,
      "memory(GiB)": 117.38,
      "step": 56550,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.67432275,
      "epoch": 1.434677828513445,
      "grad_norm": 5.84375,
      "learning_rate": 2.0298746509355326e-06,
      "loss": 1.53392601,
      "memory(GiB)": 117.38,
      "step": 56555,
      "train_speed(iter/s)": 1.637514
    },
    {
      "acc": 0.66257658,
      "epoch": 1.43480466768138,
      "grad_norm": 5.09375,
      "learning_rate": 2.0290311545833166e-06,
      "loss": 1.58483057,
      "memory(GiB)": 117.38,
      "step": 56560,
      "train_speed(iter/s)": 1.63753
    },
    {
      "acc": 0.67124596,
      "epoch": 1.4349315068493151,
      "grad_norm": 5.1875,
      "learning_rate": 2.028187788907574e-06,
      "loss": 1.58956041,
      "memory(GiB)": 117.38,
      "step": 56565,
      "train_speed(iter/s)": 1.637546
    },
    {
      "acc": 0.66505065,
      "epoch": 1.4350583460172501,
      "grad_norm": 4.625,
      "learning_rate": 2.027344553945403e-06,
      "loss": 1.60864868,
      "memory(GiB)": 117.38,
      "step": 56570,
      "train_speed(iter/s)": 1.637561
    },
    {
      "acc": 0.66054506,
      "epoch": 1.4351851851851851,
      "grad_norm": 5.09375,
      "learning_rate": 2.0265014497338868e-06,
      "loss": 1.54737091,
      "memory(GiB)": 117.38,
      "step": 56575,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.64677486,
      "epoch": 1.43531202435312,
      "grad_norm": 7.0,
      "learning_rate": 2.0256584763101145e-06,
      "loss": 1.61277962,
      "memory(GiB)": 117.38,
      "step": 56580,
      "train_speed(iter/s)": 1.637592
    },
    {
      "acc": 0.6490932,
      "epoch": 1.4354388635210553,
      "grad_norm": 6.03125,
      "learning_rate": 2.024815633711162e-06,
      "loss": 1.6407465,
      "memory(GiB)": 117.38,
      "step": 56585,
      "train_speed(iter/s)": 1.637607
    },
    {
      "acc": 0.65596676,
      "epoch": 1.4355657026889903,
      "grad_norm": 5.28125,
      "learning_rate": 2.0239729219741005e-06,
      "loss": 1.56711426,
      "memory(GiB)": 117.38,
      "step": 56590,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.65874748,
      "epoch": 1.4356925418569255,
      "grad_norm": 6.5,
      "learning_rate": 2.0231303411359975e-06,
      "loss": 1.56645145,
      "memory(GiB)": 117.38,
      "step": 56595,
      "train_speed(iter/s)": 1.637636
    },
    {
      "acc": 0.6512485,
      "epoch": 1.4358193810248605,
      "grad_norm": 6.8125,
      "learning_rate": 2.0222878912339127e-06,
      "loss": 1.62066383,
      "memory(GiB)": 117.38,
      "step": 56600,
      "train_speed(iter/s)": 1.637652
    },
    {
      "acc": 0.66414938,
      "epoch": 1.4359462201927955,
      "grad_norm": 5.4375,
      "learning_rate": 2.021445572304901e-06,
      "loss": 1.61221886,
      "memory(GiB)": 117.38,
      "step": 56605,
      "train_speed(iter/s)": 1.637667
    },
    {
      "acc": 0.66144352,
      "epoch": 1.4360730593607305,
      "grad_norm": 6.46875,
      "learning_rate": 2.0206033843860113e-06,
      "loss": 1.56715546,
      "memory(GiB)": 117.38,
      "step": 56610,
      "train_speed(iter/s)": 1.637682
    },
    {
      "acc": 0.65715666,
      "epoch": 1.4361998985286657,
      "grad_norm": 5.21875,
      "learning_rate": 2.0197613275142868e-06,
      "loss": 1.60859222,
      "memory(GiB)": 117.38,
      "step": 56615,
      "train_speed(iter/s)": 1.637698
    },
    {
      "acc": 0.66180749,
      "epoch": 1.4363267376966007,
      "grad_norm": 6.78125,
      "learning_rate": 2.018919401726765e-06,
      "loss": 1.56663246,
      "memory(GiB)": 117.38,
      "step": 56620,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.65509725,
      "epoch": 1.436453576864536,
      "grad_norm": 4.84375,
      "learning_rate": 2.0180776070604773e-06,
      "loss": 1.61059113,
      "memory(GiB)": 117.38,
      "step": 56625,
      "train_speed(iter/s)": 1.63773
    },
    {
      "acc": 0.66102705,
      "epoch": 1.4365804160324709,
      "grad_norm": 5.71875,
      "learning_rate": 2.0172359435524497e-06,
      "loss": 1.64509811,
      "memory(GiB)": 117.38,
      "step": 56630,
      "train_speed(iter/s)": 1.637745
    },
    {
      "acc": 0.65620785,
      "epoch": 1.4367072552004059,
      "grad_norm": 6.875,
      "learning_rate": 2.0163944112397027e-06,
      "loss": 1.57133694,
      "memory(GiB)": 117.38,
      "step": 56635,
      "train_speed(iter/s)": 1.63776
    },
    {
      "acc": 0.65012503,
      "epoch": 1.4368340943683409,
      "grad_norm": 5.3125,
      "learning_rate": 2.01555301015925e-06,
      "loss": 1.64002457,
      "memory(GiB)": 117.38,
      "step": 56640,
      "train_speed(iter/s)": 1.637775
    },
    {
      "acc": 0.6535697,
      "epoch": 1.436960933536276,
      "grad_norm": 4.875,
      "learning_rate": 2.0147117403480994e-06,
      "loss": 1.62286396,
      "memory(GiB)": 117.38,
      "step": 56645,
      "train_speed(iter/s)": 1.637792
    },
    {
      "acc": 0.66047297,
      "epoch": 1.437087772704211,
      "grad_norm": 5.34375,
      "learning_rate": 2.0138706018432576e-06,
      "loss": 1.54875898,
      "memory(GiB)": 117.38,
      "step": 56650,
      "train_speed(iter/s)": 1.637808
    },
    {
      "acc": 0.64496484,
      "epoch": 1.437214611872146,
      "grad_norm": 4.5625,
      "learning_rate": 2.0130295946817176e-06,
      "loss": 1.6263237,
      "memory(GiB)": 117.38,
      "step": 56655,
      "train_speed(iter/s)": 1.637823
    },
    {
      "acc": 0.66204586,
      "epoch": 1.4373414510400813,
      "grad_norm": 5.03125,
      "learning_rate": 2.0121887189004713e-06,
      "loss": 1.57126503,
      "memory(GiB)": 117.38,
      "step": 56660,
      "train_speed(iter/s)": 1.637839
    },
    {
      "acc": 0.66267447,
      "epoch": 1.4374682902080163,
      "grad_norm": 4.90625,
      "learning_rate": 2.0113479745365033e-06,
      "loss": 1.53914337,
      "memory(GiB)": 117.38,
      "step": 56665,
      "train_speed(iter/s)": 1.637854
    },
    {
      "acc": 0.65045338,
      "epoch": 1.4375951293759512,
      "grad_norm": 6.25,
      "learning_rate": 2.0105073616267984e-06,
      "loss": 1.64538155,
      "memory(GiB)": 117.38,
      "step": 56670,
      "train_speed(iter/s)": 1.637869
    },
    {
      "acc": 0.65917873,
      "epoch": 1.4377219685438862,
      "grad_norm": 6.4375,
      "learning_rate": 2.0096668802083254e-06,
      "loss": 1.62476234,
      "memory(GiB)": 117.38,
      "step": 56675,
      "train_speed(iter/s)": 1.637884
    },
    {
      "acc": 0.65910206,
      "epoch": 1.4378488077118214,
      "grad_norm": 5.4375,
      "learning_rate": 2.0088265303180516e-06,
      "loss": 1.57755928,
      "memory(GiB)": 117.38,
      "step": 56680,
      "train_speed(iter/s)": 1.637898
    },
    {
      "acc": 0.64925337,
      "epoch": 1.4379756468797564,
      "grad_norm": 6.875,
      "learning_rate": 2.0079863119929434e-06,
      "loss": 1.61306782,
      "memory(GiB)": 117.38,
      "step": 56685,
      "train_speed(iter/s)": 1.637914
    },
    {
      "acc": 0.654246,
      "epoch": 1.4381024860476916,
      "grad_norm": 5.59375,
      "learning_rate": 2.0071462252699575e-06,
      "loss": 1.65449734,
      "memory(GiB)": 117.38,
      "step": 56690,
      "train_speed(iter/s)": 1.637929
    },
    {
      "acc": 0.64911981,
      "epoch": 1.4382293252156266,
      "grad_norm": 5.5,
      "learning_rate": 2.006306270186039e-06,
      "loss": 1.58105955,
      "memory(GiB)": 117.38,
      "step": 56695,
      "train_speed(iter/s)": 1.637945
    },
    {
      "acc": 0.65274992,
      "epoch": 1.4383561643835616,
      "grad_norm": 5.90625,
      "learning_rate": 2.005466446778139e-06,
      "loss": 1.59600353,
      "memory(GiB)": 117.38,
      "step": 56700,
      "train_speed(iter/s)": 1.63796
    },
    {
      "acc": 0.64707499,
      "epoch": 1.4384830035514966,
      "grad_norm": 5.78125,
      "learning_rate": 2.0046267550831935e-06,
      "loss": 1.66900558,
      "memory(GiB)": 117.38,
      "step": 56705,
      "train_speed(iter/s)": 1.637977
    },
    {
      "acc": 0.65609694,
      "epoch": 1.4386098427194318,
      "grad_norm": 5.25,
      "learning_rate": 2.003787195138139e-06,
      "loss": 1.63397541,
      "memory(GiB)": 117.38,
      "step": 56710,
      "train_speed(iter/s)": 1.637992
    },
    {
      "acc": 0.65700374,
      "epoch": 1.4387366818873668,
      "grad_norm": 5.3125,
      "learning_rate": 2.002947766979897e-06,
      "loss": 1.63378906,
      "memory(GiB)": 117.38,
      "step": 56715,
      "train_speed(iter/s)": 1.638007
    },
    {
      "acc": 0.65730853,
      "epoch": 1.4388635210553018,
      "grad_norm": 5.28125,
      "learning_rate": 2.0021084706453945e-06,
      "loss": 1.55307751,
      "memory(GiB)": 117.38,
      "step": 56720,
      "train_speed(iter/s)": 1.638022
    },
    {
      "acc": 0.64480267,
      "epoch": 1.438990360223237,
      "grad_norm": 5.25,
      "learning_rate": 2.0012693061715467e-06,
      "loss": 1.59397993,
      "memory(GiB)": 117.38,
      "step": 56725,
      "train_speed(iter/s)": 1.638037
    },
    {
      "acc": 0.67191734,
      "epoch": 1.439117199391172,
      "grad_norm": 5.96875,
      "learning_rate": 2.000430273595263e-06,
      "loss": 1.53846283,
      "memory(GiB)": 117.38,
      "step": 56730,
      "train_speed(iter/s)": 1.638052
    },
    {
      "acc": 0.64228039,
      "epoch": 1.439244038559107,
      "grad_norm": 5.84375,
      "learning_rate": 1.9995913729534477e-06,
      "loss": 1.60652103,
      "memory(GiB)": 117.38,
      "step": 56735,
      "train_speed(iter/s)": 1.638066
    },
    {
      "acc": 0.64356232,
      "epoch": 1.439370877727042,
      "grad_norm": 4.8125,
      "learning_rate": 1.9987526042830003e-06,
      "loss": 1.64345322,
      "memory(GiB)": 117.38,
      "step": 56740,
      "train_speed(iter/s)": 1.638082
    },
    {
      "acc": 0.66152668,
      "epoch": 1.4394977168949772,
      "grad_norm": 4.875,
      "learning_rate": 1.9979139676208124e-06,
      "loss": 1.565271,
      "memory(GiB)": 117.38,
      "step": 56745,
      "train_speed(iter/s)": 1.638096
    },
    {
      "acc": 0.6499929,
      "epoch": 1.4396245560629122,
      "grad_norm": 6.9375,
      "learning_rate": 1.9970754630037718e-06,
      "loss": 1.5653801,
      "memory(GiB)": 117.38,
      "step": 56750,
      "train_speed(iter/s)": 1.638111
    },
    {
      "acc": 0.66611886,
      "epoch": 1.4397513952308474,
      "grad_norm": 5.53125,
      "learning_rate": 1.9962370904687596e-06,
      "loss": 1.5523344,
      "memory(GiB)": 117.38,
      "step": 56755,
      "train_speed(iter/s)": 1.638126
    },
    {
      "acc": 0.64757495,
      "epoch": 1.4398782343987824,
      "grad_norm": 7.0,
      "learning_rate": 1.9953988500526506e-06,
      "loss": 1.66170254,
      "memory(GiB)": 117.38,
      "step": 56760,
      "train_speed(iter/s)": 1.638141
    },
    {
      "acc": 0.66140223,
      "epoch": 1.4400050735667174,
      "grad_norm": 5.5,
      "learning_rate": 1.994560741792315e-06,
      "loss": 1.56403284,
      "memory(GiB)": 117.38,
      "step": 56765,
      "train_speed(iter/s)": 1.638157
    },
    {
      "acc": 0.65791788,
      "epoch": 1.4401319127346524,
      "grad_norm": 5.9375,
      "learning_rate": 1.993722765724616e-06,
      "loss": 1.59685392,
      "memory(GiB)": 117.38,
      "step": 56770,
      "train_speed(iter/s)": 1.638172
    },
    {
      "acc": 0.65203524,
      "epoch": 1.4402587519025876,
      "grad_norm": 5.78125,
      "learning_rate": 1.992884921886412e-06,
      "loss": 1.63024254,
      "memory(GiB)": 117.38,
      "step": 56775,
      "train_speed(iter/s)": 1.638187
    },
    {
      "acc": 0.65303907,
      "epoch": 1.4403855910705226,
      "grad_norm": 5.5625,
      "learning_rate": 1.9920472103145555e-06,
      "loss": 1.65169296,
      "memory(GiB)": 117.38,
      "step": 56780,
      "train_speed(iter/s)": 1.638202
    },
    {
      "acc": 0.65772057,
      "epoch": 1.4405124302384578,
      "grad_norm": 5.21875,
      "learning_rate": 1.99120963104589e-06,
      "loss": 1.60849762,
      "memory(GiB)": 117.38,
      "step": 56785,
      "train_speed(iter/s)": 1.638217
    },
    {
      "acc": 0.65142798,
      "epoch": 1.4406392694063928,
      "grad_norm": 6.71875,
      "learning_rate": 1.990372184117262e-06,
      "loss": 1.62114925,
      "memory(GiB)": 117.38,
      "step": 56790,
      "train_speed(iter/s)": 1.638232
    },
    {
      "acc": 0.64926682,
      "epoch": 1.4407661085743277,
      "grad_norm": 6.15625,
      "learning_rate": 1.9895348695655e-06,
      "loss": 1.60237045,
      "memory(GiB)": 117.38,
      "step": 56795,
      "train_speed(iter/s)": 1.638247
    },
    {
      "acc": 0.67513599,
      "epoch": 1.4408929477422627,
      "grad_norm": 5.34375,
      "learning_rate": 1.9886976874274356e-06,
      "loss": 1.55294781,
      "memory(GiB)": 117.38,
      "step": 56800,
      "train_speed(iter/s)": 1.638262
    },
    {
      "acc": 0.66617613,
      "epoch": 1.441019786910198,
      "grad_norm": 5.875,
      "learning_rate": 1.9878606377398895e-06,
      "loss": 1.52820234,
      "memory(GiB)": 117.38,
      "step": 56805,
      "train_speed(iter/s)": 1.638278
    },
    {
      "acc": 0.65975051,
      "epoch": 1.441146626078133,
      "grad_norm": 4.375,
      "learning_rate": 1.9870237205396844e-06,
      "loss": 1.58299789,
      "memory(GiB)": 117.38,
      "step": 56810,
      "train_speed(iter/s)": 1.638293
    },
    {
      "acc": 0.65558829,
      "epoch": 1.441273465246068,
      "grad_norm": 5.71875,
      "learning_rate": 1.986186935863626e-06,
      "loss": 1.63474922,
      "memory(GiB)": 117.38,
      "step": 56815,
      "train_speed(iter/s)": 1.638308
    },
    {
      "acc": 0.65376973,
      "epoch": 1.4414003044140031,
      "grad_norm": 5.78125,
      "learning_rate": 1.9853502837485207e-06,
      "loss": 1.60126038,
      "memory(GiB)": 117.38,
      "step": 56820,
      "train_speed(iter/s)": 1.638324
    },
    {
      "acc": 0.64889507,
      "epoch": 1.4415271435819381,
      "grad_norm": 5.5,
      "learning_rate": 1.9845137642311707e-06,
      "loss": 1.59639645,
      "memory(GiB)": 117.38,
      "step": 56825,
      "train_speed(iter/s)": 1.638339
    },
    {
      "acc": 0.66802053,
      "epoch": 1.441653982749873,
      "grad_norm": 6.28125,
      "learning_rate": 1.9836773773483704e-06,
      "loss": 1.58461447,
      "memory(GiB)": 117.38,
      "step": 56830,
      "train_speed(iter/s)": 1.638355
    },
    {
      "acc": 0.64627705,
      "epoch": 1.441780821917808,
      "grad_norm": 5.65625,
      "learning_rate": 1.982841123136904e-06,
      "loss": 1.64518509,
      "memory(GiB)": 117.38,
      "step": 56835,
      "train_speed(iter/s)": 1.63837
    },
    {
      "acc": 0.645298,
      "epoch": 1.4419076610857433,
      "grad_norm": 5.40625,
      "learning_rate": 1.982005001633554e-06,
      "loss": 1.62318802,
      "memory(GiB)": 117.38,
      "step": 56840,
      "train_speed(iter/s)": 1.638385
    },
    {
      "acc": 0.63613777,
      "epoch": 1.4420345002536783,
      "grad_norm": 5.6875,
      "learning_rate": 1.9811690128751002e-06,
      "loss": 1.7154295,
      "memory(GiB)": 117.38,
      "step": 56845,
      "train_speed(iter/s)": 1.6384
    },
    {
      "acc": 0.65435224,
      "epoch": 1.4421613394216135,
      "grad_norm": 5.25,
      "learning_rate": 1.980333156898313e-06,
      "loss": 1.57885752,
      "memory(GiB)": 117.38,
      "step": 56850,
      "train_speed(iter/s)": 1.638416
    },
    {
      "acc": 0.64419703,
      "epoch": 1.4422881785895485,
      "grad_norm": 5.46875,
      "learning_rate": 1.979497433739952e-06,
      "loss": 1.63819427,
      "memory(GiB)": 117.38,
      "step": 56855,
      "train_speed(iter/s)": 1.638432
    },
    {
      "acc": 0.66411519,
      "epoch": 1.4424150177574835,
      "grad_norm": 5.96875,
      "learning_rate": 1.9786618434367814e-06,
      "loss": 1.55240612,
      "memory(GiB)": 117.38,
      "step": 56860,
      "train_speed(iter/s)": 1.638448
    },
    {
      "acc": 0.65778127,
      "epoch": 1.4425418569254185,
      "grad_norm": 6.09375,
      "learning_rate": 1.977826386025552e-06,
      "loss": 1.59033175,
      "memory(GiB)": 117.38,
      "step": 56865,
      "train_speed(iter/s)": 1.638464
    },
    {
      "acc": 0.65405087,
      "epoch": 1.4426686960933537,
      "grad_norm": 5.25,
      "learning_rate": 1.976991061543011e-06,
      "loss": 1.59910669,
      "memory(GiB)": 117.38,
      "step": 56870,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.64710789,
      "epoch": 1.4427955352612887,
      "grad_norm": 6.15625,
      "learning_rate": 1.9761558700259e-06,
      "loss": 1.62930298,
      "memory(GiB)": 117.38,
      "step": 56875,
      "train_speed(iter/s)": 1.638494
    },
    {
      "acc": 0.65761924,
      "epoch": 1.4429223744292237,
      "grad_norm": 5.90625,
      "learning_rate": 1.9753208115109546e-06,
      "loss": 1.56891394,
      "memory(GiB)": 117.38,
      "step": 56880,
      "train_speed(iter/s)": 1.638508
    },
    {
      "acc": 0.65390573,
      "epoch": 1.4430492135971589,
      "grad_norm": 5.59375,
      "learning_rate": 1.9744858860349043e-06,
      "loss": 1.59684925,
      "memory(GiB)": 117.38,
      "step": 56885,
      "train_speed(iter/s)": 1.638524
    },
    {
      "acc": 0.66021557,
      "epoch": 1.4431760527650939,
      "grad_norm": 6.1875,
      "learning_rate": 1.9736510936344723e-06,
      "loss": 1.60303688,
      "memory(GiB)": 117.38,
      "step": 56890,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.64649925,
      "epoch": 1.4433028919330289,
      "grad_norm": 5.59375,
      "learning_rate": 1.9728164343463764e-06,
      "loss": 1.66878395,
      "memory(GiB)": 117.38,
      "step": 56895,
      "train_speed(iter/s)": 1.638322
    },
    {
      "acc": 0.67600212,
      "epoch": 1.4434297311009638,
      "grad_norm": 7.90625,
      "learning_rate": 1.97198190820733e-06,
      "loss": 1.57320032,
      "memory(GiB)": 117.38,
      "step": 56900,
      "train_speed(iter/s)": 1.638338
    },
    {
      "acc": 0.64314008,
      "epoch": 1.443556570268899,
      "grad_norm": 4.84375,
      "learning_rate": 1.9711475152540376e-06,
      "loss": 1.65345192,
      "memory(GiB)": 117.38,
      "step": 56905,
      "train_speed(iter/s)": 1.638354
    },
    {
      "acc": 0.65243139,
      "epoch": 1.443683409436834,
      "grad_norm": 5.5625,
      "learning_rate": 1.9703132555232007e-06,
      "loss": 1.66248093,
      "memory(GiB)": 117.38,
      "step": 56910,
      "train_speed(iter/s)": 1.63837
    },
    {
      "acc": 0.67233706,
      "epoch": 1.4438102486047693,
      "grad_norm": 5.4375,
      "learning_rate": 1.9694791290515135e-06,
      "loss": 1.57454777,
      "memory(GiB)": 117.38,
      "step": 56915,
      "train_speed(iter/s)": 1.638386
    },
    {
      "acc": 0.66005259,
      "epoch": 1.4439370877727042,
      "grad_norm": 6.78125,
      "learning_rate": 1.968645135875665e-06,
      "loss": 1.60908051,
      "memory(GiB)": 117.38,
      "step": 56920,
      "train_speed(iter/s)": 1.638402
    },
    {
      "acc": 0.65143685,
      "epoch": 1.4440639269406392,
      "grad_norm": 5.25,
      "learning_rate": 1.967811276032335e-06,
      "loss": 1.58872051,
      "memory(GiB)": 117.38,
      "step": 56925,
      "train_speed(iter/s)": 1.638418
    },
    {
      "acc": 0.64992266,
      "epoch": 1.4441907661085742,
      "grad_norm": 5.5625,
      "learning_rate": 1.966977549558206e-06,
      "loss": 1.64243546,
      "memory(GiB)": 117.38,
      "step": 56930,
      "train_speed(iter/s)": 1.638435
    },
    {
      "acc": 0.67083831,
      "epoch": 1.4443176052765094,
      "grad_norm": 5.03125,
      "learning_rate": 1.966143956489945e-06,
      "loss": 1.56842413,
      "memory(GiB)": 117.38,
      "step": 56935,
      "train_speed(iter/s)": 1.638449
    },
    {
      "acc": 0.67404747,
      "epoch": 1.4444444444444444,
      "grad_norm": 5.25,
      "learning_rate": 1.965310496864217e-06,
      "loss": 1.59623775,
      "memory(GiB)": 117.38,
      "step": 56940,
      "train_speed(iter/s)": 1.638464
    },
    {
      "acc": 0.66441669,
      "epoch": 1.4445712836123796,
      "grad_norm": 6.3125,
      "learning_rate": 1.9644771707176813e-06,
      "loss": 1.56829948,
      "memory(GiB)": 117.38,
      "step": 56945,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.64249125,
      "epoch": 1.4446981227803146,
      "grad_norm": 4.90625,
      "learning_rate": 1.963643978086996e-06,
      "loss": 1.60039444,
      "memory(GiB)": 117.38,
      "step": 56950,
      "train_speed(iter/s)": 1.638494
    },
    {
      "acc": 0.65368681,
      "epoch": 1.4448249619482496,
      "grad_norm": 6.125,
      "learning_rate": 1.9628109190088023e-06,
      "loss": 1.58377132,
      "memory(GiB)": 117.38,
      "step": 56955,
      "train_speed(iter/s)": 1.638509
    },
    {
      "acc": 0.65559635,
      "epoch": 1.4449518011161846,
      "grad_norm": 6.53125,
      "learning_rate": 1.961977993519743e-06,
      "loss": 1.64477673,
      "memory(GiB)": 117.38,
      "step": 56960,
      "train_speed(iter/s)": 1.638524
    },
    {
      "acc": 0.65625477,
      "epoch": 1.4450786402841198,
      "grad_norm": 5.53125,
      "learning_rate": 1.9611452016564574e-06,
      "loss": 1.54010897,
      "memory(GiB)": 117.38,
      "step": 56965,
      "train_speed(iter/s)": 1.638541
    },
    {
      "acc": 0.65136757,
      "epoch": 1.4452054794520548,
      "grad_norm": 5.84375,
      "learning_rate": 1.960312543455575e-06,
      "loss": 1.63736649,
      "memory(GiB)": 117.38,
      "step": 56970,
      "train_speed(iter/s)": 1.638556
    },
    {
      "acc": 0.6611475,
      "epoch": 1.4453323186199898,
      "grad_norm": 6.0625,
      "learning_rate": 1.959480018953716e-06,
      "loss": 1.61653004,
      "memory(GiB)": 117.38,
      "step": 56975,
      "train_speed(iter/s)": 1.638572
    },
    {
      "acc": 0.63838387,
      "epoch": 1.445459157787925,
      "grad_norm": 4.25,
      "learning_rate": 1.9586476281874994e-06,
      "loss": 1.64717407,
      "memory(GiB)": 117.38,
      "step": 56980,
      "train_speed(iter/s)": 1.638588
    },
    {
      "acc": 0.6426403,
      "epoch": 1.44558599695586,
      "grad_norm": 6.59375,
      "learning_rate": 1.9578153711935403e-06,
      "loss": 1.62657452,
      "memory(GiB)": 117.38,
      "step": 56985,
      "train_speed(iter/s)": 1.638604
    },
    {
      "acc": 0.64297981,
      "epoch": 1.445712836123795,
      "grad_norm": 6.375,
      "learning_rate": 1.9569832480084456e-06,
      "loss": 1.63859196,
      "memory(GiB)": 117.38,
      "step": 56990,
      "train_speed(iter/s)": 1.63862
    },
    {
      "acc": 0.64794579,
      "epoch": 1.44583967529173,
      "grad_norm": 7.96875,
      "learning_rate": 1.9561512586688096e-06,
      "loss": 1.67420025,
      "memory(GiB)": 117.38,
      "step": 56995,
      "train_speed(iter/s)": 1.638635
    },
    {
      "acc": 0.65910435,
      "epoch": 1.4459665144596652,
      "grad_norm": 5.65625,
      "learning_rate": 1.9553194032112334e-06,
      "loss": 1.56917953,
      "memory(GiB)": 117.38,
      "step": 57000,
      "train_speed(iter/s)": 1.638651
    },
    {
      "epoch": 1.4459665144596652,
      "eval_acc": 0.6462751928028618,
      "eval_loss": 1.5732783079147339,
      "eval_runtime": 58.8551,
      "eval_samples_per_second": 108.232,
      "eval_steps_per_second": 27.066,
      "step": 57000
    },
    {
      "acc": 0.65983486,
      "epoch": 1.4460933536276002,
      "grad_norm": 10.0625,
      "learning_rate": 1.954487681672303e-06,
      "loss": 1.63470917,
      "memory(GiB)": 117.38,
      "step": 57005,
      "train_speed(iter/s)": 1.635694
    },
    {
      "acc": 0.66568861,
      "epoch": 1.4462201927955354,
      "grad_norm": 7.96875,
      "learning_rate": 1.9536560940886033e-06,
      "loss": 1.5311286,
      "memory(GiB)": 117.38,
      "step": 57010,
      "train_speed(iter/s)": 1.635711
    },
    {
      "acc": 0.63711953,
      "epoch": 1.4463470319634704,
      "grad_norm": 5.625,
      "learning_rate": 1.9528246404967067e-06,
      "loss": 1.67905521,
      "memory(GiB)": 117.38,
      "step": 57015,
      "train_speed(iter/s)": 1.635726
    },
    {
      "acc": 0.65817933,
      "epoch": 1.4464738711314054,
      "grad_norm": 6.0,
      "learning_rate": 1.951993320933188e-06,
      "loss": 1.62685814,
      "memory(GiB)": 117.38,
      "step": 57020,
      "train_speed(iter/s)": 1.635742
    },
    {
      "acc": 0.6541234,
      "epoch": 1.4466007102993403,
      "grad_norm": 6.78125,
      "learning_rate": 1.951162135434612e-06,
      "loss": 1.66277428,
      "memory(GiB)": 117.38,
      "step": 57025,
      "train_speed(iter/s)": 1.635757
    },
    {
      "acc": 0.66161323,
      "epoch": 1.4467275494672756,
      "grad_norm": 5.6875,
      "learning_rate": 1.9503310840375374e-06,
      "loss": 1.62241707,
      "memory(GiB)": 117.38,
      "step": 57030,
      "train_speed(iter/s)": 1.635774
    },
    {
      "acc": 0.65783582,
      "epoch": 1.4468543886352105,
      "grad_norm": 6.0625,
      "learning_rate": 1.949500166778517e-06,
      "loss": 1.67539177,
      "memory(GiB)": 117.38,
      "step": 57035,
      "train_speed(iter/s)": 1.63579
    },
    {
      "acc": 0.6523447,
      "epoch": 1.4469812278031455,
      "grad_norm": 5.625,
      "learning_rate": 1.948669383694099e-06,
      "loss": 1.6104454,
      "memory(GiB)": 117.38,
      "step": 57040,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.66037321,
      "epoch": 1.4471080669710807,
      "grad_norm": 6.0,
      "learning_rate": 1.947838734820825e-06,
      "loss": 1.54887028,
      "memory(GiB)": 117.38,
      "step": 57045,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.65758562,
      "epoch": 1.4472349061390157,
      "grad_norm": 5.96875,
      "learning_rate": 1.94700822019523e-06,
      "loss": 1.566259,
      "memory(GiB)": 117.38,
      "step": 57050,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.67645731,
      "epoch": 1.4473617453069507,
      "grad_norm": 7.28125,
      "learning_rate": 1.9461778398538447e-06,
      "loss": 1.52994108,
      "memory(GiB)": 117.38,
      "step": 57055,
      "train_speed(iter/s)": 1.635848
    },
    {
      "acc": 0.62973032,
      "epoch": 1.4474885844748857,
      "grad_norm": 5.75,
      "learning_rate": 1.945347593833191e-06,
      "loss": 1.73797112,
      "memory(GiB)": 117.38,
      "step": 57060,
      "train_speed(iter/s)": 1.635864
    },
    {
      "acc": 0.64589596,
      "epoch": 1.447615423642821,
      "grad_norm": 6.375,
      "learning_rate": 1.9445174821697893e-06,
      "loss": 1.65696621,
      "memory(GiB)": 117.38,
      "step": 57065,
      "train_speed(iter/s)": 1.635879
    },
    {
      "acc": 0.62811861,
      "epoch": 1.447742262810756,
      "grad_norm": 5.125,
      "learning_rate": 1.94368750490015e-06,
      "loss": 1.6904026,
      "memory(GiB)": 117.38,
      "step": 57070,
      "train_speed(iter/s)": 1.635895
    },
    {
      "acc": 0.67131414,
      "epoch": 1.4478691019786911,
      "grad_norm": 6.8125,
      "learning_rate": 1.94285766206078e-06,
      "loss": 1.53685827,
      "memory(GiB)": 117.38,
      "step": 57075,
      "train_speed(iter/s)": 1.635911
    },
    {
      "acc": 0.657477,
      "epoch": 1.447995941146626,
      "grad_norm": 6.90625,
      "learning_rate": 1.9420279536881794e-06,
      "loss": 1.64505463,
      "memory(GiB)": 117.38,
      "step": 57080,
      "train_speed(iter/s)": 1.635927
    },
    {
      "acc": 0.65891447,
      "epoch": 1.448122780314561,
      "grad_norm": 5.0625,
      "learning_rate": 1.9411983798188398e-06,
      "loss": 1.64063225,
      "memory(GiB)": 117.38,
      "step": 57085,
      "train_speed(iter/s)": 1.635942
    },
    {
      "acc": 0.64775205,
      "epoch": 1.448249619482496,
      "grad_norm": 5.96875,
      "learning_rate": 1.940368940489256e-06,
      "loss": 1.64184151,
      "memory(GiB)": 117.38,
      "step": 57090,
      "train_speed(iter/s)": 1.635958
    },
    {
      "acc": 0.66083183,
      "epoch": 1.4483764586504313,
      "grad_norm": 6.46875,
      "learning_rate": 1.939539635735905e-06,
      "loss": 1.57934237,
      "memory(GiB)": 117.38,
      "step": 57095,
      "train_speed(iter/s)": 1.635973
    },
    {
      "acc": 0.67137265,
      "epoch": 1.4485032978183663,
      "grad_norm": 6.5,
      "learning_rate": 1.9387104655952625e-06,
      "loss": 1.58787498,
      "memory(GiB)": 117.38,
      "step": 57100,
      "train_speed(iter/s)": 1.635989
    },
    {
      "acc": 0.65753026,
      "epoch": 1.4486301369863015,
      "grad_norm": 6.0,
      "learning_rate": 1.9378814301038033e-06,
      "loss": 1.55950699,
      "memory(GiB)": 117.38,
      "step": 57105,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.65536742,
      "epoch": 1.4487569761542365,
      "grad_norm": 6.03125,
      "learning_rate": 1.937052529297992e-06,
      "loss": 1.59135695,
      "memory(GiB)": 117.38,
      "step": 57110,
      "train_speed(iter/s)": 1.636019
    },
    {
      "acc": 0.67116528,
      "epoch": 1.4488838153221715,
      "grad_norm": 6.28125,
      "learning_rate": 1.9362237632142838e-06,
      "loss": 1.55093994,
      "memory(GiB)": 117.38,
      "step": 57115,
      "train_speed(iter/s)": 1.636035
    },
    {
      "acc": 0.67192211,
      "epoch": 1.4490106544901065,
      "grad_norm": 5.4375,
      "learning_rate": 1.9353951318891313e-06,
      "loss": 1.52644215,
      "memory(GiB)": 117.38,
      "step": 57120,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.66900139,
      "epoch": 1.4491374936580417,
      "grad_norm": 6.75,
      "learning_rate": 1.9345666353589855e-06,
      "loss": 1.5844758,
      "memory(GiB)": 117.38,
      "step": 57125,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.66390681,
      "epoch": 1.4492643328259767,
      "grad_norm": 6.5,
      "learning_rate": 1.9337382736602868e-06,
      "loss": 1.57343369,
      "memory(GiB)": 117.38,
      "step": 57130,
      "train_speed(iter/s)": 1.63608
    },
    {
      "acc": 0.65626774,
      "epoch": 1.4493911719939117,
      "grad_norm": 7.3125,
      "learning_rate": 1.9329100468294646e-06,
      "loss": 1.59688339,
      "memory(GiB)": 117.38,
      "step": 57135,
      "train_speed(iter/s)": 1.636095
    },
    {
      "acc": 0.66577759,
      "epoch": 1.4495180111618469,
      "grad_norm": 5.0,
      "learning_rate": 1.9320819549029546e-06,
      "loss": 1.54080753,
      "memory(GiB)": 117.38,
      "step": 57140,
      "train_speed(iter/s)": 1.636111
    },
    {
      "acc": 0.65928102,
      "epoch": 1.4496448503297819,
      "grad_norm": 5.46875,
      "learning_rate": 1.9312539979171774e-06,
      "loss": 1.5902153,
      "memory(GiB)": 117.38,
      "step": 57145,
      "train_speed(iter/s)": 1.636126
    },
    {
      "acc": 0.64968801,
      "epoch": 1.4497716894977168,
      "grad_norm": 4.8125,
      "learning_rate": 1.9304261759085525e-06,
      "loss": 1.57117939,
      "memory(GiB)": 117.38,
      "step": 57150,
      "train_speed(iter/s)": 1.636141
    },
    {
      "acc": 0.66583667,
      "epoch": 1.4498985286656518,
      "grad_norm": 6.21875,
      "learning_rate": 1.929598488913485e-06,
      "loss": 1.59090929,
      "memory(GiB)": 117.38,
      "step": 57155,
      "train_speed(iter/s)": 1.636157
    },
    {
      "acc": 0.66483445,
      "epoch": 1.450025367833587,
      "grad_norm": 4.78125,
      "learning_rate": 1.928770936968386e-06,
      "loss": 1.54363451,
      "memory(GiB)": 117.38,
      "step": 57160,
      "train_speed(iter/s)": 1.636172
    },
    {
      "acc": 0.67496901,
      "epoch": 1.450152207001522,
      "grad_norm": 5.09375,
      "learning_rate": 1.927943520109653e-06,
      "loss": 1.51304855,
      "memory(GiB)": 117.38,
      "step": 57165,
      "train_speed(iter/s)": 1.636189
    },
    {
      "acc": 0.65530767,
      "epoch": 1.4502790461694572,
      "grad_norm": 6.1875,
      "learning_rate": 1.9271162383736804e-06,
      "loss": 1.65991783,
      "memory(GiB)": 117.38,
      "step": 57170,
      "train_speed(iter/s)": 1.636205
    },
    {
      "acc": 0.66496162,
      "epoch": 1.4504058853373922,
      "grad_norm": 6.3125,
      "learning_rate": 1.9262890917968547e-06,
      "loss": 1.58260727,
      "memory(GiB)": 117.38,
      "step": 57175,
      "train_speed(iter/s)": 1.63622
    },
    {
      "acc": 0.64246774,
      "epoch": 1.4505327245053272,
      "grad_norm": 6.0625,
      "learning_rate": 1.925462080415558e-06,
      "loss": 1.72915134,
      "memory(GiB)": 117.38,
      "step": 57180,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.64967732,
      "epoch": 1.4506595636732622,
      "grad_norm": 5.90625,
      "learning_rate": 1.924635204266166e-06,
      "loss": 1.63484974,
      "memory(GiB)": 117.38,
      "step": 57185,
      "train_speed(iter/s)": 1.636251
    },
    {
      "acc": 0.66788197,
      "epoch": 1.4507864028411974,
      "grad_norm": 4.71875,
      "learning_rate": 1.923808463385048e-06,
      "loss": 1.56585808,
      "memory(GiB)": 117.38,
      "step": 57190,
      "train_speed(iter/s)": 1.636266
    },
    {
      "acc": 0.64836664,
      "epoch": 1.4509132420091324,
      "grad_norm": 5.15625,
      "learning_rate": 1.922981857808568e-06,
      "loss": 1.63585682,
      "memory(GiB)": 117.38,
      "step": 57195,
      "train_speed(iter/s)": 1.636282
    },
    {
      "acc": 0.64952722,
      "epoch": 1.4510400811770674,
      "grad_norm": 4.6875,
      "learning_rate": 1.9221553875730835e-06,
      "loss": 1.63347435,
      "memory(GiB)": 117.38,
      "step": 57200,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.6486619,
      "epoch": 1.4511669203450026,
      "grad_norm": 5.3125,
      "learning_rate": 1.921329052714947e-06,
      "loss": 1.60259781,
      "memory(GiB)": 117.38,
      "step": 57205,
      "train_speed(iter/s)": 1.636314
    },
    {
      "acc": 0.66280117,
      "epoch": 1.4512937595129376,
      "grad_norm": 6.34375,
      "learning_rate": 1.920502853270504e-06,
      "loss": 1.57172356,
      "memory(GiB)": 117.38,
      "step": 57210,
      "train_speed(iter/s)": 1.63633
    },
    {
      "acc": 0.65556793,
      "epoch": 1.4514205986808726,
      "grad_norm": 5.1875,
      "learning_rate": 1.919676789276094e-06,
      "loss": 1.63801041,
      "memory(GiB)": 117.38,
      "step": 57215,
      "train_speed(iter/s)": 1.636346
    },
    {
      "acc": 0.65053039,
      "epoch": 1.4515474378488076,
      "grad_norm": 6.71875,
      "learning_rate": 1.918850860768052e-06,
      "loss": 1.67173576,
      "memory(GiB)": 117.38,
      "step": 57220,
      "train_speed(iter/s)": 1.636361
    },
    {
      "acc": 0.65983396,
      "epoch": 1.4516742770167428,
      "grad_norm": 6.3125,
      "learning_rate": 1.918025067782704e-06,
      "loss": 1.57795067,
      "memory(GiB)": 117.38,
      "step": 57225,
      "train_speed(iter/s)": 1.636377
    },
    {
      "acc": 0.6572083,
      "epoch": 1.4518011161846778,
      "grad_norm": 5.59375,
      "learning_rate": 1.9171994103563766e-06,
      "loss": 1.63053951,
      "memory(GiB)": 117.38,
      "step": 57230,
      "train_speed(iter/s)": 1.636392
    },
    {
      "acc": 0.63794103,
      "epoch": 1.451927955352613,
      "grad_norm": 5.84375,
      "learning_rate": 1.916373888525381e-06,
      "loss": 1.71828842,
      "memory(GiB)": 117.38,
      "step": 57235,
      "train_speed(iter/s)": 1.636407
    },
    {
      "acc": 0.63756962,
      "epoch": 1.452054794520548,
      "grad_norm": 6.3125,
      "learning_rate": 1.9155485023260294e-06,
      "loss": 1.65989151,
      "memory(GiB)": 117.38,
      "step": 57240,
      "train_speed(iter/s)": 1.636423
    },
    {
      "acc": 0.64877052,
      "epoch": 1.452181633688483,
      "grad_norm": 5.21875,
      "learning_rate": 1.914723251794624e-06,
      "loss": 1.57665539,
      "memory(GiB)": 117.38,
      "step": 57245,
      "train_speed(iter/s)": 1.636437
    },
    {
      "acc": 0.66091881,
      "epoch": 1.452308472856418,
      "grad_norm": 6.0,
      "learning_rate": 1.9138981369674688e-06,
      "loss": 1.64759636,
      "memory(GiB)": 117.38,
      "step": 57250,
      "train_speed(iter/s)": 1.636453
    },
    {
      "acc": 0.64368081,
      "epoch": 1.4524353120243532,
      "grad_norm": 5.25,
      "learning_rate": 1.9130731578808493e-06,
      "loss": 1.63277321,
      "memory(GiB)": 117.38,
      "step": 57255,
      "train_speed(iter/s)": 1.636469
    },
    {
      "acc": 0.64925041,
      "epoch": 1.4525621511922882,
      "grad_norm": 5.5,
      "learning_rate": 1.912248314571053e-06,
      "loss": 1.63647041,
      "memory(GiB)": 117.38,
      "step": 57260,
      "train_speed(iter/s)": 1.636484
    },
    {
      "acc": 0.6674449,
      "epoch": 1.4526889903602234,
      "grad_norm": 6.375,
      "learning_rate": 1.9114236070743638e-06,
      "loss": 1.58786201,
      "memory(GiB)": 117.38,
      "step": 57265,
      "train_speed(iter/s)": 1.636499
    },
    {
      "acc": 0.66013389,
      "epoch": 1.4528158295281584,
      "grad_norm": 5.46875,
      "learning_rate": 1.910599035427055e-06,
      "loss": 1.60119324,
      "memory(GiB)": 117.38,
      "step": 57270,
      "train_speed(iter/s)": 1.636515
    },
    {
      "acc": 0.6530014,
      "epoch": 1.4529426686960933,
      "grad_norm": 7.125,
      "learning_rate": 1.909774599665392e-06,
      "loss": 1.55767431,
      "memory(GiB)": 117.38,
      "step": 57275,
      "train_speed(iter/s)": 1.636531
    },
    {
      "acc": 0.66979566,
      "epoch": 1.4530695078640283,
      "grad_norm": 7.25,
      "learning_rate": 1.9089502998256382e-06,
      "loss": 1.53461914,
      "memory(GiB)": 117.38,
      "step": 57280,
      "train_speed(iter/s)": 1.636546
    },
    {
      "acc": 0.65918865,
      "epoch": 1.4531963470319635,
      "grad_norm": 6.71875,
      "learning_rate": 1.9081261359440517e-06,
      "loss": 1.58309984,
      "memory(GiB)": 117.38,
      "step": 57285,
      "train_speed(iter/s)": 1.636561
    },
    {
      "acc": 0.67190862,
      "epoch": 1.4533231861998985,
      "grad_norm": 6.3125,
      "learning_rate": 1.9073021080568837e-06,
      "loss": 1.55598526,
      "memory(GiB)": 117.38,
      "step": 57290,
      "train_speed(iter/s)": 1.636577
    },
    {
      "acc": 0.65918016,
      "epoch": 1.4534500253678335,
      "grad_norm": 6.5,
      "learning_rate": 1.9064782162003737e-06,
      "loss": 1.55874691,
      "memory(GiB)": 117.38,
      "step": 57295,
      "train_speed(iter/s)": 1.636594
    },
    {
      "acc": 0.65292559,
      "epoch": 1.4535768645357687,
      "grad_norm": 5.21875,
      "learning_rate": 1.9056544604107646e-06,
      "loss": 1.56381721,
      "memory(GiB)": 117.38,
      "step": 57300,
      "train_speed(iter/s)": 1.636609
    },
    {
      "acc": 0.65367217,
      "epoch": 1.4537037037037037,
      "grad_norm": 6.28125,
      "learning_rate": 1.9048308407242882e-06,
      "loss": 1.59002361,
      "memory(GiB)": 117.38,
      "step": 57305,
      "train_speed(iter/s)": 1.636625
    },
    {
      "acc": 0.65160046,
      "epoch": 1.4538305428716387,
      "grad_norm": 6.96875,
      "learning_rate": 1.90400735717717e-06,
      "loss": 1.62826805,
      "memory(GiB)": 117.38,
      "step": 57310,
      "train_speed(iter/s)": 1.63664
    },
    {
      "acc": 0.64498286,
      "epoch": 1.4539573820395737,
      "grad_norm": 5.90625,
      "learning_rate": 1.903184009805631e-06,
      "loss": 1.57962074,
      "memory(GiB)": 117.38,
      "step": 57315,
      "train_speed(iter/s)": 1.636655
    },
    {
      "acc": 0.64771957,
      "epoch": 1.454084221207509,
      "grad_norm": 5.25,
      "learning_rate": 1.9023607986458854e-06,
      "loss": 1.70104218,
      "memory(GiB)": 117.38,
      "step": 57320,
      "train_speed(iter/s)": 1.636671
    },
    {
      "acc": 0.66043119,
      "epoch": 1.454211060375444,
      "grad_norm": 5.4375,
      "learning_rate": 1.901537723734142e-06,
      "loss": 1.5954442,
      "memory(GiB)": 117.38,
      "step": 57325,
      "train_speed(iter/s)": 1.636687
    },
    {
      "acc": 0.66115427,
      "epoch": 1.454337899543379,
      "grad_norm": 4.84375,
      "learning_rate": 1.9007147851066031e-06,
      "loss": 1.56853418,
      "memory(GiB)": 117.38,
      "step": 57330,
      "train_speed(iter/s)": 1.636702
    },
    {
      "acc": 0.67260337,
      "epoch": 1.454464738711314,
      "grad_norm": 5.625,
      "learning_rate": 1.8998919827994654e-06,
      "loss": 1.49298887,
      "memory(GiB)": 117.38,
      "step": 57335,
      "train_speed(iter/s)": 1.636718
    },
    {
      "acc": 0.65983915,
      "epoch": 1.454591577879249,
      "grad_norm": 5.8125,
      "learning_rate": 1.899069316848919e-06,
      "loss": 1.64879665,
      "memory(GiB)": 117.38,
      "step": 57340,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.66017113,
      "epoch": 1.454718417047184,
      "grad_norm": 5.0,
      "learning_rate": 1.8982467872911486e-06,
      "loss": 1.68707008,
      "memory(GiB)": 117.38,
      "step": 57345,
      "train_speed(iter/s)": 1.636749
    },
    {
      "acc": 0.65039816,
      "epoch": 1.4548452562151193,
      "grad_norm": 5.15625,
      "learning_rate": 1.8974243941623332e-06,
      "loss": 1.61987057,
      "memory(GiB)": 117.38,
      "step": 57350,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.64611359,
      "epoch": 1.4549720953830543,
      "grad_norm": 6.59375,
      "learning_rate": 1.896602137498645e-06,
      "loss": 1.68173237,
      "memory(GiB)": 117.38,
      "step": 57355,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.65224314,
      "epoch": 1.4550989345509893,
      "grad_norm": 5.75,
      "learning_rate": 1.89578001733625e-06,
      "loss": 1.60310459,
      "memory(GiB)": 117.38,
      "step": 57360,
      "train_speed(iter/s)": 1.636796
    },
    {
      "acc": 0.66426563,
      "epoch": 1.4552257737189245,
      "grad_norm": 5.65625,
      "learning_rate": 1.8949580337113078e-06,
      "loss": 1.56433716,
      "memory(GiB)": 117.38,
      "step": 57365,
      "train_speed(iter/s)": 1.636811
    },
    {
      "acc": 0.64145288,
      "epoch": 1.4553526128868595,
      "grad_norm": 6.40625,
      "learning_rate": 1.8941361866599778e-06,
      "loss": 1.66888504,
      "memory(GiB)": 117.38,
      "step": 57370,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.64828243,
      "epoch": 1.4554794520547945,
      "grad_norm": 5.125,
      "learning_rate": 1.893314476218403e-06,
      "loss": 1.638554,
      "memory(GiB)": 117.38,
      "step": 57375,
      "train_speed(iter/s)": 1.636842
    },
    {
      "acc": 0.65154319,
      "epoch": 1.4556062912227294,
      "grad_norm": 5.0,
      "learning_rate": 1.8924929024227279e-06,
      "loss": 1.60959511,
      "memory(GiB)": 117.38,
      "step": 57380,
      "train_speed(iter/s)": 1.636858
    },
    {
      "acc": 0.64478154,
      "epoch": 1.4557331303906647,
      "grad_norm": 6.0625,
      "learning_rate": 1.8916714653090874e-06,
      "loss": 1.65695953,
      "memory(GiB)": 117.38,
      "step": 57385,
      "train_speed(iter/s)": 1.636873
    },
    {
      "acc": 0.65126963,
      "epoch": 1.4558599695585996,
      "grad_norm": 5.09375,
      "learning_rate": 1.8908501649136174e-06,
      "loss": 1.6137291,
      "memory(GiB)": 117.38,
      "step": 57390,
      "train_speed(iter/s)": 1.636889
    },
    {
      "acc": 0.6604413,
      "epoch": 1.4559868087265349,
      "grad_norm": 5.625,
      "learning_rate": 1.8900290012724358e-06,
      "loss": 1.57032309,
      "memory(GiB)": 117.38,
      "step": 57395,
      "train_speed(iter/s)": 1.636905
    },
    {
      "acc": 0.66509604,
      "epoch": 1.4561136478944698,
      "grad_norm": 6.125,
      "learning_rate": 1.889207974421663e-06,
      "loss": 1.53082504,
      "memory(GiB)": 117.38,
      "step": 57400,
      "train_speed(iter/s)": 1.63692
    },
    {
      "acc": 0.67246957,
      "epoch": 1.4562404870624048,
      "grad_norm": 14.0,
      "learning_rate": 1.8883870843974134e-06,
      "loss": 1.51925716,
      "memory(GiB)": 117.38,
      "step": 57405,
      "train_speed(iter/s)": 1.636935
    },
    {
      "acc": 0.63677759,
      "epoch": 1.4563673262303398,
      "grad_norm": 6.0,
      "learning_rate": 1.887566331235794e-06,
      "loss": 1.71867752,
      "memory(GiB)": 117.38,
      "step": 57410,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.64237099,
      "epoch": 1.456494165398275,
      "grad_norm": 7.34375,
      "learning_rate": 1.8867457149729013e-06,
      "loss": 1.74663906,
      "memory(GiB)": 117.38,
      "step": 57415,
      "train_speed(iter/s)": 1.636967
    },
    {
      "acc": 0.65436869,
      "epoch": 1.45662100456621,
      "grad_norm": 5.34375,
      "learning_rate": 1.8859252356448305e-06,
      "loss": 1.60367146,
      "memory(GiB)": 117.38,
      "step": 57420,
      "train_speed(iter/s)": 1.636982
    },
    {
      "acc": 0.66943574,
      "epoch": 1.4567478437341452,
      "grad_norm": 6.625,
      "learning_rate": 1.8851048932876725e-06,
      "loss": 1.52174606,
      "memory(GiB)": 117.38,
      "step": 57425,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.67208447,
      "epoch": 1.4568746829020802,
      "grad_norm": 5.59375,
      "learning_rate": 1.8842846879375104e-06,
      "loss": 1.5853405,
      "memory(GiB)": 117.38,
      "step": 57430,
      "train_speed(iter/s)": 1.637012
    },
    {
      "acc": 0.66352468,
      "epoch": 1.4570015220700152,
      "grad_norm": 6.5625,
      "learning_rate": 1.8834646196304146e-06,
      "loss": 1.58954268,
      "memory(GiB)": 117.38,
      "step": 57435,
      "train_speed(iter/s)": 1.637027
    },
    {
      "acc": 0.6562253,
      "epoch": 1.4571283612379502,
      "grad_norm": 6.59375,
      "learning_rate": 1.8826446884024612e-06,
      "loss": 1.68271179,
      "memory(GiB)": 117.38,
      "step": 57440,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.66146083,
      "epoch": 1.4572552004058854,
      "grad_norm": 4.90625,
      "learning_rate": 1.8818248942897122e-06,
      "loss": 1.5480011,
      "memory(GiB)": 117.38,
      "step": 57445,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.64915457,
      "epoch": 1.4573820395738204,
      "grad_norm": 11.625,
      "learning_rate": 1.8810052373282277e-06,
      "loss": 1.6108326,
      "memory(GiB)": 117.38,
      "step": 57450,
      "train_speed(iter/s)": 1.637074
    },
    {
      "acc": 0.65405035,
      "epoch": 1.4575088787417554,
      "grad_norm": 5.78125,
      "learning_rate": 1.880185717554055e-06,
      "loss": 1.62377968,
      "memory(GiB)": 117.38,
      "step": 57455,
      "train_speed(iter/s)": 1.637089
    },
    {
      "acc": 0.64473839,
      "epoch": 1.4576357179096906,
      "grad_norm": 5.625,
      "learning_rate": 1.879366335003245e-06,
      "loss": 1.64802246,
      "memory(GiB)": 117.38,
      "step": 57460,
      "train_speed(iter/s)": 1.637104
    },
    {
      "acc": 0.6480298,
      "epoch": 1.4577625570776256,
      "grad_norm": 6.53125,
      "learning_rate": 1.8785470897118362e-06,
      "loss": 1.64533787,
      "memory(GiB)": 117.38,
      "step": 57465,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.65634394,
      "epoch": 1.4578893962455606,
      "grad_norm": 5.9375,
      "learning_rate": 1.8777279817158627e-06,
      "loss": 1.59434662,
      "memory(GiB)": 117.38,
      "step": 57470,
      "train_speed(iter/s)": 1.637136
    },
    {
      "acc": 0.66031961,
      "epoch": 1.4580162354134956,
      "grad_norm": 5.15625,
      "learning_rate": 1.8769090110513522e-06,
      "loss": 1.59686947,
      "memory(GiB)": 117.38,
      "step": 57475,
      "train_speed(iter/s)": 1.637152
    },
    {
      "acc": 0.642873,
      "epoch": 1.4581430745814308,
      "grad_norm": 6.09375,
      "learning_rate": 1.8760901777543273e-06,
      "loss": 1.67422428,
      "memory(GiB)": 117.38,
      "step": 57480,
      "train_speed(iter/s)": 1.637167
    },
    {
      "acc": 0.66464052,
      "epoch": 1.4582699137493658,
      "grad_norm": 5.8125,
      "learning_rate": 1.8752714818608036e-06,
      "loss": 1.58630896,
      "memory(GiB)": 117.38,
      "step": 57485,
      "train_speed(iter/s)": 1.637182
    },
    {
      "acc": 0.66656523,
      "epoch": 1.458396752917301,
      "grad_norm": 5.40625,
      "learning_rate": 1.874452923406791e-06,
      "loss": 1.57180309,
      "memory(GiB)": 117.38,
      "step": 57490,
      "train_speed(iter/s)": 1.637198
    },
    {
      "acc": 0.65065546,
      "epoch": 1.458523592085236,
      "grad_norm": 5.4375,
      "learning_rate": 1.8736345024282937e-06,
      "loss": 1.63041382,
      "memory(GiB)": 117.38,
      "step": 57495,
      "train_speed(iter/s)": 1.637214
    },
    {
      "acc": 0.6569746,
      "epoch": 1.458650431253171,
      "grad_norm": 6.78125,
      "learning_rate": 1.8728162189613085e-06,
      "loss": 1.56969585,
      "memory(GiB)": 117.38,
      "step": 57500,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.66193199,
      "epoch": 1.458777270421106,
      "grad_norm": 7.21875,
      "learning_rate": 1.8719980730418285e-06,
      "loss": 1.61484318,
      "memory(GiB)": 117.38,
      "step": 57505,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.64654274,
      "epoch": 1.4589041095890412,
      "grad_norm": 6.875,
      "learning_rate": 1.8711800647058388e-06,
      "loss": 1.60167351,
      "memory(GiB)": 117.38,
      "step": 57510,
      "train_speed(iter/s)": 1.637259
    },
    {
      "acc": 0.64001398,
      "epoch": 1.4590309487569761,
      "grad_norm": 5.8125,
      "learning_rate": 1.8703621939893185e-06,
      "loss": 1.6887989,
      "memory(GiB)": 117.38,
      "step": 57515,
      "train_speed(iter/s)": 1.637274
    },
    {
      "acc": 0.64813566,
      "epoch": 1.4591577879249111,
      "grad_norm": 7.0625,
      "learning_rate": 1.869544460928242e-06,
      "loss": 1.58926125,
      "memory(GiB)": 117.38,
      "step": 57520,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.65228615,
      "epoch": 1.4592846270928463,
      "grad_norm": 5.5,
      "learning_rate": 1.868726865558575e-06,
      "loss": 1.59925556,
      "memory(GiB)": 117.38,
      "step": 57525,
      "train_speed(iter/s)": 1.637304
    },
    {
      "acc": 0.65611172,
      "epoch": 1.4594114662607813,
      "grad_norm": 5.0,
      "learning_rate": 1.8679094079162835e-06,
      "loss": 1.55919685,
      "memory(GiB)": 117.38,
      "step": 57530,
      "train_speed(iter/s)": 1.637319
    },
    {
      "acc": 0.65540648,
      "epoch": 1.4595383054287163,
      "grad_norm": 7.59375,
      "learning_rate": 1.867092088037319e-06,
      "loss": 1.57990208,
      "memory(GiB)": 117.38,
      "step": 57535,
      "train_speed(iter/s)": 1.637335
    },
    {
      "acc": 0.64379635,
      "epoch": 1.4596651445966513,
      "grad_norm": 5.9375,
      "learning_rate": 1.8662749059576296e-06,
      "loss": 1.65772171,
      "memory(GiB)": 117.38,
      "step": 57540,
      "train_speed(iter/s)": 1.637352
    },
    {
      "acc": 0.65981836,
      "epoch": 1.4597919837645865,
      "grad_norm": 7.28125,
      "learning_rate": 1.865457861713163e-06,
      "loss": 1.64376106,
      "memory(GiB)": 117.38,
      "step": 57545,
      "train_speed(iter/s)": 1.637367
    },
    {
      "acc": 0.67210875,
      "epoch": 1.4599188229325215,
      "grad_norm": 5.84375,
      "learning_rate": 1.8646409553398558e-06,
      "loss": 1.54521103,
      "memory(GiB)": 117.38,
      "step": 57550,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.66094265,
      "epoch": 1.4600456621004567,
      "grad_norm": 5.40625,
      "learning_rate": 1.8638241868736367e-06,
      "loss": 1.57712822,
      "memory(GiB)": 117.38,
      "step": 57555,
      "train_speed(iter/s)": 1.637397
    },
    {
      "acc": 0.65462923,
      "epoch": 1.4601725012683917,
      "grad_norm": 6.0625,
      "learning_rate": 1.8630075563504297e-06,
      "loss": 1.65783138,
      "memory(GiB)": 117.38,
      "step": 57560,
      "train_speed(iter/s)": 1.637412
    },
    {
      "acc": 0.6534667,
      "epoch": 1.4602993404363267,
      "grad_norm": 5.28125,
      "learning_rate": 1.8621910638061575e-06,
      "loss": 1.61833553,
      "memory(GiB)": 117.38,
      "step": 57565,
      "train_speed(iter/s)": 1.637428
    },
    {
      "acc": 0.64576778,
      "epoch": 1.4604261796042617,
      "grad_norm": 5.8125,
      "learning_rate": 1.8613747092767336e-06,
      "loss": 1.62031097,
      "memory(GiB)": 117.38,
      "step": 57570,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.65791631,
      "epoch": 1.460553018772197,
      "grad_norm": 5.5,
      "learning_rate": 1.8605584927980596e-06,
      "loss": 1.58129625,
      "memory(GiB)": 117.38,
      "step": 57575,
      "train_speed(iter/s)": 1.637458
    },
    {
      "acc": 0.65510769,
      "epoch": 1.4606798579401319,
      "grad_norm": 6.0625,
      "learning_rate": 1.859742414406041e-06,
      "loss": 1.54510098,
      "memory(GiB)": 117.38,
      "step": 57580,
      "train_speed(iter/s)": 1.637473
    },
    {
      "acc": 0.68095303,
      "epoch": 1.460806697108067,
      "grad_norm": 5.71875,
      "learning_rate": 1.8589264741365714e-06,
      "loss": 1.45404854,
      "memory(GiB)": 117.38,
      "step": 57585,
      "train_speed(iter/s)": 1.637489
    },
    {
      "acc": 0.64815416,
      "epoch": 1.460933536276002,
      "grad_norm": 4.71875,
      "learning_rate": 1.8581106720255414e-06,
      "loss": 1.6676609,
      "memory(GiB)": 117.38,
      "step": 57590,
      "train_speed(iter/s)": 1.637504
    },
    {
      "acc": 0.65925474,
      "epoch": 1.461060375443937,
      "grad_norm": 6.3125,
      "learning_rate": 1.8572950081088282e-06,
      "loss": 1.58584385,
      "memory(GiB)": 117.38,
      "step": 57595,
      "train_speed(iter/s)": 1.637519
    },
    {
      "acc": 0.65165644,
      "epoch": 1.461187214611872,
      "grad_norm": 6.15625,
      "learning_rate": 1.856479482422313e-06,
      "loss": 1.57179966,
      "memory(GiB)": 117.38,
      "step": 57600,
      "train_speed(iter/s)": 1.637535
    },
    {
      "acc": 0.66127176,
      "epoch": 1.4613140537798073,
      "grad_norm": 6.71875,
      "learning_rate": 1.8556640950018651e-06,
      "loss": 1.5802146,
      "memory(GiB)": 117.38,
      "step": 57605,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.67094517,
      "epoch": 1.4614408929477423,
      "grad_norm": 6.8125,
      "learning_rate": 1.8548488458833485e-06,
      "loss": 1.55746155,
      "memory(GiB)": 117.38,
      "step": 57610,
      "train_speed(iter/s)": 1.637565
    },
    {
      "acc": 0.66236334,
      "epoch": 1.4615677321156773,
      "grad_norm": 5.15625,
      "learning_rate": 1.854033735102622e-06,
      "loss": 1.60938721,
      "memory(GiB)": 117.38,
      "step": 57615,
      "train_speed(iter/s)": 1.63758
    },
    {
      "acc": 0.65521197,
      "epoch": 1.4616945712836125,
      "grad_norm": 6.90625,
      "learning_rate": 1.8532187626955377e-06,
      "loss": 1.6438303,
      "memory(GiB)": 117.38,
      "step": 57620,
      "train_speed(iter/s)": 1.637597
    },
    {
      "acc": 0.66102042,
      "epoch": 1.4618214104515475,
      "grad_norm": 5.6875,
      "learning_rate": 1.8524039286979417e-06,
      "loss": 1.62839165,
      "memory(GiB)": 117.38,
      "step": 57625,
      "train_speed(iter/s)": 1.637612
    },
    {
      "acc": 0.66023965,
      "epoch": 1.4619482496194824,
      "grad_norm": 5.65625,
      "learning_rate": 1.8515892331456736e-06,
      "loss": 1.61627388,
      "memory(GiB)": 117.38,
      "step": 57630,
      "train_speed(iter/s)": 1.637627
    },
    {
      "acc": 0.68634062,
      "epoch": 1.4620750887874174,
      "grad_norm": 5.53125,
      "learning_rate": 1.850774676074568e-06,
      "loss": 1.47016926,
      "memory(GiB)": 117.38,
      "step": 57635,
      "train_speed(iter/s)": 1.637643
    },
    {
      "acc": 0.65195532,
      "epoch": 1.4622019279553526,
      "grad_norm": 5.625,
      "learning_rate": 1.8499602575204522e-06,
      "loss": 1.64178848,
      "memory(GiB)": 117.38,
      "step": 57640,
      "train_speed(iter/s)": 1.637657
    },
    {
      "acc": 0.64850588,
      "epoch": 1.4623287671232876,
      "grad_norm": 5.3125,
      "learning_rate": 1.8491459775191484e-06,
      "loss": 1.58326969,
      "memory(GiB)": 117.38,
      "step": 57645,
      "train_speed(iter/s)": 1.637673
    },
    {
      "acc": 0.64244986,
      "epoch": 1.4624556062912228,
      "grad_norm": 6.09375,
      "learning_rate": 1.8483318361064716e-06,
      "loss": 1.6479784,
      "memory(GiB)": 117.38,
      "step": 57650,
      "train_speed(iter/s)": 1.637689
    },
    {
      "acc": 0.66887412,
      "epoch": 1.4625824454591578,
      "grad_norm": 6.09375,
      "learning_rate": 1.847517833318232e-06,
      "loss": 1.58741856,
      "memory(GiB)": 117.38,
      "step": 57655,
      "train_speed(iter/s)": 1.637705
    },
    {
      "acc": 0.67089691,
      "epoch": 1.4627092846270928,
      "grad_norm": 6.03125,
      "learning_rate": 1.8467039691902334e-06,
      "loss": 1.49151707,
      "memory(GiB)": 117.38,
      "step": 57660,
      "train_speed(iter/s)": 1.63772
    },
    {
      "acc": 0.66514549,
      "epoch": 1.4628361237950278,
      "grad_norm": 5.65625,
      "learning_rate": 1.8458902437582705e-06,
      "loss": 1.57454929,
      "memory(GiB)": 117.38,
      "step": 57665,
      "train_speed(iter/s)": 1.637735
    },
    {
      "acc": 0.65667682,
      "epoch": 1.462962962962963,
      "grad_norm": 5.25,
      "learning_rate": 1.8450766570581402e-06,
      "loss": 1.63363724,
      "memory(GiB)": 117.38,
      "step": 57670,
      "train_speed(iter/s)": 1.63775
    },
    {
      "acc": 0.65358829,
      "epoch": 1.463089802130898,
      "grad_norm": 5.9375,
      "learning_rate": 1.8442632091256223e-06,
      "loss": 1.60633087,
      "memory(GiB)": 117.38,
      "step": 57675,
      "train_speed(iter/s)": 1.637766
    },
    {
      "acc": 0.65549574,
      "epoch": 1.463216641298833,
      "grad_norm": 5.46875,
      "learning_rate": 1.8434498999964983e-06,
      "loss": 1.59054184,
      "memory(GiB)": 117.38,
      "step": 57680,
      "train_speed(iter/s)": 1.63778
    },
    {
      "acc": 0.66804919,
      "epoch": 1.4633434804667682,
      "grad_norm": 5.625,
      "learning_rate": 1.8426367297065384e-06,
      "loss": 1.56434231,
      "memory(GiB)": 117.38,
      "step": 57685,
      "train_speed(iter/s)": 1.637795
    },
    {
      "acc": 0.64665184,
      "epoch": 1.4634703196347032,
      "grad_norm": 5.25,
      "learning_rate": 1.841823698291516e-06,
      "loss": 1.66826744,
      "memory(GiB)": 117.38,
      "step": 57690,
      "train_speed(iter/s)": 1.63781
    },
    {
      "acc": 0.66567688,
      "epoch": 1.4635971588026382,
      "grad_norm": 5.8125,
      "learning_rate": 1.8410108057871851e-06,
      "loss": 1.53914165,
      "memory(GiB)": 117.38,
      "step": 57695,
      "train_speed(iter/s)": 1.637825
    },
    {
      "acc": 0.66354566,
      "epoch": 1.4637239979705732,
      "grad_norm": 8.0625,
      "learning_rate": 1.8401980522293017e-06,
      "loss": 1.55408611,
      "memory(GiB)": 117.38,
      "step": 57700,
      "train_speed(iter/s)": 1.63784
    },
    {
      "acc": 0.64570875,
      "epoch": 1.4638508371385084,
      "grad_norm": 8.125,
      "learning_rate": 1.839385437653617e-06,
      "loss": 1.68113461,
      "memory(GiB)": 117.38,
      "step": 57705,
      "train_speed(iter/s)": 1.637855
    },
    {
      "acc": 0.65801697,
      "epoch": 1.4639776763064434,
      "grad_norm": 4.78125,
      "learning_rate": 1.8385729620958731e-06,
      "loss": 1.64358425,
      "memory(GiB)": 117.38,
      "step": 57710,
      "train_speed(iter/s)": 1.63787
    },
    {
      "acc": 0.65845752,
      "epoch": 1.4641045154743786,
      "grad_norm": 6.6875,
      "learning_rate": 1.8377606255918024e-06,
      "loss": 1.59352703,
      "memory(GiB)": 117.38,
      "step": 57715,
      "train_speed(iter/s)": 1.637885
    },
    {
      "acc": 0.64405642,
      "epoch": 1.4642313546423136,
      "grad_norm": 5.90625,
      "learning_rate": 1.8369484281771388e-06,
      "loss": 1.65216274,
      "memory(GiB)": 117.38,
      "step": 57720,
      "train_speed(iter/s)": 1.6379
    },
    {
      "acc": 0.65615058,
      "epoch": 1.4643581938102486,
      "grad_norm": 5.28125,
      "learning_rate": 1.836136369887606e-06,
      "loss": 1.57136021,
      "memory(GiB)": 117.38,
      "step": 57725,
      "train_speed(iter/s)": 1.637915
    },
    {
      "acc": 0.66528053,
      "epoch": 1.4644850329781836,
      "grad_norm": 12.0,
      "learning_rate": 1.8353244507589225e-06,
      "loss": 1.58454514,
      "memory(GiB)": 117.38,
      "step": 57730,
      "train_speed(iter/s)": 1.63793
    },
    {
      "acc": 0.6485703,
      "epoch": 1.4646118721461188,
      "grad_norm": 5.3125,
      "learning_rate": 1.8345126708267958e-06,
      "loss": 1.60334282,
      "memory(GiB)": 117.38,
      "step": 57735,
      "train_speed(iter/s)": 1.637945
    },
    {
      "acc": 0.6711556,
      "epoch": 1.4647387113140538,
      "grad_norm": 6.25,
      "learning_rate": 1.8337010301269364e-06,
      "loss": 1.54415798,
      "memory(GiB)": 117.38,
      "step": 57740,
      "train_speed(iter/s)": 1.63796
    },
    {
      "acc": 0.65799427,
      "epoch": 1.464865550481989,
      "grad_norm": 7.125,
      "learning_rate": 1.8328895286950422e-06,
      "loss": 1.61259537,
      "memory(GiB)": 117.38,
      "step": 57745,
      "train_speed(iter/s)": 1.637976
    },
    {
      "acc": 0.64889278,
      "epoch": 1.464992389649924,
      "grad_norm": 5.78125,
      "learning_rate": 1.8320781665668063e-06,
      "loss": 1.61780052,
      "memory(GiB)": 117.38,
      "step": 57750,
      "train_speed(iter/s)": 1.637991
    },
    {
      "acc": 0.63451982,
      "epoch": 1.465119228817859,
      "grad_norm": 5.6875,
      "learning_rate": 1.8312669437779167e-06,
      "loss": 1.71501007,
      "memory(GiB)": 117.38,
      "step": 57755,
      "train_speed(iter/s)": 1.638006
    },
    {
      "acc": 0.65335474,
      "epoch": 1.465246067985794,
      "grad_norm": 5.59375,
      "learning_rate": 1.8304558603640544e-06,
      "loss": 1.64657059,
      "memory(GiB)": 117.38,
      "step": 57760,
      "train_speed(iter/s)": 1.638021
    },
    {
      "acc": 0.64696302,
      "epoch": 1.4653729071537291,
      "grad_norm": 5.125,
      "learning_rate": 1.8296449163608942e-06,
      "loss": 1.67038918,
      "memory(GiB)": 117.38,
      "step": 57765,
      "train_speed(iter/s)": 1.638035
    },
    {
      "acc": 0.67117066,
      "epoch": 1.4654997463216641,
      "grad_norm": 5.09375,
      "learning_rate": 1.8288341118041052e-06,
      "loss": 1.4601757,
      "memory(GiB)": 117.38,
      "step": 57770,
      "train_speed(iter/s)": 1.63805
    },
    {
      "acc": 0.66161356,
      "epoch": 1.4656265854895991,
      "grad_norm": 6.34375,
      "learning_rate": 1.82802344672935e-06,
      "loss": 1.62998199,
      "memory(GiB)": 117.38,
      "step": 57775,
      "train_speed(iter/s)": 1.638066
    },
    {
      "acc": 0.64639921,
      "epoch": 1.4657534246575343,
      "grad_norm": 5.40625,
      "learning_rate": 1.8272129211722855e-06,
      "loss": 1.60705299,
      "memory(GiB)": 117.38,
      "step": 57780,
      "train_speed(iter/s)": 1.638081
    },
    {
      "acc": 0.65424089,
      "epoch": 1.4658802638254693,
      "grad_norm": 6.4375,
      "learning_rate": 1.8264025351685627e-06,
      "loss": 1.63999844,
      "memory(GiB)": 117.38,
      "step": 57785,
      "train_speed(iter/s)": 1.638096
    },
    {
      "acc": 0.65920467,
      "epoch": 1.4660071029934043,
      "grad_norm": 6.0625,
      "learning_rate": 1.8255922887538251e-06,
      "loss": 1.56166487,
      "memory(GiB)": 117.38,
      "step": 57790,
      "train_speed(iter/s)": 1.638111
    },
    {
      "acc": 0.65805445,
      "epoch": 1.4661339421613393,
      "grad_norm": 9.625,
      "learning_rate": 1.8247821819637112e-06,
      "loss": 1.57964792,
      "memory(GiB)": 117.38,
      "step": 57795,
      "train_speed(iter/s)": 1.638127
    },
    {
      "acc": 0.65947523,
      "epoch": 1.4662607813292745,
      "grad_norm": 7.5625,
      "learning_rate": 1.8239722148338534e-06,
      "loss": 1.55246401,
      "memory(GiB)": 117.38,
      "step": 57800,
      "train_speed(iter/s)": 1.638143
    },
    {
      "acc": 0.65045786,
      "epoch": 1.4663876204972095,
      "grad_norm": 5.5625,
      "learning_rate": 1.823162387399876e-06,
      "loss": 1.64657764,
      "memory(GiB)": 117.38,
      "step": 57805,
      "train_speed(iter/s)": 1.638157
    },
    {
      "acc": 0.67498369,
      "epoch": 1.4665144596651447,
      "grad_norm": 5.90625,
      "learning_rate": 1.822352699697404e-06,
      "loss": 1.55461931,
      "memory(GiB)": 117.38,
      "step": 57810,
      "train_speed(iter/s)": 1.638173
    },
    {
      "acc": 0.66299582,
      "epoch": 1.4666412988330797,
      "grad_norm": 5.59375,
      "learning_rate": 1.8215431517620452e-06,
      "loss": 1.602038,
      "memory(GiB)": 117.38,
      "step": 57815,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.64773974,
      "epoch": 1.4667681380010147,
      "grad_norm": 5.8125,
      "learning_rate": 1.8207337436294097e-06,
      "loss": 1.71699619,
      "memory(GiB)": 117.38,
      "step": 57820,
      "train_speed(iter/s)": 1.638203
    },
    {
      "acc": 0.65654716,
      "epoch": 1.4668949771689497,
      "grad_norm": 6.53125,
      "learning_rate": 1.819924475335097e-06,
      "loss": 1.59594221,
      "memory(GiB)": 117.38,
      "step": 57825,
      "train_speed(iter/s)": 1.638218
    },
    {
      "acc": 0.65022173,
      "epoch": 1.4670218163368849,
      "grad_norm": 6.125,
      "learning_rate": 1.8191153469147065e-06,
      "loss": 1.57646351,
      "memory(GiB)": 117.38,
      "step": 57830,
      "train_speed(iter/s)": 1.638233
    },
    {
      "acc": 0.64668159,
      "epoch": 1.4671486555048199,
      "grad_norm": 5.28125,
      "learning_rate": 1.8183063584038236e-06,
      "loss": 1.68735371,
      "memory(GiB)": 117.38,
      "step": 57835,
      "train_speed(iter/s)": 1.638249
    },
    {
      "acc": 0.6517664,
      "epoch": 1.4672754946727549,
      "grad_norm": 5.5625,
      "learning_rate": 1.8174975098380304e-06,
      "loss": 1.62523823,
      "memory(GiB)": 117.38,
      "step": 57840,
      "train_speed(iter/s)": 1.638265
    },
    {
      "acc": 0.66627073,
      "epoch": 1.46740233384069,
      "grad_norm": 5.53125,
      "learning_rate": 1.8166888012529078e-06,
      "loss": 1.57713547,
      "memory(GiB)": 117.38,
      "step": 57845,
      "train_speed(iter/s)": 1.638279
    },
    {
      "acc": 0.65476637,
      "epoch": 1.467529173008625,
      "grad_norm": 6.0,
      "learning_rate": 1.8158802326840252e-06,
      "loss": 1.55509501,
      "memory(GiB)": 117.38,
      "step": 57850,
      "train_speed(iter/s)": 1.638294
    },
    {
      "acc": 0.6728941,
      "epoch": 1.46765601217656,
      "grad_norm": 4.625,
      "learning_rate": 1.8150718041669447e-06,
      "loss": 1.53818569,
      "memory(GiB)": 117.38,
      "step": 57855,
      "train_speed(iter/s)": 1.638308
    },
    {
      "acc": 0.65525231,
      "epoch": 1.467782851344495,
      "grad_norm": 5.75,
      "learning_rate": 1.814263515737224e-06,
      "loss": 1.61912804,
      "memory(GiB)": 117.38,
      "step": 57860,
      "train_speed(iter/s)": 1.638322
    },
    {
      "acc": 0.65495143,
      "epoch": 1.4679096905124303,
      "grad_norm": 4.5,
      "learning_rate": 1.813455367430419e-06,
      "loss": 1.55582657,
      "memory(GiB)": 117.38,
      "step": 57865,
      "train_speed(iter/s)": 1.638337
    },
    {
      "acc": 0.65118151,
      "epoch": 1.4680365296803652,
      "grad_norm": 5.46875,
      "learning_rate": 1.812647359282076e-06,
      "loss": 1.7126833,
      "memory(GiB)": 117.38,
      "step": 57870,
      "train_speed(iter/s)": 1.638352
    },
    {
      "acc": 0.64035082,
      "epoch": 1.4681633688483005,
      "grad_norm": 5.15625,
      "learning_rate": 1.8118394913277287e-06,
      "loss": 1.60532188,
      "memory(GiB)": 117.38,
      "step": 57875,
      "train_speed(iter/s)": 1.638121
    },
    {
      "acc": 0.64400587,
      "epoch": 1.4682902080162354,
      "grad_norm": 5.9375,
      "learning_rate": 1.8110317636029162e-06,
      "loss": 1.59559479,
      "memory(GiB)": 117.38,
      "step": 57880,
      "train_speed(iter/s)": 1.638136
    },
    {
      "acc": 0.66109982,
      "epoch": 1.4684170471841704,
      "grad_norm": 5.90625,
      "learning_rate": 1.810224176143165e-06,
      "loss": 1.64881477,
      "memory(GiB)": 117.38,
      "step": 57885,
      "train_speed(iter/s)": 1.63815
    },
    {
      "acc": 0.6538269,
      "epoch": 1.4685438863521054,
      "grad_norm": 6.34375,
      "learning_rate": 1.8094167289839953e-06,
      "loss": 1.52976017,
      "memory(GiB)": 117.38,
      "step": 57890,
      "train_speed(iter/s)": 1.638164
    },
    {
      "acc": 0.64798522,
      "epoch": 1.4686707255200406,
      "grad_norm": 6.28125,
      "learning_rate": 1.808609422160923e-06,
      "loss": 1.66295319,
      "memory(GiB)": 117.38,
      "step": 57895,
      "train_speed(iter/s)": 1.638179
    },
    {
      "acc": 0.65884228,
      "epoch": 1.4687975646879756,
      "grad_norm": 6.03125,
      "learning_rate": 1.8078022557094571e-06,
      "loss": 1.59045715,
      "memory(GiB)": 117.38,
      "step": 57900,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.6462811,
      "epoch": 1.4689244038559108,
      "grad_norm": 5.96875,
      "learning_rate": 1.8069952296651e-06,
      "loss": 1.63971462,
      "memory(GiB)": 117.38,
      "step": 57905,
      "train_speed(iter/s)": 1.638208
    },
    {
      "acc": 0.64991851,
      "epoch": 1.4690512430238458,
      "grad_norm": 6.9375,
      "learning_rate": 1.8061883440633481e-06,
      "loss": 1.6211853,
      "memory(GiB)": 117.38,
      "step": 57910,
      "train_speed(iter/s)": 1.638223
    },
    {
      "acc": 0.6713758,
      "epoch": 1.4691780821917808,
      "grad_norm": 5.78125,
      "learning_rate": 1.8053815989396927e-06,
      "loss": 1.58132153,
      "memory(GiB)": 117.38,
      "step": 57915,
      "train_speed(iter/s)": 1.638238
    },
    {
      "acc": 0.66793137,
      "epoch": 1.4693049213597158,
      "grad_norm": 5.8125,
      "learning_rate": 1.8045749943296171e-06,
      "loss": 1.54448557,
      "memory(GiB)": 117.38,
      "step": 57920,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.65428467,
      "epoch": 1.469431760527651,
      "grad_norm": 6.53125,
      "learning_rate": 1.8037685302686003e-06,
      "loss": 1.6049469,
      "memory(GiB)": 117.38,
      "step": 57925,
      "train_speed(iter/s)": 1.638268
    },
    {
      "acc": 0.65100832,
      "epoch": 1.469558599695586,
      "grad_norm": 5.84375,
      "learning_rate": 1.8029622067921133e-06,
      "loss": 1.67594109,
      "memory(GiB)": 117.38,
      "step": 57930,
      "train_speed(iter/s)": 1.638283
    },
    {
      "acc": 0.66949358,
      "epoch": 1.469685438863521,
      "grad_norm": 6.8125,
      "learning_rate": 1.8021560239356223e-06,
      "loss": 1.52854862,
      "memory(GiB)": 117.38,
      "step": 57935,
      "train_speed(iter/s)": 1.638298
    },
    {
      "acc": 0.66745014,
      "epoch": 1.4698122780314562,
      "grad_norm": 5.15625,
      "learning_rate": 1.8013499817345865e-06,
      "loss": 1.57253742,
      "memory(GiB)": 117.38,
      "step": 57940,
      "train_speed(iter/s)": 1.638314
    },
    {
      "acc": 0.66670198,
      "epoch": 1.4699391171993912,
      "grad_norm": 5.78125,
      "learning_rate": 1.8005440802244595e-06,
      "loss": 1.60223846,
      "memory(GiB)": 117.38,
      "step": 57945,
      "train_speed(iter/s)": 1.638329
    },
    {
      "acc": 0.66258121,
      "epoch": 1.4700659563673262,
      "grad_norm": 4.90625,
      "learning_rate": 1.7997383194406887e-06,
      "loss": 1.55077305,
      "memory(GiB)": 117.38,
      "step": 57950,
      "train_speed(iter/s)": 1.638344
    },
    {
      "acc": 0.66343145,
      "epoch": 1.4701927955352612,
      "grad_norm": 5.0625,
      "learning_rate": 1.7989326994187146e-06,
      "loss": 1.60424995,
      "memory(GiB)": 117.38,
      "step": 57955,
      "train_speed(iter/s)": 1.638358
    },
    {
      "acc": 0.66509609,
      "epoch": 1.4703196347031964,
      "grad_norm": 6.1875,
      "learning_rate": 1.798127220193972e-06,
      "loss": 1.61679001,
      "memory(GiB)": 117.38,
      "step": 57960,
      "train_speed(iter/s)": 1.638374
    },
    {
      "acc": 0.64575863,
      "epoch": 1.4704464738711314,
      "grad_norm": 6.59375,
      "learning_rate": 1.7973218818018878e-06,
      "loss": 1.60709801,
      "memory(GiB)": 117.38,
      "step": 57965,
      "train_speed(iter/s)": 1.63839
    },
    {
      "acc": 0.65588737,
      "epoch": 1.4705733130390666,
      "grad_norm": 5.40625,
      "learning_rate": 1.7965166842778897e-06,
      "loss": 1.60618763,
      "memory(GiB)": 117.38,
      "step": 57970,
      "train_speed(iter/s)": 1.638405
    },
    {
      "acc": 0.66633224,
      "epoch": 1.4707001522070016,
      "grad_norm": 5.15625,
      "learning_rate": 1.7957116276573888e-06,
      "loss": 1.60000439,
      "memory(GiB)": 117.38,
      "step": 57975,
      "train_speed(iter/s)": 1.638419
    },
    {
      "acc": 0.6571146,
      "epoch": 1.4708269913749366,
      "grad_norm": 6.5,
      "learning_rate": 1.7949067119757951e-06,
      "loss": 1.56607904,
      "memory(GiB)": 117.38,
      "step": 57980,
      "train_speed(iter/s)": 1.638433
    },
    {
      "acc": 0.64114771,
      "epoch": 1.4709538305428715,
      "grad_norm": 5.875,
      "learning_rate": 1.7941019372685154e-06,
      "loss": 1.65334702,
      "memory(GiB)": 117.38,
      "step": 57985,
      "train_speed(iter/s)": 1.638448
    },
    {
      "acc": 0.65712395,
      "epoch": 1.4710806697108068,
      "grad_norm": 7.1875,
      "learning_rate": 1.7932973035709471e-06,
      "loss": 1.53186407,
      "memory(GiB)": 117.38,
      "step": 57990,
      "train_speed(iter/s)": 1.638462
    },
    {
      "acc": 0.65712624,
      "epoch": 1.4712075088787417,
      "grad_norm": 6.875,
      "learning_rate": 1.792492810918479e-06,
      "loss": 1.60296822,
      "memory(GiB)": 117.38,
      "step": 57995,
      "train_speed(iter/s)": 1.638477
    },
    {
      "acc": 0.64174623,
      "epoch": 1.4713343480466767,
      "grad_norm": 5.53125,
      "learning_rate": 1.7916884593464957e-06,
      "loss": 1.651194,
      "memory(GiB)": 117.38,
      "step": 58000,
      "train_speed(iter/s)": 1.638491
    },
    {
      "epoch": 1.4713343480466767,
      "eval_acc": 0.6463060997333233,
      "eval_loss": 1.5731773376464844,
      "eval_runtime": 58.4715,
      "eval_samples_per_second": 108.942,
      "eval_steps_per_second": 27.244,
      "step": 58000
    },
    {
      "acc": 0.65707092,
      "epoch": 1.471461187214612,
      "grad_norm": 6.59375,
      "learning_rate": 1.79088424889038e-06,
      "loss": 1.65774002,
      "memory(GiB)": 117.38,
      "step": 58005,
      "train_speed(iter/s)": 1.635603
    },
    {
      "acc": 0.65804262,
      "epoch": 1.471588026382547,
      "grad_norm": 6.78125,
      "learning_rate": 1.7900801795855043e-06,
      "loss": 1.56835489,
      "memory(GiB)": 117.38,
      "step": 58010,
      "train_speed(iter/s)": 1.635619
    },
    {
      "acc": 0.65304799,
      "epoch": 1.471714865550482,
      "grad_norm": 6.5,
      "learning_rate": 1.7892762514672303e-06,
      "loss": 1.59533587,
      "memory(GiB)": 117.38,
      "step": 58015,
      "train_speed(iter/s)": 1.635634
    },
    {
      "acc": 0.64941578,
      "epoch": 1.471841704718417,
      "grad_norm": 6.28125,
      "learning_rate": 1.7884724645709228e-06,
      "loss": 1.66165295,
      "memory(GiB)": 117.38,
      "step": 58020,
      "train_speed(iter/s)": 1.635649
    },
    {
      "acc": 0.66016474,
      "epoch": 1.4719685438863521,
      "grad_norm": 5.46875,
      "learning_rate": 1.7876688189319353e-06,
      "loss": 1.62394047,
      "memory(GiB)": 117.38,
      "step": 58025,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.66501589,
      "epoch": 1.472095383054287,
      "grad_norm": 4.59375,
      "learning_rate": 1.7868653145856163e-06,
      "loss": 1.54129066,
      "memory(GiB)": 117.38,
      "step": 58030,
      "train_speed(iter/s)": 1.63568
    },
    {
      "acc": 0.67722635,
      "epoch": 1.4722222222222223,
      "grad_norm": 6.5,
      "learning_rate": 1.7860619515673034e-06,
      "loss": 1.46738453,
      "memory(GiB)": 117.38,
      "step": 58035,
      "train_speed(iter/s)": 1.635697
    },
    {
      "acc": 0.67051888,
      "epoch": 1.4723490613901573,
      "grad_norm": 5.1875,
      "learning_rate": 1.785258729912337e-06,
      "loss": 1.58444004,
      "memory(GiB)": 117.38,
      "step": 58040,
      "train_speed(iter/s)": 1.635712
    },
    {
      "acc": 0.63772449,
      "epoch": 1.4724759005580923,
      "grad_norm": 5.59375,
      "learning_rate": 1.784455649656044e-06,
      "loss": 1.70479698,
      "memory(GiB)": 117.38,
      "step": 58045,
      "train_speed(iter/s)": 1.635727
    },
    {
      "acc": 0.66457853,
      "epoch": 1.4726027397260273,
      "grad_norm": 4.71875,
      "learning_rate": 1.7836527108337482e-06,
      "loss": 1.58913021,
      "memory(GiB)": 117.38,
      "step": 58050,
      "train_speed(iter/s)": 1.635742
    },
    {
      "acc": 0.66410203,
      "epoch": 1.4727295788939625,
      "grad_norm": 4.8125,
      "learning_rate": 1.782849913480766e-06,
      "loss": 1.58055525,
      "memory(GiB)": 117.38,
      "step": 58055,
      "train_speed(iter/s)": 1.635758
    },
    {
      "acc": 0.65571756,
      "epoch": 1.4728564180618975,
      "grad_norm": 5.34375,
      "learning_rate": 1.7820472576324078e-06,
      "loss": 1.60898399,
      "memory(GiB)": 117.38,
      "step": 58060,
      "train_speed(iter/s)": 1.635774
    },
    {
      "acc": 0.66803703,
      "epoch": 1.4729832572298327,
      "grad_norm": 5.5625,
      "learning_rate": 1.7812447433239789e-06,
      "loss": 1.54877119,
      "memory(GiB)": 117.38,
      "step": 58065,
      "train_speed(iter/s)": 1.63579
    },
    {
      "acc": 0.64985929,
      "epoch": 1.4731100963977677,
      "grad_norm": 5.46875,
      "learning_rate": 1.7804423705907764e-06,
      "loss": 1.71767006,
      "memory(GiB)": 117.38,
      "step": 58070,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.65960436,
      "epoch": 1.4732369355657027,
      "grad_norm": 5.03125,
      "learning_rate": 1.779640139468093e-06,
      "loss": 1.5335762,
      "memory(GiB)": 117.38,
      "step": 58075,
      "train_speed(iter/s)": 1.635822
    },
    {
      "acc": 0.65477872,
      "epoch": 1.4733637747336377,
      "grad_norm": 6.15625,
      "learning_rate": 1.778838049991214e-06,
      "loss": 1.63821888,
      "memory(GiB)": 117.38,
      "step": 58080,
      "train_speed(iter/s)": 1.635838
    },
    {
      "acc": 0.64955473,
      "epoch": 1.4734906139015729,
      "grad_norm": 5.53125,
      "learning_rate": 1.778036102195419e-06,
      "loss": 1.58565264,
      "memory(GiB)": 117.38,
      "step": 58085,
      "train_speed(iter/s)": 1.635853
    },
    {
      "acc": 0.65545125,
      "epoch": 1.4736174530695079,
      "grad_norm": 5.34375,
      "learning_rate": 1.7772342961159817e-06,
      "loss": 1.60906105,
      "memory(GiB)": 117.38,
      "step": 58090,
      "train_speed(iter/s)": 1.635868
    },
    {
      "acc": 0.65875525,
      "epoch": 1.4737442922374429,
      "grad_norm": 6.375,
      "learning_rate": 1.7764326317881681e-06,
      "loss": 1.59845238,
      "memory(GiB)": 117.38,
      "step": 58095,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.66537104,
      "epoch": 1.473871131405378,
      "grad_norm": 6.90625,
      "learning_rate": 1.77563110924724e-06,
      "loss": 1.59925146,
      "memory(GiB)": 117.38,
      "step": 58100,
      "train_speed(iter/s)": 1.6359
    },
    {
      "acc": 0.64920645,
      "epoch": 1.473997970573313,
      "grad_norm": 5.6875,
      "learning_rate": 1.7748297285284494e-06,
      "loss": 1.61591263,
      "memory(GiB)": 117.38,
      "step": 58105,
      "train_speed(iter/s)": 1.635916
    },
    {
      "acc": 0.66434312,
      "epoch": 1.474124809741248,
      "grad_norm": 7.3125,
      "learning_rate": 1.7740284896670507e-06,
      "loss": 1.56024122,
      "memory(GiB)": 117.38,
      "step": 58110,
      "train_speed(iter/s)": 1.635932
    },
    {
      "acc": 0.66220188,
      "epoch": 1.474251648909183,
      "grad_norm": 5.9375,
      "learning_rate": 1.7732273926982796e-06,
      "loss": 1.58656502,
      "memory(GiB)": 117.38,
      "step": 58115,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.65404601,
      "epoch": 1.4743784880771182,
      "grad_norm": 5.21875,
      "learning_rate": 1.7724264376573747e-06,
      "loss": 1.54699917,
      "memory(GiB)": 117.38,
      "step": 58120,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.64706726,
      "epoch": 1.4745053272450532,
      "grad_norm": 5.65625,
      "learning_rate": 1.7716256245795631e-06,
      "loss": 1.65383949,
      "memory(GiB)": 117.38,
      "step": 58125,
      "train_speed(iter/s)": 1.63598
    },
    {
      "acc": 0.65086303,
      "epoch": 1.4746321664129884,
      "grad_norm": 5.8125,
      "learning_rate": 1.7708249535000737e-06,
      "loss": 1.63307858,
      "memory(GiB)": 117.38,
      "step": 58130,
      "train_speed(iter/s)": 1.635996
    },
    {
      "acc": 0.64020905,
      "epoch": 1.4747590055809234,
      "grad_norm": 5.75,
      "learning_rate": 1.7700244244541182e-06,
      "loss": 1.61890812,
      "memory(GiB)": 117.38,
      "step": 58135,
      "train_speed(iter/s)": 1.636012
    },
    {
      "acc": 0.66835909,
      "epoch": 1.4748858447488584,
      "grad_norm": 6.1875,
      "learning_rate": 1.7692240374769081e-06,
      "loss": 1.59846554,
      "memory(GiB)": 117.38,
      "step": 58140,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.65756869,
      "epoch": 1.4750126839167934,
      "grad_norm": 6.90625,
      "learning_rate": 1.7684237926036507e-06,
      "loss": 1.65764236,
      "memory(GiB)": 117.38,
      "step": 58145,
      "train_speed(iter/s)": 1.636043
    },
    {
      "acc": 0.64486589,
      "epoch": 1.4751395230847286,
      "grad_norm": 5.40625,
      "learning_rate": 1.7676236898695442e-06,
      "loss": 1.6357933,
      "memory(GiB)": 117.38,
      "step": 58150,
      "train_speed(iter/s)": 1.63606
    },
    {
      "acc": 0.65813298,
      "epoch": 1.4752663622526636,
      "grad_norm": 5.78125,
      "learning_rate": 1.7668237293097762e-06,
      "loss": 1.63641014,
      "memory(GiB)": 117.38,
      "step": 58155,
      "train_speed(iter/s)": 1.636077
    },
    {
      "acc": 0.66899223,
      "epoch": 1.4753932014205986,
      "grad_norm": 5.71875,
      "learning_rate": 1.7660239109595374e-06,
      "loss": 1.54526062,
      "memory(GiB)": 117.38,
      "step": 58160,
      "train_speed(iter/s)": 1.636091
    },
    {
      "acc": 0.63701143,
      "epoch": 1.4755200405885338,
      "grad_norm": 8.9375,
      "learning_rate": 1.7652242348540056e-06,
      "loss": 1.69027996,
      "memory(GiB)": 117.38,
      "step": 58165,
      "train_speed(iter/s)": 1.636108
    },
    {
      "acc": 0.65848742,
      "epoch": 1.4756468797564688,
      "grad_norm": 5.40625,
      "learning_rate": 1.764424701028356e-06,
      "loss": 1.5350193,
      "memory(GiB)": 117.38,
      "step": 58170,
      "train_speed(iter/s)": 1.636123
    },
    {
      "acc": 0.64417906,
      "epoch": 1.4757737189244038,
      "grad_norm": 5.625,
      "learning_rate": 1.7636253095177507e-06,
      "loss": 1.6628521,
      "memory(GiB)": 117.38,
      "step": 58175,
      "train_speed(iter/s)": 1.636139
    },
    {
      "acc": 0.63725128,
      "epoch": 1.4759005580923388,
      "grad_norm": 5.5,
      "learning_rate": 1.762826060357355e-06,
      "loss": 1.59830732,
      "memory(GiB)": 117.38,
      "step": 58180,
      "train_speed(iter/s)": 1.636155
    },
    {
      "acc": 0.65749836,
      "epoch": 1.476027397260274,
      "grad_norm": 5.84375,
      "learning_rate": 1.762026953582322e-06,
      "loss": 1.56753473,
      "memory(GiB)": 117.38,
      "step": 58185,
      "train_speed(iter/s)": 1.636167
    },
    {
      "acc": 0.65688167,
      "epoch": 1.476154236428209,
      "grad_norm": 6.53125,
      "learning_rate": 1.7612279892278006e-06,
      "loss": 1.58181448,
      "memory(GiB)": 117.38,
      "step": 58190,
      "train_speed(iter/s)": 1.636183
    },
    {
      "acc": 0.66077704,
      "epoch": 1.4762810755961442,
      "grad_norm": 6.53125,
      "learning_rate": 1.7604291673289314e-06,
      "loss": 1.58817921,
      "memory(GiB)": 117.38,
      "step": 58195,
      "train_speed(iter/s)": 1.636199
    },
    {
      "acc": 0.66322813,
      "epoch": 1.4764079147640792,
      "grad_norm": 5.625,
      "learning_rate": 1.759630487920852e-06,
      "loss": 1.51315727,
      "memory(GiB)": 117.38,
      "step": 58200,
      "train_speed(iter/s)": 1.636214
    },
    {
      "acc": 0.65955381,
      "epoch": 1.4765347539320142,
      "grad_norm": 5.0,
      "learning_rate": 1.7588319510386903e-06,
      "loss": 1.564571,
      "memory(GiB)": 117.38,
      "step": 58205,
      "train_speed(iter/s)": 1.63623
    },
    {
      "acc": 0.65465112,
      "epoch": 1.4766615930999492,
      "grad_norm": 5.03125,
      "learning_rate": 1.7580335567175704e-06,
      "loss": 1.58541908,
      "memory(GiB)": 117.38,
      "step": 58210,
      "train_speed(iter/s)": 1.636246
    },
    {
      "acc": 0.63913035,
      "epoch": 1.4767884322678844,
      "grad_norm": 5.1875,
      "learning_rate": 1.7572353049926094e-06,
      "loss": 1.71614113,
      "memory(GiB)": 117.38,
      "step": 58215,
      "train_speed(iter/s)": 1.636261
    },
    {
      "acc": 0.66926317,
      "epoch": 1.4769152714358194,
      "grad_norm": 5.03125,
      "learning_rate": 1.7564371958989173e-06,
      "loss": 1.52571945,
      "memory(GiB)": 117.38,
      "step": 58220,
      "train_speed(iter/s)": 1.636277
    },
    {
      "acc": 0.66025009,
      "epoch": 1.4770421106037546,
      "grad_norm": 7.78125,
      "learning_rate": 1.7556392294715984e-06,
      "loss": 1.60552883,
      "memory(GiB)": 117.38,
      "step": 58225,
      "train_speed(iter/s)": 1.636293
    },
    {
      "acc": 0.64934721,
      "epoch": 1.4771689497716896,
      "grad_norm": 6.15625,
      "learning_rate": 1.7548414057457518e-06,
      "loss": 1.60746765,
      "memory(GiB)": 117.38,
      "step": 58230,
      "train_speed(iter/s)": 1.636309
    },
    {
      "acc": 0.65709004,
      "epoch": 1.4772957889396245,
      "grad_norm": 5.0,
      "learning_rate": 1.7540437247564685e-06,
      "loss": 1.60906181,
      "memory(GiB)": 117.38,
      "step": 58235,
      "train_speed(iter/s)": 1.636325
    },
    {
      "acc": 0.67588587,
      "epoch": 1.4774226281075595,
      "grad_norm": 4.65625,
      "learning_rate": 1.7532461865388345e-06,
      "loss": 1.51192713,
      "memory(GiB)": 117.38,
      "step": 58240,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.6588172,
      "epoch": 1.4775494672754947,
      "grad_norm": 5.875,
      "learning_rate": 1.752448791127927e-06,
      "loss": 1.59473209,
      "memory(GiB)": 117.38,
      "step": 58245,
      "train_speed(iter/s)": 1.636357
    },
    {
      "acc": 0.66801414,
      "epoch": 1.4776763064434297,
      "grad_norm": 5.875,
      "learning_rate": 1.7516515385588245e-06,
      "loss": 1.59867077,
      "memory(GiB)": 117.38,
      "step": 58250,
      "train_speed(iter/s)": 1.636372
    },
    {
      "acc": 0.64722624,
      "epoch": 1.4778031456113647,
      "grad_norm": 6.65625,
      "learning_rate": 1.7508544288665885e-06,
      "loss": 1.6347002,
      "memory(GiB)": 117.38,
      "step": 58255,
      "train_speed(iter/s)": 1.636389
    },
    {
      "acc": 0.66878166,
      "epoch": 1.4779299847793,
      "grad_norm": 5.21875,
      "learning_rate": 1.750057462086281e-06,
      "loss": 1.57623711,
      "memory(GiB)": 117.38,
      "step": 58260,
      "train_speed(iter/s)": 1.636405
    },
    {
      "acc": 0.66516342,
      "epoch": 1.478056823947235,
      "grad_norm": 5.375,
      "learning_rate": 1.7492606382529542e-06,
      "loss": 1.58060951,
      "memory(GiB)": 117.38,
      "step": 58265,
      "train_speed(iter/s)": 1.636421
    },
    {
      "acc": 0.66816592,
      "epoch": 1.47818366311517,
      "grad_norm": 6.15625,
      "learning_rate": 1.748463957401662e-06,
      "loss": 1.59277229,
      "memory(GiB)": 117.38,
      "step": 58270,
      "train_speed(iter/s)": 1.636438
    },
    {
      "acc": 0.65403109,
      "epoch": 1.478310502283105,
      "grad_norm": 6.28125,
      "learning_rate": 1.7476674195674404e-06,
      "loss": 1.60921288,
      "memory(GiB)": 117.38,
      "step": 58275,
      "train_speed(iter/s)": 1.636454
    },
    {
      "acc": 0.66395659,
      "epoch": 1.47843734145104,
      "grad_norm": 7.21875,
      "learning_rate": 1.7468710247853244e-06,
      "loss": 1.60885086,
      "memory(GiB)": 117.38,
      "step": 58280,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.65570054,
      "epoch": 1.478564180618975,
      "grad_norm": 5.78125,
      "learning_rate": 1.7460747730903466e-06,
      "loss": 1.54865913,
      "memory(GiB)": 117.38,
      "step": 58285,
      "train_speed(iter/s)": 1.636486
    },
    {
      "acc": 0.65777349,
      "epoch": 1.4786910197869103,
      "grad_norm": 7.71875,
      "learning_rate": 1.7452786645175297e-06,
      "loss": 1.59784489,
      "memory(GiB)": 117.38,
      "step": 58290,
      "train_speed(iter/s)": 1.636503
    },
    {
      "acc": 0.64883413,
      "epoch": 1.4788178589548453,
      "grad_norm": 5.46875,
      "learning_rate": 1.7444826991018864e-06,
      "loss": 1.65592422,
      "memory(GiB)": 117.38,
      "step": 58295,
      "train_speed(iter/s)": 1.636518
    },
    {
      "acc": 0.65616913,
      "epoch": 1.4789446981227803,
      "grad_norm": 5.5,
      "learning_rate": 1.7436868768784276e-06,
      "loss": 1.59346552,
      "memory(GiB)": 117.38,
      "step": 58300,
      "train_speed(iter/s)": 1.636534
    },
    {
      "acc": 0.66765242,
      "epoch": 1.4790715372907153,
      "grad_norm": 6.84375,
      "learning_rate": 1.7428911978821594e-06,
      "loss": 1.59195023,
      "memory(GiB)": 117.38,
      "step": 58305,
      "train_speed(iter/s)": 1.63655
    },
    {
      "acc": 0.64725127,
      "epoch": 1.4791983764586505,
      "grad_norm": 6.1875,
      "learning_rate": 1.7420956621480806e-06,
      "loss": 1.63658428,
      "memory(GiB)": 117.38,
      "step": 58310,
      "train_speed(iter/s)": 1.636566
    },
    {
      "acc": 0.66420803,
      "epoch": 1.4793252156265855,
      "grad_norm": 4.875,
      "learning_rate": 1.7413002697111765e-06,
      "loss": 1.53049049,
      "memory(GiB)": 117.38,
      "step": 58315,
      "train_speed(iter/s)": 1.636582
    },
    {
      "acc": 0.66570177,
      "epoch": 1.4794520547945205,
      "grad_norm": 5.96875,
      "learning_rate": 1.7405050206064372e-06,
      "loss": 1.60504017,
      "memory(GiB)": 117.38,
      "step": 58320,
      "train_speed(iter/s)": 1.636597
    },
    {
      "acc": 0.65544472,
      "epoch": 1.4795788939624557,
      "grad_norm": 5.375,
      "learning_rate": 1.73970991486884e-06,
      "loss": 1.61635227,
      "memory(GiB)": 117.38,
      "step": 58325,
      "train_speed(iter/s)": 1.636614
    },
    {
      "acc": 0.64594116,
      "epoch": 1.4797057331303907,
      "grad_norm": 8.25,
      "learning_rate": 1.7389149525333565e-06,
      "loss": 1.64471931,
      "memory(GiB)": 117.38,
      "step": 58330,
      "train_speed(iter/s)": 1.636628
    },
    {
      "acc": 0.63516736,
      "epoch": 1.4798325722983257,
      "grad_norm": 5.25,
      "learning_rate": 1.7381201336349535e-06,
      "loss": 1.60180054,
      "memory(GiB)": 117.38,
      "step": 58335,
      "train_speed(iter/s)": 1.636644
    },
    {
      "acc": 0.65250211,
      "epoch": 1.4799594114662606,
      "grad_norm": 5.625,
      "learning_rate": 1.7373254582085896e-06,
      "loss": 1.60495605,
      "memory(GiB)": 117.38,
      "step": 58340,
      "train_speed(iter/s)": 1.636659
    },
    {
      "acc": 0.65782995,
      "epoch": 1.4800862506341959,
      "grad_norm": 7.375,
      "learning_rate": 1.7365309262892194e-06,
      "loss": 1.56491375,
      "memory(GiB)": 117.38,
      "step": 58345,
      "train_speed(iter/s)": 1.636675
    },
    {
      "acc": 0.63674431,
      "epoch": 1.4802130898021308,
      "grad_norm": 6.90625,
      "learning_rate": 1.735736537911789e-06,
      "loss": 1.67274647,
      "memory(GiB)": 117.38,
      "step": 58350,
      "train_speed(iter/s)": 1.636692
    },
    {
      "acc": 0.63673859,
      "epoch": 1.480339928970066,
      "grad_norm": 4.875,
      "learning_rate": 1.7349422931112403e-06,
      "loss": 1.66101303,
      "memory(GiB)": 117.38,
      "step": 58355,
      "train_speed(iter/s)": 1.636708
    },
    {
      "acc": 0.66338797,
      "epoch": 1.480466768138001,
      "grad_norm": 4.65625,
      "learning_rate": 1.7341481919225062e-06,
      "loss": 1.59977922,
      "memory(GiB)": 117.38,
      "step": 58360,
      "train_speed(iter/s)": 1.636724
    },
    {
      "acc": 0.65562878,
      "epoch": 1.480593607305936,
      "grad_norm": 5.46875,
      "learning_rate": 1.733354234380516e-06,
      "loss": 1.5597949,
      "memory(GiB)": 117.38,
      "step": 58365,
      "train_speed(iter/s)": 1.63674
    },
    {
      "acc": 0.65842876,
      "epoch": 1.480720446473871,
      "grad_norm": 5.25,
      "learning_rate": 1.7325604205201912e-06,
      "loss": 1.53835945,
      "memory(GiB)": 117.38,
      "step": 58370,
      "train_speed(iter/s)": 1.636755
    },
    {
      "acc": 0.6519269,
      "epoch": 1.4808472856418062,
      "grad_norm": 4.875,
      "learning_rate": 1.7317667503764468e-06,
      "loss": 1.5427165,
      "memory(GiB)": 117.38,
      "step": 58375,
      "train_speed(iter/s)": 1.636771
    },
    {
      "acc": 0.63990617,
      "epoch": 1.4809741248097412,
      "grad_norm": 5.75,
      "learning_rate": 1.7309732239841926e-06,
      "loss": 1.73910656,
      "memory(GiB)": 117.38,
      "step": 58380,
      "train_speed(iter/s)": 1.636787
    },
    {
      "acc": 0.66312122,
      "epoch": 1.4811009639776764,
      "grad_norm": 5.3125,
      "learning_rate": 1.730179841378331e-06,
      "loss": 1.5415411,
      "memory(GiB)": 117.38,
      "step": 58385,
      "train_speed(iter/s)": 1.636803
    },
    {
      "acc": 0.65903182,
      "epoch": 1.4812278031456114,
      "grad_norm": 5.21875,
      "learning_rate": 1.7293866025937589e-06,
      "loss": 1.57861357,
      "memory(GiB)": 117.38,
      "step": 58390,
      "train_speed(iter/s)": 1.636819
    },
    {
      "acc": 0.66187086,
      "epoch": 1.4813546423135464,
      "grad_norm": 5.65625,
      "learning_rate": 1.7285935076653659e-06,
      "loss": 1.56488838,
      "memory(GiB)": 117.38,
      "step": 58395,
      "train_speed(iter/s)": 1.636835
    },
    {
      "acc": 0.67188573,
      "epoch": 1.4814814814814814,
      "grad_norm": 6.15625,
      "learning_rate": 1.7278005566280365e-06,
      "loss": 1.56783524,
      "memory(GiB)": 117.38,
      "step": 58400,
      "train_speed(iter/s)": 1.636851
    },
    {
      "acc": 0.64539118,
      "epoch": 1.4816083206494166,
      "grad_norm": 6.15625,
      "learning_rate": 1.727007749516646e-06,
      "loss": 1.67626705,
      "memory(GiB)": 117.38,
      "step": 58405,
      "train_speed(iter/s)": 1.636867
    },
    {
      "acc": 0.65323706,
      "epoch": 1.4817351598173516,
      "grad_norm": 5.125,
      "learning_rate": 1.7262150863660709e-06,
      "loss": 1.66554279,
      "memory(GiB)": 117.38,
      "step": 58410,
      "train_speed(iter/s)": 1.636883
    },
    {
      "acc": 0.65039778,
      "epoch": 1.4818619989852866,
      "grad_norm": 5.21875,
      "learning_rate": 1.7254225672111713e-06,
      "loss": 1.62035866,
      "memory(GiB)": 117.38,
      "step": 58415,
      "train_speed(iter/s)": 1.636899
    },
    {
      "acc": 0.65661621,
      "epoch": 1.4819888381532218,
      "grad_norm": 6.65625,
      "learning_rate": 1.7246301920868052e-06,
      "loss": 1.58803453,
      "memory(GiB)": 117.38,
      "step": 58420,
      "train_speed(iter/s)": 1.636915
    },
    {
      "acc": 0.66044226,
      "epoch": 1.4821156773211568,
      "grad_norm": 6.59375,
      "learning_rate": 1.723837961027829e-06,
      "loss": 1.4941515,
      "memory(GiB)": 117.38,
      "step": 58425,
      "train_speed(iter/s)": 1.636931
    },
    {
      "acc": 0.65878887,
      "epoch": 1.4822425164890918,
      "grad_norm": 6.84375,
      "learning_rate": 1.723045874069087e-06,
      "loss": 1.60147572,
      "memory(GiB)": 117.38,
      "step": 58430,
      "train_speed(iter/s)": 1.636947
    },
    {
      "acc": 0.66305676,
      "epoch": 1.4823693556570268,
      "grad_norm": 5.375,
      "learning_rate": 1.7222539312454167e-06,
      "loss": 1.64099464,
      "memory(GiB)": 117.38,
      "step": 58435,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.6523262,
      "epoch": 1.482496194824962,
      "grad_norm": 6.65625,
      "learning_rate": 1.7214621325916515e-06,
      "loss": 1.63348141,
      "memory(GiB)": 117.38,
      "step": 58440,
      "train_speed(iter/s)": 1.636981
    },
    {
      "acc": 0.67395077,
      "epoch": 1.482623033992897,
      "grad_norm": 9.0625,
      "learning_rate": 1.7206704781426204e-06,
      "loss": 1.55686607,
      "memory(GiB)": 117.38,
      "step": 58445,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.66666398,
      "epoch": 1.4827498731608322,
      "grad_norm": 5.90625,
      "learning_rate": 1.7198789679331445e-06,
      "loss": 1.58234825,
      "memory(GiB)": 117.38,
      "step": 58450,
      "train_speed(iter/s)": 1.637012
    },
    {
      "acc": 0.67018414,
      "epoch": 1.4828767123287672,
      "grad_norm": 4.09375,
      "learning_rate": 1.7190876019980329e-06,
      "loss": 1.52543058,
      "memory(GiB)": 117.38,
      "step": 58455,
      "train_speed(iter/s)": 1.637027
    },
    {
      "acc": 0.65292039,
      "epoch": 1.4830035514967022,
      "grad_norm": 5.5,
      "learning_rate": 1.7182963803720987e-06,
      "loss": 1.58325291,
      "memory(GiB)": 117.38,
      "step": 58460,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.63819237,
      "epoch": 1.4831303906646371,
      "grad_norm": 7.0625,
      "learning_rate": 1.7175053030901418e-06,
      "loss": 1.67336922,
      "memory(GiB)": 117.38,
      "step": 58465,
      "train_speed(iter/s)": 1.637059
    },
    {
      "acc": 0.66205082,
      "epoch": 1.4832572298325724,
      "grad_norm": 5.8125,
      "learning_rate": 1.7167143701869582e-06,
      "loss": 1.59941053,
      "memory(GiB)": 117.38,
      "step": 58470,
      "train_speed(iter/s)": 1.637074
    },
    {
      "acc": 0.65847273,
      "epoch": 1.4833840690005073,
      "grad_norm": 6.53125,
      "learning_rate": 1.7159235816973318e-06,
      "loss": 1.5616128,
      "memory(GiB)": 117.38,
      "step": 58475,
      "train_speed(iter/s)": 1.637091
    },
    {
      "acc": 0.66114902,
      "epoch": 1.4835109081684423,
      "grad_norm": 4.875,
      "learning_rate": 1.7151329376560506e-06,
      "loss": 1.62488689,
      "memory(GiB)": 117.38,
      "step": 58480,
      "train_speed(iter/s)": 1.637106
    },
    {
      "acc": 0.65020003,
      "epoch": 1.4836377473363775,
      "grad_norm": 4.625,
      "learning_rate": 1.7143424380978885e-06,
      "loss": 1.59014368,
      "memory(GiB)": 117.38,
      "step": 58485,
      "train_speed(iter/s)": 1.637122
    },
    {
      "acc": 0.65273342,
      "epoch": 1.4837645865043125,
      "grad_norm": 7.34375,
      "learning_rate": 1.7135520830576157e-06,
      "loss": 1.66563416,
      "memory(GiB)": 117.38,
      "step": 58490,
      "train_speed(iter/s)": 1.637137
    },
    {
      "acc": 0.66002617,
      "epoch": 1.4838914256722475,
      "grad_norm": 5.15625,
      "learning_rate": 1.712761872569995e-06,
      "loss": 1.58130112,
      "memory(GiB)": 117.38,
      "step": 58495,
      "train_speed(iter/s)": 1.637153
    },
    {
      "acc": 0.65367932,
      "epoch": 1.4840182648401825,
      "grad_norm": 5.375,
      "learning_rate": 1.7119718066697838e-06,
      "loss": 1.64396267,
      "memory(GiB)": 117.38,
      "step": 58500,
      "train_speed(iter/s)": 1.637169
    },
    {
      "acc": 0.65111408,
      "epoch": 1.4841451040081177,
      "grad_norm": 5.3125,
      "learning_rate": 1.7111818853917323e-06,
      "loss": 1.59455919,
      "memory(GiB)": 117.38,
      "step": 58505,
      "train_speed(iter/s)": 1.637184
    },
    {
      "acc": 0.6549758,
      "epoch": 1.4842719431760527,
      "grad_norm": 5.21875,
      "learning_rate": 1.710392108770585e-06,
      "loss": 1.65722713,
      "memory(GiB)": 117.38,
      "step": 58510,
      "train_speed(iter/s)": 1.6372
    },
    {
      "acc": 0.65516052,
      "epoch": 1.484398782343988,
      "grad_norm": 5.625,
      "learning_rate": 1.7096024768410796e-06,
      "loss": 1.65799637,
      "memory(GiB)": 117.38,
      "step": 58515,
      "train_speed(iter/s)": 1.637216
    },
    {
      "acc": 0.64082503,
      "epoch": 1.484525621511923,
      "grad_norm": 4.6875,
      "learning_rate": 1.7088129896379484e-06,
      "loss": 1.66720009,
      "memory(GiB)": 117.38,
      "step": 58520,
      "train_speed(iter/s)": 1.637231
    },
    {
      "acc": 0.63991313,
      "epoch": 1.484652460679858,
      "grad_norm": 5.375,
      "learning_rate": 1.7080236471959155e-06,
      "loss": 1.61568604,
      "memory(GiB)": 117.38,
      "step": 58525,
      "train_speed(iter/s)": 1.637247
    },
    {
      "acc": 0.65633616,
      "epoch": 1.4847792998477929,
      "grad_norm": 6.09375,
      "learning_rate": 1.7072344495497007e-06,
      "loss": 1.61981888,
      "memory(GiB)": 117.38,
      "step": 58530,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.68699226,
      "epoch": 1.484906139015728,
      "grad_norm": 5.6875,
      "learning_rate": 1.7064453967340155e-06,
      "loss": 1.50519199,
      "memory(GiB)": 117.38,
      "step": 58535,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.65298004,
      "epoch": 1.485032978183663,
      "grad_norm": 6.0,
      "learning_rate": 1.7056564887835664e-06,
      "loss": 1.5819809,
      "memory(GiB)": 117.38,
      "step": 58540,
      "train_speed(iter/s)": 1.637294
    },
    {
      "acc": 0.64004951,
      "epoch": 1.4851598173515983,
      "grad_norm": 5.65625,
      "learning_rate": 1.704867725733052e-06,
      "loss": 1.69057503,
      "memory(GiB)": 117.38,
      "step": 58545,
      "train_speed(iter/s)": 1.637311
    },
    {
      "acc": 0.65730515,
      "epoch": 1.4852866565195333,
      "grad_norm": 6.53125,
      "learning_rate": 1.7040791076171692e-06,
      "loss": 1.55900059,
      "memory(GiB)": 117.38,
      "step": 58550,
      "train_speed(iter/s)": 1.637327
    },
    {
      "acc": 0.65295768,
      "epoch": 1.4854134956874683,
      "grad_norm": 5.28125,
      "learning_rate": 1.7032906344706017e-06,
      "loss": 1.60283566,
      "memory(GiB)": 117.38,
      "step": 58555,
      "train_speed(iter/s)": 1.637343
    },
    {
      "acc": 0.63999052,
      "epoch": 1.4855403348554033,
      "grad_norm": 5.375,
      "learning_rate": 1.7025023063280306e-06,
      "loss": 1.64884109,
      "memory(GiB)": 117.38,
      "step": 58560,
      "train_speed(iter/s)": 1.63736
    },
    {
      "acc": 0.64783883,
      "epoch": 1.4856671740233385,
      "grad_norm": 5.53125,
      "learning_rate": 1.701714123224128e-06,
      "loss": 1.65695744,
      "memory(GiB)": 117.38,
      "step": 58565,
      "train_speed(iter/s)": 1.637375
    },
    {
      "acc": 0.64992628,
      "epoch": 1.4857940131912735,
      "grad_norm": 6.25,
      "learning_rate": 1.7009260851935684e-06,
      "loss": 1.6325901,
      "memory(GiB)": 117.38,
      "step": 58570,
      "train_speed(iter/s)": 1.637392
    },
    {
      "acc": 0.63948412,
      "epoch": 1.4859208523592085,
      "grad_norm": 5.0,
      "learning_rate": 1.7001381922710064e-06,
      "loss": 1.70411797,
      "memory(GiB)": 117.38,
      "step": 58575,
      "train_speed(iter/s)": 1.637407
    },
    {
      "acc": 0.65391884,
      "epoch": 1.4860476915271437,
      "grad_norm": 6.125,
      "learning_rate": 1.699350444491098e-06,
      "loss": 1.58855133,
      "memory(GiB)": 117.38,
      "step": 58580,
      "train_speed(iter/s)": 1.637422
    },
    {
      "acc": 0.64425001,
      "epoch": 1.4861745306950787,
      "grad_norm": 5.25,
      "learning_rate": 1.6985628418884953e-06,
      "loss": 1.63907661,
      "memory(GiB)": 117.38,
      "step": 58585,
      "train_speed(iter/s)": 1.637438
    },
    {
      "acc": 0.66493506,
      "epoch": 1.4863013698630136,
      "grad_norm": 5.84375,
      "learning_rate": 1.6977753844978406e-06,
      "loss": 1.59763718,
      "memory(GiB)": 117.38,
      "step": 58590,
      "train_speed(iter/s)": 1.637454
    },
    {
      "acc": 0.65835514,
      "epoch": 1.4864282090309486,
      "grad_norm": 5.6875,
      "learning_rate": 1.696988072353764e-06,
      "loss": 1.58148785,
      "memory(GiB)": 117.38,
      "step": 58595,
      "train_speed(iter/s)": 1.63747
    },
    {
      "acc": 0.64658823,
      "epoch": 1.4865550481988838,
      "grad_norm": 5.46875,
      "learning_rate": 1.6962009054909007e-06,
      "loss": 1.62080841,
      "memory(GiB)": 117.38,
      "step": 58600,
      "train_speed(iter/s)": 1.637487
    },
    {
      "acc": 0.66238937,
      "epoch": 1.4866818873668188,
      "grad_norm": 4.5625,
      "learning_rate": 1.6954138839438723e-06,
      "loss": 1.56551476,
      "memory(GiB)": 117.38,
      "step": 58605,
      "train_speed(iter/s)": 1.637504
    },
    {
      "acc": 0.65568867,
      "epoch": 1.486808726534754,
      "grad_norm": 5.96875,
      "learning_rate": 1.6946270077472966e-06,
      "loss": 1.63482723,
      "memory(GiB)": 117.38,
      "step": 58610,
      "train_speed(iter/s)": 1.63752
    },
    {
      "acc": 0.66745057,
      "epoch": 1.486935565702689,
      "grad_norm": 5.375,
      "learning_rate": 1.6938402769357787e-06,
      "loss": 1.53524981,
      "memory(GiB)": 117.38,
      "step": 58615,
      "train_speed(iter/s)": 1.637537
    },
    {
      "acc": 0.65915036,
      "epoch": 1.487062404870624,
      "grad_norm": 10.0625,
      "learning_rate": 1.6930536915439288e-06,
      "loss": 1.60668488,
      "memory(GiB)": 117.38,
      "step": 58620,
      "train_speed(iter/s)": 1.637552
    },
    {
      "acc": 0.65312009,
      "epoch": 1.487189244038559,
      "grad_norm": 6.625,
      "learning_rate": 1.6922672516063415e-06,
      "loss": 1.60617638,
      "memory(GiB)": 117.38,
      "step": 58625,
      "train_speed(iter/s)": 1.637568
    },
    {
      "acc": 0.6509902,
      "epoch": 1.4873160832064942,
      "grad_norm": 4.84375,
      "learning_rate": 1.6914809571576086e-06,
      "loss": 1.59868441,
      "memory(GiB)": 117.38,
      "step": 58630,
      "train_speed(iter/s)": 1.637584
    },
    {
      "acc": 0.6462038,
      "epoch": 1.4874429223744292,
      "grad_norm": 5.3125,
      "learning_rate": 1.6906948082323149e-06,
      "loss": 1.69357262,
      "memory(GiB)": 117.38,
      "step": 58635,
      "train_speed(iter/s)": 1.6376
    },
    {
      "acc": 0.65631628,
      "epoch": 1.4875697615423642,
      "grad_norm": 5.59375,
      "learning_rate": 1.689908804865038e-06,
      "loss": 1.64427738,
      "memory(GiB)": 117.38,
      "step": 58640,
      "train_speed(iter/s)": 1.637616
    },
    {
      "acc": 0.65549636,
      "epoch": 1.4876966007102994,
      "grad_norm": 5.375,
      "learning_rate": 1.6891229470903509e-06,
      "loss": 1.64407234,
      "memory(GiB)": 117.38,
      "step": 58645,
      "train_speed(iter/s)": 1.637631
    },
    {
      "acc": 0.647651,
      "epoch": 1.4878234398782344,
      "grad_norm": 6.15625,
      "learning_rate": 1.6883372349428184e-06,
      "loss": 1.67279606,
      "memory(GiB)": 117.38,
      "step": 58650,
      "train_speed(iter/s)": 1.637649
    },
    {
      "acc": 0.67380109,
      "epoch": 1.4879502790461694,
      "grad_norm": 6.46875,
      "learning_rate": 1.6875516684569999e-06,
      "loss": 1.56411819,
      "memory(GiB)": 117.38,
      "step": 58655,
      "train_speed(iter/s)": 1.637665
    },
    {
      "acc": 0.66175508,
      "epoch": 1.4880771182141044,
      "grad_norm": 5.5625,
      "learning_rate": 1.686766247667448e-06,
      "loss": 1.63842773,
      "memory(GiB)": 117.38,
      "step": 58660,
      "train_speed(iter/s)": 1.637681
    },
    {
      "acc": 0.65478916,
      "epoch": 1.4882039573820396,
      "grad_norm": 6.03125,
      "learning_rate": 1.685980972608709e-06,
      "loss": 1.6224247,
      "memory(GiB)": 117.38,
      "step": 58665,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.65100369,
      "epoch": 1.4883307965499746,
      "grad_norm": 5.0,
      "learning_rate": 1.6851958433153227e-06,
      "loss": 1.5648324,
      "memory(GiB)": 117.38,
      "step": 58670,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.64601846,
      "epoch": 1.4884576357179098,
      "grad_norm": 6.34375,
      "learning_rate": 1.684410859821823e-06,
      "loss": 1.67129402,
      "memory(GiB)": 117.38,
      "step": 58675,
      "train_speed(iter/s)": 1.637727
    },
    {
      "acc": 0.67288537,
      "epoch": 1.4885844748858448,
      "grad_norm": 5.1875,
      "learning_rate": 1.6836260221627364e-06,
      "loss": 1.56544971,
      "memory(GiB)": 117.38,
      "step": 58680,
      "train_speed(iter/s)": 1.637741
    },
    {
      "acc": 0.65607591,
      "epoch": 1.4887113140537798,
      "grad_norm": 5.15625,
      "learning_rate": 1.682841330372582e-06,
      "loss": 1.5784235,
      "memory(GiB)": 117.38,
      "step": 58685,
      "train_speed(iter/s)": 1.637755
    },
    {
      "acc": 0.66048188,
      "epoch": 1.4888381532217148,
      "grad_norm": 4.5625,
      "learning_rate": 1.6820567844858798e-06,
      "loss": 1.52764893,
      "memory(GiB)": 117.38,
      "step": 58690,
      "train_speed(iter/s)": 1.63777
    },
    {
      "acc": 0.65201206,
      "epoch": 1.48896499238965,
      "grad_norm": 6.03125,
      "learning_rate": 1.681272384537132e-06,
      "loss": 1.58908329,
      "memory(GiB)": 117.38,
      "step": 58695,
      "train_speed(iter/s)": 1.637783
    },
    {
      "acc": 0.6660645,
      "epoch": 1.489091831557585,
      "grad_norm": 5.375,
      "learning_rate": 1.6804881305608423e-06,
      "loss": 1.54748325,
      "memory(GiB)": 117.38,
      "step": 58700,
      "train_speed(iter/s)": 1.637796
    },
    {
      "acc": 0.64187741,
      "epoch": 1.4892186707255202,
      "grad_norm": 7.03125,
      "learning_rate": 1.679704022591503e-06,
      "loss": 1.65535927,
      "memory(GiB)": 117.38,
      "step": 58705,
      "train_speed(iter/s)": 1.637813
    },
    {
      "acc": 0.64683318,
      "epoch": 1.4893455098934552,
      "grad_norm": 6.34375,
      "learning_rate": 1.678920060663608e-06,
      "loss": 1.60943108,
      "memory(GiB)": 117.38,
      "step": 58710,
      "train_speed(iter/s)": 1.637828
    },
    {
      "acc": 0.65541849,
      "epoch": 1.4894723490613901,
      "grad_norm": 5.34375,
      "learning_rate": 1.6781362448116344e-06,
      "loss": 1.5401967,
      "memory(GiB)": 117.38,
      "step": 58715,
      "train_speed(iter/s)": 1.637844
    },
    {
      "acc": 0.64753494,
      "epoch": 1.4895991882293251,
      "grad_norm": 5.0625,
      "learning_rate": 1.6773525750700586e-06,
      "loss": 1.61058998,
      "memory(GiB)": 117.38,
      "step": 58720,
      "train_speed(iter/s)": 1.63786
    },
    {
      "acc": 0.65694828,
      "epoch": 1.4897260273972603,
      "grad_norm": 6.34375,
      "learning_rate": 1.676569051473353e-06,
      "loss": 1.57637215,
      "memory(GiB)": 117.38,
      "step": 58725,
      "train_speed(iter/s)": 1.637876
    },
    {
      "acc": 0.65121851,
      "epoch": 1.4898528665651953,
      "grad_norm": 5.28125,
      "learning_rate": 1.6757856740559796e-06,
      "loss": 1.587288,
      "memory(GiB)": 117.38,
      "step": 58730,
      "train_speed(iter/s)": 1.637891
    },
    {
      "acc": 0.6652545,
      "epoch": 1.4899797057331303,
      "grad_norm": 5.125,
      "learning_rate": 1.6750024428523926e-06,
      "loss": 1.53108788,
      "memory(GiB)": 117.38,
      "step": 58735,
      "train_speed(iter/s)": 1.637907
    },
    {
      "acc": 0.66659002,
      "epoch": 1.4901065449010655,
      "grad_norm": 4.96875,
      "learning_rate": 1.6742193578970418e-06,
      "loss": 1.60661926,
      "memory(GiB)": 117.38,
      "step": 58740,
      "train_speed(iter/s)": 1.637923
    },
    {
      "acc": 0.66636882,
      "epoch": 1.4902333840690005,
      "grad_norm": 7.1875,
      "learning_rate": 1.673436419224373e-06,
      "loss": 1.5998848,
      "memory(GiB)": 117.38,
      "step": 58745,
      "train_speed(iter/s)": 1.637939
    },
    {
      "acc": 0.64998679,
      "epoch": 1.4903602232369355,
      "grad_norm": 5.25,
      "learning_rate": 1.6726536268688248e-06,
      "loss": 1.68369675,
      "memory(GiB)": 117.38,
      "step": 58750,
      "train_speed(iter/s)": 1.637955
    },
    {
      "acc": 0.6552546,
      "epoch": 1.4904870624048705,
      "grad_norm": 6.71875,
      "learning_rate": 1.671870980864822e-06,
      "loss": 1.62197609,
      "memory(GiB)": 117.38,
      "step": 58755,
      "train_speed(iter/s)": 1.63797
    },
    {
      "acc": 0.66752291,
      "epoch": 1.4906139015728057,
      "grad_norm": 5.84375,
      "learning_rate": 1.6710884812467943e-06,
      "loss": 1.50893412,
      "memory(GiB)": 117.38,
      "step": 58760,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.65801306,
      "epoch": 1.4907407407407407,
      "grad_norm": 5.6875,
      "learning_rate": 1.6703061280491579e-06,
      "loss": 1.62512569,
      "memory(GiB)": 117.38,
      "step": 58765,
      "train_speed(iter/s)": 1.638001
    },
    {
      "acc": 0.66275606,
      "epoch": 1.490867579908676,
      "grad_norm": 5.40625,
      "learning_rate": 1.6695239213063237e-06,
      "loss": 1.56917524,
      "memory(GiB)": 117.38,
      "step": 58770,
      "train_speed(iter/s)": 1.638017
    },
    {
      "acc": 0.66071,
      "epoch": 1.490994419076611,
      "grad_norm": 6.5,
      "learning_rate": 1.6687418610526972e-06,
      "loss": 1.59299698,
      "memory(GiB)": 117.38,
      "step": 58775,
      "train_speed(iter/s)": 1.638033
    },
    {
      "acc": 0.6698513,
      "epoch": 1.4911212582445459,
      "grad_norm": 6.25,
      "learning_rate": 1.6679599473226766e-06,
      "loss": 1.51940384,
      "memory(GiB)": 117.38,
      "step": 58780,
      "train_speed(iter/s)": 1.638047
    },
    {
      "acc": 0.65332928,
      "epoch": 1.4912480974124809,
      "grad_norm": 6.46875,
      "learning_rate": 1.6671781801506536e-06,
      "loss": 1.65746822,
      "memory(GiB)": 117.38,
      "step": 58785,
      "train_speed(iter/s)": 1.638062
    },
    {
      "acc": 0.65322771,
      "epoch": 1.491374936580416,
      "grad_norm": 4.6875,
      "learning_rate": 1.6663965595710147e-06,
      "loss": 1.58195763,
      "memory(GiB)": 117.38,
      "step": 58790,
      "train_speed(iter/s)": 1.638078
    },
    {
      "acc": 0.66882024,
      "epoch": 1.491501775748351,
      "grad_norm": 7.28125,
      "learning_rate": 1.6656150856181386e-06,
      "loss": 1.5820878,
      "memory(GiB)": 117.38,
      "step": 58795,
      "train_speed(iter/s)": 1.638094
    },
    {
      "acc": 0.65017233,
      "epoch": 1.491628614916286,
      "grad_norm": 5.375,
      "learning_rate": 1.6648337583263974e-06,
      "loss": 1.65157356,
      "memory(GiB)": 117.38,
      "step": 58800,
      "train_speed(iter/s)": 1.63811
    },
    {
      "acc": 0.65503259,
      "epoch": 1.4917554540842213,
      "grad_norm": 5.1875,
      "learning_rate": 1.6640525777301586e-06,
      "loss": 1.61745586,
      "memory(GiB)": 117.38,
      "step": 58805,
      "train_speed(iter/s)": 1.638125
    },
    {
      "acc": 0.66357312,
      "epoch": 1.4918822932521563,
      "grad_norm": 6.65625,
      "learning_rate": 1.663271543863781e-06,
      "loss": 1.59421711,
      "memory(GiB)": 117.38,
      "step": 58810,
      "train_speed(iter/s)": 1.638142
    },
    {
      "acc": 0.66036906,
      "epoch": 1.4920091324200913,
      "grad_norm": 6.3125,
      "learning_rate": 1.6624906567616183e-06,
      "loss": 1.63050842,
      "memory(GiB)": 117.38,
      "step": 58815,
      "train_speed(iter/s)": 1.638157
    },
    {
      "acc": 0.65754991,
      "epoch": 1.4921359715880262,
      "grad_norm": 5.59375,
      "learning_rate": 1.6617099164580175e-06,
      "loss": 1.56847639,
      "memory(GiB)": 117.38,
      "step": 58820,
      "train_speed(iter/s)": 1.638173
    },
    {
      "acc": 0.65467749,
      "epoch": 1.4922628107559615,
      "grad_norm": 6.4375,
      "learning_rate": 1.660929322987319e-06,
      "loss": 1.58660259,
      "memory(GiB)": 117.38,
      "step": 58825,
      "train_speed(iter/s)": 1.638188
    },
    {
      "acc": 0.65053029,
      "epoch": 1.4923896499238964,
      "grad_norm": 7.0,
      "learning_rate": 1.660148876383857e-06,
      "loss": 1.59802055,
      "memory(GiB)": 117.38,
      "step": 58830,
      "train_speed(iter/s)": 1.638204
    },
    {
      "acc": 0.66431561,
      "epoch": 1.4925164890918317,
      "grad_norm": 6.0,
      "learning_rate": 1.6593685766819584e-06,
      "loss": 1.55291405,
      "memory(GiB)": 117.38,
      "step": 58835,
      "train_speed(iter/s)": 1.63822
    },
    {
      "acc": 0.65234203,
      "epoch": 1.4926433282597666,
      "grad_norm": 5.34375,
      "learning_rate": 1.658588423915945e-06,
      "loss": 1.56638832,
      "memory(GiB)": 117.38,
      "step": 58840,
      "train_speed(iter/s)": 1.638235
    },
    {
      "acc": 0.66264772,
      "epoch": 1.4927701674277016,
      "grad_norm": 5.125,
      "learning_rate": 1.6578084181201293e-06,
      "loss": 1.57011166,
      "memory(GiB)": 117.38,
      "step": 58845,
      "train_speed(iter/s)": 1.63825
    },
    {
      "acc": 0.65326414,
      "epoch": 1.4928970065956366,
      "grad_norm": 6.25,
      "learning_rate": 1.6570285593288242e-06,
      "loss": 1.64257793,
      "memory(GiB)": 117.38,
      "step": 58850,
      "train_speed(iter/s)": 1.638265
    },
    {
      "acc": 0.64946322,
      "epoch": 1.4930238457635718,
      "grad_norm": 6.375,
      "learning_rate": 1.6562488475763267e-06,
      "loss": 1.59806156,
      "memory(GiB)": 117.38,
      "step": 58855,
      "train_speed(iter/s)": 1.638281
    },
    {
      "acc": 0.65304618,
      "epoch": 1.4931506849315068,
      "grad_norm": 7.40625,
      "learning_rate": 1.6554692828969321e-06,
      "loss": 1.62616596,
      "memory(GiB)": 117.38,
      "step": 58860,
      "train_speed(iter/s)": 1.638296
    },
    {
      "acc": 0.64253006,
      "epoch": 1.493277524099442,
      "grad_norm": 7.65625,
      "learning_rate": 1.6546898653249326e-06,
      "loss": 1.60036926,
      "memory(GiB)": 117.38,
      "step": 58865,
      "train_speed(iter/s)": 1.638311
    },
    {
      "acc": 0.64423571,
      "epoch": 1.493404363267377,
      "grad_norm": 5.46875,
      "learning_rate": 1.65391059489461e-06,
      "loss": 1.68086586,
      "memory(GiB)": 117.38,
      "step": 58870,
      "train_speed(iter/s)": 1.638327
    },
    {
      "acc": 0.64976792,
      "epoch": 1.493531202435312,
      "grad_norm": 4.6875,
      "learning_rate": 1.6531314716402369e-06,
      "loss": 1.60481491,
      "memory(GiB)": 117.38,
      "step": 58875,
      "train_speed(iter/s)": 1.638343
    },
    {
      "acc": 0.6548851,
      "epoch": 1.493658041603247,
      "grad_norm": 5.8125,
      "learning_rate": 1.652352495596083e-06,
      "loss": 1.59254417,
      "memory(GiB)": 117.38,
      "step": 58880,
      "train_speed(iter/s)": 1.638358
    },
    {
      "acc": 0.65525436,
      "epoch": 1.4937848807711822,
      "grad_norm": 5.28125,
      "learning_rate": 1.6515736667964144e-06,
      "loss": 1.59278698,
      "memory(GiB)": 117.38,
      "step": 58885,
      "train_speed(iter/s)": 1.638373
    },
    {
      "acc": 0.65833759,
      "epoch": 1.4939117199391172,
      "grad_norm": 6.0625,
      "learning_rate": 1.6507949852754867e-06,
      "loss": 1.64661064,
      "memory(GiB)": 117.38,
      "step": 58890,
      "train_speed(iter/s)": 1.638389
    },
    {
      "acc": 0.65120153,
      "epoch": 1.4940385591070522,
      "grad_norm": 5.21875,
      "learning_rate": 1.6500164510675453e-06,
      "loss": 1.61090374,
      "memory(GiB)": 117.38,
      "step": 58895,
      "train_speed(iter/s)": 1.638404
    },
    {
      "acc": 0.66952782,
      "epoch": 1.4941653982749874,
      "grad_norm": 6.625,
      "learning_rate": 1.649238064206839e-06,
      "loss": 1.5535059,
      "memory(GiB)": 117.38,
      "step": 58900,
      "train_speed(iter/s)": 1.638419
    },
    {
      "acc": 0.65716972,
      "epoch": 1.4942922374429224,
      "grad_norm": 5.03125,
      "learning_rate": 1.6484598247276023e-06,
      "loss": 1.60048904,
      "memory(GiB)": 117.38,
      "step": 58905,
      "train_speed(iter/s)": 1.638434
    },
    {
      "acc": 0.63739233,
      "epoch": 1.4944190766108574,
      "grad_norm": 5.6875,
      "learning_rate": 1.6476817326640682e-06,
      "loss": 1.6312542,
      "memory(GiB)": 117.38,
      "step": 58910,
      "train_speed(iter/s)": 1.638449
    },
    {
      "acc": 0.66941986,
      "epoch": 1.4945459157787924,
      "grad_norm": 7.3125,
      "learning_rate": 1.646903788050455e-06,
      "loss": 1.54043446,
      "memory(GiB)": 117.38,
      "step": 58915,
      "train_speed(iter/s)": 1.638464
    },
    {
      "acc": 0.65407443,
      "epoch": 1.4946727549467276,
      "grad_norm": 5.0,
      "learning_rate": 1.6461259909209853e-06,
      "loss": 1.58598537,
      "memory(GiB)": 117.38,
      "step": 58920,
      "train_speed(iter/s)": 1.638479
    },
    {
      "acc": 0.66881928,
      "epoch": 1.4947995941146626,
      "grad_norm": 5.4375,
      "learning_rate": 1.6453483413098687e-06,
      "loss": 1.59661674,
      "memory(GiB)": 117.38,
      "step": 58925,
      "train_speed(iter/s)": 1.638495
    },
    {
      "acc": 0.65527649,
      "epoch": 1.4949264332825978,
      "grad_norm": 5.53125,
      "learning_rate": 1.6445708392513093e-06,
      "loss": 1.64659462,
      "memory(GiB)": 117.38,
      "step": 58930,
      "train_speed(iter/s)": 1.63851
    },
    {
      "acc": 0.64791303,
      "epoch": 1.4950532724505328,
      "grad_norm": 5.375,
      "learning_rate": 1.643793484779505e-06,
      "loss": 1.67163944,
      "memory(GiB)": 117.38,
      "step": 58935,
      "train_speed(iter/s)": 1.638524
    },
    {
      "acc": 0.66956196,
      "epoch": 1.4951801116184678,
      "grad_norm": 7.46875,
      "learning_rate": 1.6430162779286484e-06,
      "loss": 1.56772776,
      "memory(GiB)": 117.38,
      "step": 58940,
      "train_speed(iter/s)": 1.638539
    },
    {
      "acc": 0.6662281,
      "epoch": 1.4953069507864027,
      "grad_norm": 4.84375,
      "learning_rate": 1.6422392187329233e-06,
      "loss": 1.53606396,
      "memory(GiB)": 117.38,
      "step": 58945,
      "train_speed(iter/s)": 1.638555
    },
    {
      "acc": 0.64550695,
      "epoch": 1.495433789954338,
      "grad_norm": 7.15625,
      "learning_rate": 1.6414623072265085e-06,
      "loss": 1.69089851,
      "memory(GiB)": 117.38,
      "step": 58950,
      "train_speed(iter/s)": 1.63857
    },
    {
      "acc": 0.66228223,
      "epoch": 1.495560629122273,
      "grad_norm": 5.65625,
      "learning_rate": 1.6406855434435765e-06,
      "loss": 1.58152933,
      "memory(GiB)": 117.38,
      "step": 58955,
      "train_speed(iter/s)": 1.638586
    },
    {
      "acc": 0.64446435,
      "epoch": 1.495687468290208,
      "grad_norm": 4.90625,
      "learning_rate": 1.6399089274182922e-06,
      "loss": 1.62432594,
      "memory(GiB)": 117.38,
      "step": 58960,
      "train_speed(iter/s)": 1.638601
    },
    {
      "acc": 0.66495709,
      "epoch": 1.4958143074581431,
      "grad_norm": 4.75,
      "learning_rate": 1.6391324591848156e-06,
      "loss": 1.59021397,
      "memory(GiB)": 117.38,
      "step": 58965,
      "train_speed(iter/s)": 1.638617
    },
    {
      "acc": 0.66466942,
      "epoch": 1.4959411466260781,
      "grad_norm": 5.375,
      "learning_rate": 1.6383561387772984e-06,
      "loss": 1.5989727,
      "memory(GiB)": 117.38,
      "step": 58970,
      "train_speed(iter/s)": 1.638632
    },
    {
      "acc": 0.65867825,
      "epoch": 1.4960679857940131,
      "grad_norm": 5.1875,
      "learning_rate": 1.6375799662298868e-06,
      "loss": 1.59532871,
      "memory(GiB)": 117.38,
      "step": 58975,
      "train_speed(iter/s)": 1.638647
    },
    {
      "acc": 0.64182115,
      "epoch": 1.4961948249619481,
      "grad_norm": 6.125,
      "learning_rate": 1.6368039415767201e-06,
      "loss": 1.64483757,
      "memory(GiB)": 117.38,
      "step": 58980,
      "train_speed(iter/s)": 1.638662
    },
    {
      "acc": 0.64847851,
      "epoch": 1.4963216641298833,
      "grad_norm": 5.8125,
      "learning_rate": 1.6360280648519305e-06,
      "loss": 1.61589432,
      "memory(GiB)": 117.38,
      "step": 58985,
      "train_speed(iter/s)": 1.638678
    },
    {
      "acc": 0.67074327,
      "epoch": 1.4964485032978183,
      "grad_norm": 6.09375,
      "learning_rate": 1.6352523360896488e-06,
      "loss": 1.54374924,
      "memory(GiB)": 117.38,
      "step": 58990,
      "train_speed(iter/s)": 1.638692
    },
    {
      "acc": 0.65385571,
      "epoch": 1.4965753424657535,
      "grad_norm": 5.09375,
      "learning_rate": 1.63447675532399e-06,
      "loss": 1.61351757,
      "memory(GiB)": 117.38,
      "step": 58995,
      "train_speed(iter/s)": 1.638708
    },
    {
      "acc": 0.66785917,
      "epoch": 1.4967021816336885,
      "grad_norm": 6.03125,
      "learning_rate": 1.6337013225890698e-06,
      "loss": 1.64189816,
      "memory(GiB)": 117.38,
      "step": 59000,
      "train_speed(iter/s)": 1.638723
    },
    {
      "epoch": 1.4967021816336885,
      "eval_acc": 0.6462726868355271,
      "eval_loss": 1.5733360052108765,
      "eval_runtime": 58.5083,
      "eval_samples_per_second": 108.873,
      "eval_steps_per_second": 27.227,
      "step": 59000
    },
    {
      "acc": 0.66507397,
      "epoch": 1.4968290208016235,
      "grad_norm": 4.71875,
      "learning_rate": 1.6329260379189932e-06,
      "loss": 1.5519351,
      "memory(GiB)": 117.38,
      "step": 59005,
      "train_speed(iter/s)": 1.635881
    },
    {
      "acc": 0.64430628,
      "epoch": 1.4969558599695585,
      "grad_norm": 5.78125,
      "learning_rate": 1.6321509013478653e-06,
      "loss": 1.58763018,
      "memory(GiB)": 117.38,
      "step": 59010,
      "train_speed(iter/s)": 1.635894
    },
    {
      "acc": 0.6571928,
      "epoch": 1.4970826991374937,
      "grad_norm": 5.28125,
      "learning_rate": 1.6313759129097757e-06,
      "loss": 1.62825317,
      "memory(GiB)": 117.38,
      "step": 59015,
      "train_speed(iter/s)": 1.635907
    },
    {
      "acc": 0.64946027,
      "epoch": 1.4972095383054287,
      "grad_norm": 5.53125,
      "learning_rate": 1.6306010726388117e-06,
      "loss": 1.6424118,
      "memory(GiB)": 117.38,
      "step": 59020,
      "train_speed(iter/s)": 1.635921
    },
    {
      "acc": 0.65573363,
      "epoch": 1.497336377473364,
      "grad_norm": 6.53125,
      "learning_rate": 1.6298263805690573e-06,
      "loss": 1.65517006,
      "memory(GiB)": 117.38,
      "step": 59025,
      "train_speed(iter/s)": 1.635935
    },
    {
      "acc": 0.65678153,
      "epoch": 1.4974632166412989,
      "grad_norm": 5.65625,
      "learning_rate": 1.629051836734587e-06,
      "loss": 1.58574963,
      "memory(GiB)": 117.38,
      "step": 59030,
      "train_speed(iter/s)": 1.635949
    },
    {
      "acc": 0.68190155,
      "epoch": 1.4975900558092339,
      "grad_norm": 7.03125,
      "learning_rate": 1.6282774411694641e-06,
      "loss": 1.54179535,
      "memory(GiB)": 117.38,
      "step": 59035,
      "train_speed(iter/s)": 1.635964
    },
    {
      "acc": 0.66086006,
      "epoch": 1.4977168949771689,
      "grad_norm": 7.0,
      "learning_rate": 1.6275031939077545e-06,
      "loss": 1.60549583,
      "memory(GiB)": 117.38,
      "step": 59040,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.66562696,
      "epoch": 1.497843734145104,
      "grad_norm": 5.375,
      "learning_rate": 1.6267290949835119e-06,
      "loss": 1.52672119,
      "memory(GiB)": 117.38,
      "step": 59045,
      "train_speed(iter/s)": 1.635992
    },
    {
      "acc": 0.64777427,
      "epoch": 1.497970573313039,
      "grad_norm": 5.0625,
      "learning_rate": 1.6259551444307852e-06,
      "loss": 1.5901722,
      "memory(GiB)": 117.38,
      "step": 59050,
      "train_speed(iter/s)": 1.636007
    },
    {
      "acc": 0.64258065,
      "epoch": 1.498097412480974,
      "grad_norm": 5.75,
      "learning_rate": 1.6251813422836127e-06,
      "loss": 1.62169666,
      "memory(GiB)": 117.38,
      "step": 59055,
      "train_speed(iter/s)": 1.636023
    },
    {
      "acc": 0.65736609,
      "epoch": 1.4982242516489093,
      "grad_norm": 6.71875,
      "learning_rate": 1.6244076885760334e-06,
      "loss": 1.56433506,
      "memory(GiB)": 117.38,
      "step": 59060,
      "train_speed(iter/s)": 1.636038
    },
    {
      "acc": 0.65047703,
      "epoch": 1.4983510908168443,
      "grad_norm": 5.75,
      "learning_rate": 1.6236341833420755e-06,
      "loss": 1.66114349,
      "memory(GiB)": 117.38,
      "step": 59065,
      "train_speed(iter/s)": 1.636049
    },
    {
      "acc": 0.65166845,
      "epoch": 1.4984779299847792,
      "grad_norm": 6.03125,
      "learning_rate": 1.6228608266157596e-06,
      "loss": 1.57607479,
      "memory(GiB)": 117.38,
      "step": 59070,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.66794491,
      "epoch": 1.4986047691527142,
      "grad_norm": 5.375,
      "learning_rate": 1.6220876184311034e-06,
      "loss": 1.56458521,
      "memory(GiB)": 117.38,
      "step": 59075,
      "train_speed(iter/s)": 1.63608
    },
    {
      "acc": 0.64701505,
      "epoch": 1.4987316083206494,
      "grad_norm": 7.90625,
      "learning_rate": 1.6213145588221146e-06,
      "loss": 1.57172394,
      "memory(GiB)": 117.38,
      "step": 59080,
      "train_speed(iter/s)": 1.636088
    },
    {
      "acc": 0.66153364,
      "epoch": 1.4988584474885844,
      "grad_norm": 8.1875,
      "learning_rate": 1.620541647822796e-06,
      "loss": 1.5491334,
      "memory(GiB)": 117.38,
      "step": 59085,
      "train_speed(iter/s)": 1.636103
    },
    {
      "acc": 0.65983672,
      "epoch": 1.4989852866565196,
      "grad_norm": 6.625,
      "learning_rate": 1.6197688854671444e-06,
      "loss": 1.60656357,
      "memory(GiB)": 117.38,
      "step": 59090,
      "train_speed(iter/s)": 1.636119
    },
    {
      "acc": 0.66430807,
      "epoch": 1.4991121258244546,
      "grad_norm": 5.34375,
      "learning_rate": 1.6189962717891484e-06,
      "loss": 1.52204475,
      "memory(GiB)": 117.38,
      "step": 59095,
      "train_speed(iter/s)": 1.636135
    },
    {
      "acc": 0.63608198,
      "epoch": 1.4992389649923896,
      "grad_norm": 5.5625,
      "learning_rate": 1.6182238068227917e-06,
      "loss": 1.68204269,
      "memory(GiB)": 117.38,
      "step": 59100,
      "train_speed(iter/s)": 1.63615
    },
    {
      "acc": 0.64643054,
      "epoch": 1.4993658041603246,
      "grad_norm": 5.53125,
      "learning_rate": 1.6174514906020505e-06,
      "loss": 1.62971668,
      "memory(GiB)": 117.38,
      "step": 59105,
      "train_speed(iter/s)": 1.636166
    },
    {
      "acc": 0.66912565,
      "epoch": 1.4994926433282598,
      "grad_norm": 4.71875,
      "learning_rate": 1.6166793231608952e-06,
      "loss": 1.5243782,
      "memory(GiB)": 117.38,
      "step": 59110,
      "train_speed(iter/s)": 1.63618
    },
    {
      "acc": 0.64613824,
      "epoch": 1.4996194824961948,
      "grad_norm": 5.65625,
      "learning_rate": 1.615907304533288e-06,
      "loss": 1.63399925,
      "memory(GiB)": 117.38,
      "step": 59115,
      "train_speed(iter/s)": 1.636195
    },
    {
      "acc": 0.65427275,
      "epoch": 1.4997463216641298,
      "grad_norm": 5.40625,
      "learning_rate": 1.6151354347531868e-06,
      "loss": 1.62950878,
      "memory(GiB)": 117.38,
      "step": 59120,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.66316547,
      "epoch": 1.499873160832065,
      "grad_norm": 6.0625,
      "learning_rate": 1.61436371385454e-06,
      "loss": 1.61837769,
      "memory(GiB)": 117.38,
      "step": 59125,
      "train_speed(iter/s)": 1.636227
    },
    {
      "acc": 0.65671606,
      "epoch": 1.5,
      "grad_norm": 6.21875,
      "learning_rate": 1.6135921418712959e-06,
      "loss": 1.60000172,
      "memory(GiB)": 117.38,
      "step": 59130,
      "train_speed(iter/s)": 1.636242
    },
    {
      "acc": 0.6654242,
      "epoch": 1.500126839167935,
      "grad_norm": 5.9375,
      "learning_rate": 1.6128207188373867e-06,
      "loss": 1.58890524,
      "memory(GiB)": 117.38,
      "step": 59135,
      "train_speed(iter/s)": 1.636257
    },
    {
      "acc": 0.64930043,
      "epoch": 1.50025367833587,
      "grad_norm": 5.0,
      "learning_rate": 1.6120494447867451e-06,
      "loss": 1.616535,
      "memory(GiB)": 117.38,
      "step": 59140,
      "train_speed(iter/s)": 1.636272
    },
    {
      "acc": 0.66388211,
      "epoch": 1.5003805175038052,
      "grad_norm": 9.0,
      "learning_rate": 1.6112783197532932e-06,
      "loss": 1.54436531,
      "memory(GiB)": 117.38,
      "step": 59145,
      "train_speed(iter/s)": 1.636286
    },
    {
      "acc": 0.65104041,
      "epoch": 1.5005073566717404,
      "grad_norm": 5.59375,
      "learning_rate": 1.6105073437709545e-06,
      "loss": 1.60603886,
      "memory(GiB)": 117.38,
      "step": 59150,
      "train_speed(iter/s)": 1.636301
    },
    {
      "acc": 0.64264188,
      "epoch": 1.5006341958396754,
      "grad_norm": 5.84375,
      "learning_rate": 1.6097365168736335e-06,
      "loss": 1.63858948,
      "memory(GiB)": 117.38,
      "step": 59155,
      "train_speed(iter/s)": 1.636317
    },
    {
      "acc": 0.64266996,
      "epoch": 1.5007610350076104,
      "grad_norm": 5.6875,
      "learning_rate": 1.6089658390952351e-06,
      "loss": 1.63354034,
      "memory(GiB)": 117.38,
      "step": 59160,
      "train_speed(iter/s)": 1.636332
    },
    {
      "acc": 0.65540562,
      "epoch": 1.5008878741755454,
      "grad_norm": 5.71875,
      "learning_rate": 1.6081953104696612e-06,
      "loss": 1.61162224,
      "memory(GiB)": 117.38,
      "step": 59165,
      "train_speed(iter/s)": 1.636347
    },
    {
      "acc": 0.66873426,
      "epoch": 1.5010147133434804,
      "grad_norm": 5.21875,
      "learning_rate": 1.6074249310308021e-06,
      "loss": 1.58953333,
      "memory(GiB)": 117.38,
      "step": 59170,
      "train_speed(iter/s)": 1.636361
    },
    {
      "acc": 0.6648479,
      "epoch": 1.5011415525114156,
      "grad_norm": 5.875,
      "learning_rate": 1.6066547008125399e-06,
      "loss": 1.55940399,
      "memory(GiB)": 117.38,
      "step": 59175,
      "train_speed(iter/s)": 1.636376
    },
    {
      "acc": 0.65285358,
      "epoch": 1.5012683916793506,
      "grad_norm": 6.6875,
      "learning_rate": 1.6058846198487522e-06,
      "loss": 1.6120573,
      "memory(GiB)": 117.38,
      "step": 59180,
      "train_speed(iter/s)": 1.636391
    },
    {
      "acc": 0.64392085,
      "epoch": 1.5013952308472858,
      "grad_norm": 5.21875,
      "learning_rate": 1.6051146881733142e-06,
      "loss": 1.5530077,
      "memory(GiB)": 117.38,
      "step": 59185,
      "train_speed(iter/s)": 1.636407
    },
    {
      "acc": 0.65379677,
      "epoch": 1.5015220700152208,
      "grad_norm": 6.9375,
      "learning_rate": 1.6043449058200916e-06,
      "loss": 1.61747742,
      "memory(GiB)": 117.38,
      "step": 59190,
      "train_speed(iter/s)": 1.636421
    },
    {
      "acc": 0.65701532,
      "epoch": 1.5016489091831557,
      "grad_norm": 5.59375,
      "learning_rate": 1.6035752728229364e-06,
      "loss": 1.59188852,
      "memory(GiB)": 117.38,
      "step": 59195,
      "train_speed(iter/s)": 1.636436
    },
    {
      "acc": 0.65050678,
      "epoch": 1.5017757483510907,
      "grad_norm": 4.9375,
      "learning_rate": 1.6028057892157067e-06,
      "loss": 1.69276161,
      "memory(GiB)": 117.38,
      "step": 59200,
      "train_speed(iter/s)": 1.636452
    },
    {
      "acc": 0.65947394,
      "epoch": 1.5019025875190257,
      "grad_norm": 6.21875,
      "learning_rate": 1.602036455032246e-06,
      "loss": 1.58296204,
      "memory(GiB)": 117.38,
      "step": 59205,
      "train_speed(iter/s)": 1.636466
    },
    {
      "acc": 0.65613728,
      "epoch": 1.502029426686961,
      "grad_norm": 6.40625,
      "learning_rate": 1.6012672703063925e-06,
      "loss": 1.63416252,
      "memory(GiB)": 117.38,
      "step": 59210,
      "train_speed(iter/s)": 1.636481
    },
    {
      "acc": 0.64894209,
      "epoch": 1.5021562658548961,
      "grad_norm": 6.25,
      "learning_rate": 1.600498235071979e-06,
      "loss": 1.64049835,
      "memory(GiB)": 117.38,
      "step": 59215,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.63984246,
      "epoch": 1.5022831050228311,
      "grad_norm": 5.625,
      "learning_rate": 1.5997293493628301e-06,
      "loss": 1.65934372,
      "memory(GiB)": 117.38,
      "step": 59220,
      "train_speed(iter/s)": 1.636511
    },
    {
      "acc": 0.65474558,
      "epoch": 1.5024099441907661,
      "grad_norm": 6.03125,
      "learning_rate": 1.598960613212766e-06,
      "loss": 1.60389862,
      "memory(GiB)": 117.38,
      "step": 59225,
      "train_speed(iter/s)": 1.636527
    },
    {
      "acc": 0.66082439,
      "epoch": 1.5025367833587011,
      "grad_norm": 5.90625,
      "learning_rate": 1.598192026655599e-06,
      "loss": 1.63202114,
      "memory(GiB)": 117.38,
      "step": 59230,
      "train_speed(iter/s)": 1.636542
    },
    {
      "acc": 0.64346704,
      "epoch": 1.502663622526636,
      "grad_norm": 5.6875,
      "learning_rate": 1.5974235897251344e-06,
      "loss": 1.60650558,
      "memory(GiB)": 117.38,
      "step": 59235,
      "train_speed(iter/s)": 1.636555
    },
    {
      "acc": 0.65211668,
      "epoch": 1.5027904616945713,
      "grad_norm": 7.1875,
      "learning_rate": 1.5966553024551717e-06,
      "loss": 1.63110733,
      "memory(GiB)": 117.38,
      "step": 59240,
      "train_speed(iter/s)": 1.636571
    },
    {
      "acc": 0.65274272,
      "epoch": 1.5029173008625063,
      "grad_norm": 5.53125,
      "learning_rate": 1.5958871648795032e-06,
      "loss": 1.62780304,
      "memory(GiB)": 117.38,
      "step": 59245,
      "train_speed(iter/s)": 1.636587
    },
    {
      "acc": 0.66678596,
      "epoch": 1.5030441400304415,
      "grad_norm": 5.59375,
      "learning_rate": 1.5951191770319164e-06,
      "loss": 1.55899258,
      "memory(GiB)": 117.38,
      "step": 59250,
      "train_speed(iter/s)": 1.636602
    },
    {
      "acc": 0.6561008,
      "epoch": 1.5031709791983765,
      "grad_norm": 8.3125,
      "learning_rate": 1.594351338946189e-06,
      "loss": 1.57421989,
      "memory(GiB)": 117.38,
      "step": 59255,
      "train_speed(iter/s)": 1.636618
    },
    {
      "acc": 0.66732426,
      "epoch": 1.5032978183663115,
      "grad_norm": 5.9375,
      "learning_rate": 1.5935836506560953e-06,
      "loss": 1.58876553,
      "memory(GiB)": 117.38,
      "step": 59260,
      "train_speed(iter/s)": 1.636633
    },
    {
      "acc": 0.65766935,
      "epoch": 1.5034246575342465,
      "grad_norm": 4.96875,
      "learning_rate": 1.5928161121954012e-06,
      "loss": 1.56830015,
      "memory(GiB)": 117.38,
      "step": 59265,
      "train_speed(iter/s)": 1.636647
    },
    {
      "acc": 0.6670042,
      "epoch": 1.5035514967021817,
      "grad_norm": 6.0,
      "learning_rate": 1.592048723597866e-06,
      "loss": 1.54631958,
      "memory(GiB)": 117.38,
      "step": 59270,
      "train_speed(iter/s)": 1.636662
    },
    {
      "acc": 0.67544322,
      "epoch": 1.5036783358701167,
      "grad_norm": 7.375,
      "learning_rate": 1.591281484897244e-06,
      "loss": 1.54305363,
      "memory(GiB)": 117.38,
      "step": 59275,
      "train_speed(iter/s)": 1.636678
    },
    {
      "acc": 0.66650209,
      "epoch": 1.5038051750380519,
      "grad_norm": 6.34375,
      "learning_rate": 1.5905143961272807e-06,
      "loss": 1.61135864,
      "memory(GiB)": 117.38,
      "step": 59280,
      "train_speed(iter/s)": 1.636693
    },
    {
      "acc": 0.66681828,
      "epoch": 1.5039320142059869,
      "grad_norm": 6.4375,
      "learning_rate": 1.5897474573217153e-06,
      "loss": 1.5059103,
      "memory(GiB)": 117.38,
      "step": 59285,
      "train_speed(iter/s)": 1.636707
    },
    {
      "acc": 0.64960423,
      "epoch": 1.5040588533739219,
      "grad_norm": 7.21875,
      "learning_rate": 1.588980668514285e-06,
      "loss": 1.60573177,
      "memory(GiB)": 117.38,
      "step": 59290,
      "train_speed(iter/s)": 1.636722
    },
    {
      "acc": 0.64444962,
      "epoch": 1.5041856925418569,
      "grad_norm": 5.3125,
      "learning_rate": 1.5882140297387127e-06,
      "loss": 1.58477182,
      "memory(GiB)": 117.38,
      "step": 59295,
      "train_speed(iter/s)": 1.636737
    },
    {
      "acc": 0.65833349,
      "epoch": 1.5043125317097918,
      "grad_norm": 5.375,
      "learning_rate": 1.5874475410287189e-06,
      "loss": 1.64329796,
      "memory(GiB)": 117.38,
      "step": 59300,
      "train_speed(iter/s)": 1.636752
    },
    {
      "acc": 0.66364546,
      "epoch": 1.504439370877727,
      "grad_norm": 5.625,
      "learning_rate": 1.586681202418019e-06,
      "loss": 1.59969435,
      "memory(GiB)": 117.38,
      "step": 59305,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.65419703,
      "epoch": 1.5045662100456623,
      "grad_norm": 6.40625,
      "learning_rate": 1.5859150139403212e-06,
      "loss": 1.69760265,
      "memory(GiB)": 117.38,
      "step": 59310,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.66216364,
      "epoch": 1.5046930492135973,
      "grad_norm": 5.71875,
      "learning_rate": 1.585148975629322e-06,
      "loss": 1.54835224,
      "memory(GiB)": 117.38,
      "step": 59315,
      "train_speed(iter/s)": 1.636795
    },
    {
      "acc": 0.65177455,
      "epoch": 1.5048198883815322,
      "grad_norm": 5.71875,
      "learning_rate": 1.5843830875187155e-06,
      "loss": 1.59192553,
      "memory(GiB)": 117.38,
      "step": 59320,
      "train_speed(iter/s)": 1.63681
    },
    {
      "acc": 0.66120262,
      "epoch": 1.5049467275494672,
      "grad_norm": 5.9375,
      "learning_rate": 1.583617349642192e-06,
      "loss": 1.63134613,
      "memory(GiB)": 117.38,
      "step": 59325,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.64850907,
      "epoch": 1.5050735667174022,
      "grad_norm": 4.8125,
      "learning_rate": 1.5828517620334322e-06,
      "loss": 1.63779488,
      "memory(GiB)": 117.38,
      "step": 59330,
      "train_speed(iter/s)": 1.636841
    },
    {
      "acc": 0.66626501,
      "epoch": 1.5052004058853374,
      "grad_norm": 5.15625,
      "learning_rate": 1.5820863247261054e-06,
      "loss": 1.59134541,
      "memory(GiB)": 117.38,
      "step": 59335,
      "train_speed(iter/s)": 1.636857
    },
    {
      "acc": 0.66665621,
      "epoch": 1.5053272450532724,
      "grad_norm": 5.46875,
      "learning_rate": 1.5813210377538834e-06,
      "loss": 1.58963318,
      "memory(GiB)": 117.38,
      "step": 59340,
      "train_speed(iter/s)": 1.636872
    },
    {
      "acc": 0.65015426,
      "epoch": 1.5054540842212076,
      "grad_norm": 5.1875,
      "learning_rate": 1.5805559011504252e-06,
      "loss": 1.64164085,
      "memory(GiB)": 117.38,
      "step": 59345,
      "train_speed(iter/s)": 1.636887
    },
    {
      "acc": 0.66221218,
      "epoch": 1.5055809233891426,
      "grad_norm": 5.40625,
      "learning_rate": 1.5797909149493873e-06,
      "loss": 1.51169243,
      "memory(GiB)": 117.38,
      "step": 59350,
      "train_speed(iter/s)": 1.636903
    },
    {
      "acc": 0.63912196,
      "epoch": 1.5057077625570776,
      "grad_norm": 5.6875,
      "learning_rate": 1.5790260791844114e-06,
      "loss": 1.70168667,
      "memory(GiB)": 117.38,
      "step": 59355,
      "train_speed(iter/s)": 1.636918
    },
    {
      "acc": 0.66426086,
      "epoch": 1.5058346017250126,
      "grad_norm": 5.21875,
      "learning_rate": 1.5782613938891438e-06,
      "loss": 1.57795753,
      "memory(GiB)": 117.38,
      "step": 59360,
      "train_speed(iter/s)": 1.636933
    },
    {
      "acc": 0.67836018,
      "epoch": 1.5059614408929476,
      "grad_norm": 7.46875,
      "learning_rate": 1.5774968590972172e-06,
      "loss": 1.58142729,
      "memory(GiB)": 117.38,
      "step": 59365,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.65407772,
      "epoch": 1.5060882800608828,
      "grad_norm": 6.46875,
      "learning_rate": 1.5767324748422592e-06,
      "loss": 1.60446033,
      "memory(GiB)": 117.38,
      "step": 59370,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.65910501,
      "epoch": 1.506215119228818,
      "grad_norm": 6.90625,
      "learning_rate": 1.5759682411578909e-06,
      "loss": 1.59832134,
      "memory(GiB)": 117.38,
      "step": 59375,
      "train_speed(iter/s)": 1.636978
    },
    {
      "acc": 0.66232061,
      "epoch": 1.506341958396753,
      "grad_norm": 5.5,
      "learning_rate": 1.575204158077726e-06,
      "loss": 1.63297272,
      "memory(GiB)": 117.38,
      "step": 59380,
      "train_speed(iter/s)": 1.636994
    },
    {
      "acc": 0.66719317,
      "epoch": 1.506468797564688,
      "grad_norm": 5.75,
      "learning_rate": 1.574440225635373e-06,
      "loss": 1.56606274,
      "memory(GiB)": 117.38,
      "step": 59385,
      "train_speed(iter/s)": 1.63701
    },
    {
      "acc": 0.65939445,
      "epoch": 1.506595636732623,
      "grad_norm": 5.21875,
      "learning_rate": 1.5736764438644332e-06,
      "loss": 1.57893143,
      "memory(GiB)": 117.38,
      "step": 59390,
      "train_speed(iter/s)": 1.637025
    },
    {
      "acc": 0.66406002,
      "epoch": 1.506722475900558,
      "grad_norm": 6.8125,
      "learning_rate": 1.5729128127985004e-06,
      "loss": 1.53627806,
      "memory(GiB)": 117.38,
      "step": 59395,
      "train_speed(iter/s)": 1.637042
    },
    {
      "acc": 0.63740206,
      "epoch": 1.5068493150684932,
      "grad_norm": 5.3125,
      "learning_rate": 1.5721493324711633e-06,
      "loss": 1.63831291,
      "memory(GiB)": 117.38,
      "step": 59400,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.65534267,
      "epoch": 1.5069761542364282,
      "grad_norm": 5.6875,
      "learning_rate": 1.5713860029160028e-06,
      "loss": 1.60846481,
      "memory(GiB)": 117.38,
      "step": 59405,
      "train_speed(iter/s)": 1.637073
    },
    {
      "acc": 0.64372668,
      "epoch": 1.5071029934043634,
      "grad_norm": 5.9375,
      "learning_rate": 1.5706228241665932e-06,
      "loss": 1.64875259,
      "memory(GiB)": 117.38,
      "step": 59410,
      "train_speed(iter/s)": 1.637088
    },
    {
      "acc": 0.66544375,
      "epoch": 1.5072298325722984,
      "grad_norm": 5.09375,
      "learning_rate": 1.5698597962565032e-06,
      "loss": 1.60581703,
      "memory(GiB)": 117.38,
      "step": 59415,
      "train_speed(iter/s)": 1.637104
    },
    {
      "acc": 0.66465173,
      "epoch": 1.5073566717402334,
      "grad_norm": 6.9375,
      "learning_rate": 1.5690969192192933e-06,
      "loss": 1.56793594,
      "memory(GiB)": 117.38,
      "step": 59420,
      "train_speed(iter/s)": 1.63712
    },
    {
      "acc": 0.66263733,
      "epoch": 1.5074835109081683,
      "grad_norm": 4.625,
      "learning_rate": 1.5683341930885183e-06,
      "loss": 1.56303272,
      "memory(GiB)": 117.38,
      "step": 59425,
      "train_speed(iter/s)": 1.637135
    },
    {
      "acc": 0.64411612,
      "epoch": 1.5076103500761036,
      "grad_norm": 5.90625,
      "learning_rate": 1.567571617897729e-06,
      "loss": 1.65568466,
      "memory(GiB)": 117.38,
      "step": 59430,
      "train_speed(iter/s)": 1.637151
    },
    {
      "acc": 0.66698713,
      "epoch": 1.5077371892440385,
      "grad_norm": 4.71875,
      "learning_rate": 1.566809193680463e-06,
      "loss": 1.53366222,
      "memory(GiB)": 117.38,
      "step": 59435,
      "train_speed(iter/s)": 1.637167
    },
    {
      "acc": 0.64838991,
      "epoch": 1.5078640284119738,
      "grad_norm": 7.15625,
      "learning_rate": 1.566046920470257e-06,
      "loss": 1.67905045,
      "memory(GiB)": 117.38,
      "step": 59440,
      "train_speed(iter/s)": 1.637184
    },
    {
      "acc": 0.67054396,
      "epoch": 1.5079908675799087,
      "grad_norm": 5.03125,
      "learning_rate": 1.5652847983006376e-06,
      "loss": 1.53800392,
      "memory(GiB)": 117.38,
      "step": 59445,
      "train_speed(iter/s)": 1.637199
    },
    {
      "acc": 0.66188765,
      "epoch": 1.5081177067478437,
      "grad_norm": 4.75,
      "learning_rate": 1.564522827205131e-06,
      "loss": 1.5667593,
      "memory(GiB)": 117.38,
      "step": 59450,
      "train_speed(iter/s)": 1.637215
    },
    {
      "acc": 0.64526963,
      "epoch": 1.5082445459157787,
      "grad_norm": 7.3125,
      "learning_rate": 1.5637610072172464e-06,
      "loss": 1.60111561,
      "memory(GiB)": 117.38,
      "step": 59455,
      "train_speed(iter/s)": 1.637231
    },
    {
      "acc": 0.65367398,
      "epoch": 1.5083713850837137,
      "grad_norm": 6.0625,
      "learning_rate": 1.5629993383704933e-06,
      "loss": 1.69005527,
      "memory(GiB)": 117.38,
      "step": 59460,
      "train_speed(iter/s)": 1.637246
    },
    {
      "acc": 0.66148043,
      "epoch": 1.508498224251649,
      "grad_norm": 6.15625,
      "learning_rate": 1.5622378206983764e-06,
      "loss": 1.62536736,
      "memory(GiB)": 117.38,
      "step": 59465,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.6600419,
      "epoch": 1.5086250634195841,
      "grad_norm": 5.8125,
      "learning_rate": 1.5614764542343896e-06,
      "loss": 1.56128855,
      "memory(GiB)": 117.38,
      "step": 59470,
      "train_speed(iter/s)": 1.637277
    },
    {
      "acc": 0.65574646,
      "epoch": 1.5087519025875191,
      "grad_norm": 5.59375,
      "learning_rate": 1.5607152390120173e-06,
      "loss": 1.54552221,
      "memory(GiB)": 117.38,
      "step": 59475,
      "train_speed(iter/s)": 1.637293
    },
    {
      "acc": 0.64856749,
      "epoch": 1.5088787417554541,
      "grad_norm": 5.71875,
      "learning_rate": 1.5599541750647457e-06,
      "loss": 1.61822491,
      "memory(GiB)": 117.38,
      "step": 59480,
      "train_speed(iter/s)": 1.637308
    },
    {
      "acc": 0.65341444,
      "epoch": 1.509005580923389,
      "grad_norm": 5.21875,
      "learning_rate": 1.559193262426048e-06,
      "loss": 1.56120615,
      "memory(GiB)": 117.38,
      "step": 59485,
      "train_speed(iter/s)": 1.637324
    },
    {
      "acc": 0.65711451,
      "epoch": 1.509132420091324,
      "grad_norm": 5.40625,
      "learning_rate": 1.5584325011293943e-06,
      "loss": 1.59803238,
      "memory(GiB)": 117.38,
      "step": 59490,
      "train_speed(iter/s)": 1.637338
    },
    {
      "acc": 0.64721985,
      "epoch": 1.5092592592592593,
      "grad_norm": 5.53125,
      "learning_rate": 1.5576718912082417e-06,
      "loss": 1.63914948,
      "memory(GiB)": 117.38,
      "step": 59495,
      "train_speed(iter/s)": 1.637353
    },
    {
      "acc": 0.65776649,
      "epoch": 1.5093860984271943,
      "grad_norm": 6.25,
      "learning_rate": 1.5569114326960494e-06,
      "loss": 1.60691013,
      "memory(GiB)": 117.38,
      "step": 59500,
      "train_speed(iter/s)": 1.637368
    },
    {
      "acc": 0.66006384,
      "epoch": 1.5095129375951295,
      "grad_norm": 6.125,
      "learning_rate": 1.5561511256262651e-06,
      "loss": 1.58847275,
      "memory(GiB)": 117.38,
      "step": 59505,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.65260458,
      "epoch": 1.5096397767630645,
      "grad_norm": 5.25,
      "learning_rate": 1.55539097003233e-06,
      "loss": 1.61360741,
      "memory(GiB)": 117.38,
      "step": 59510,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.66343279,
      "epoch": 1.5097666159309995,
      "grad_norm": 5.25,
      "learning_rate": 1.5546309659476788e-06,
      "loss": 1.58285294,
      "memory(GiB)": 117.38,
      "step": 59515,
      "train_speed(iter/s)": 1.637413
    },
    {
      "acc": 0.67283044,
      "epoch": 1.5098934550989345,
      "grad_norm": 5.21875,
      "learning_rate": 1.55387111340574e-06,
      "loss": 1.58267479,
      "memory(GiB)": 117.38,
      "step": 59520,
      "train_speed(iter/s)": 1.637428
    },
    {
      "acc": 0.66473532,
      "epoch": 1.5100202942668695,
      "grad_norm": 6.09375,
      "learning_rate": 1.553111412439936e-06,
      "loss": 1.56194286,
      "memory(GiB)": 117.38,
      "step": 59525,
      "train_speed(iter/s)": 1.637442
    },
    {
      "acc": 0.65723758,
      "epoch": 1.5101471334348047,
      "grad_norm": 5.34375,
      "learning_rate": 1.5523518630836809e-06,
      "loss": 1.64816818,
      "memory(GiB)": 117.38,
      "step": 59530,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.67080235,
      "epoch": 1.5102739726027399,
      "grad_norm": 6.8125,
      "learning_rate": 1.551592465370384e-06,
      "loss": 1.50460768,
      "memory(GiB)": 117.38,
      "step": 59535,
      "train_speed(iter/s)": 1.637472
    },
    {
      "acc": 0.63776093,
      "epoch": 1.5104008117706749,
      "grad_norm": 6.0,
      "learning_rate": 1.5508332193334457e-06,
      "loss": 1.60111427,
      "memory(GiB)": 117.38,
      "step": 59540,
      "train_speed(iter/s)": 1.637487
    },
    {
      "acc": 0.64542737,
      "epoch": 1.5105276509386099,
      "grad_norm": 5.46875,
      "learning_rate": 1.5500741250062628e-06,
      "loss": 1.68193874,
      "memory(GiB)": 117.38,
      "step": 59545,
      "train_speed(iter/s)": 1.637502
    },
    {
      "acc": 0.68000145,
      "epoch": 1.5106544901065448,
      "grad_norm": 5.9375,
      "learning_rate": 1.549315182422222e-06,
      "loss": 1.53957243,
      "memory(GiB)": 117.38,
      "step": 59550,
      "train_speed(iter/s)": 1.637517
    },
    {
      "acc": 0.66647639,
      "epoch": 1.5107813292744798,
      "grad_norm": 4.90625,
      "learning_rate": 1.5485563916147062e-06,
      "loss": 1.60158615,
      "memory(GiB)": 117.38,
      "step": 59555,
      "train_speed(iter/s)": 1.637531
    },
    {
      "acc": 0.63762536,
      "epoch": 1.510908168442415,
      "grad_norm": 5.8125,
      "learning_rate": 1.5477977526170895e-06,
      "loss": 1.59889803,
      "memory(GiB)": 117.38,
      "step": 59560,
      "train_speed(iter/s)": 1.637546
    },
    {
      "acc": 0.65561848,
      "epoch": 1.51103500761035,
      "grad_norm": 5.875,
      "learning_rate": 1.5470392654627392e-06,
      "loss": 1.62687626,
      "memory(GiB)": 117.38,
      "step": 59565,
      "train_speed(iter/s)": 1.637561
    },
    {
      "acc": 0.65773377,
      "epoch": 1.5111618467782852,
      "grad_norm": 5.375,
      "learning_rate": 1.5462809301850212e-06,
      "loss": 1.60167675,
      "memory(GiB)": 117.38,
      "step": 59570,
      "train_speed(iter/s)": 1.637576
    },
    {
      "acc": 0.65607738,
      "epoch": 1.5112886859462202,
      "grad_norm": 5.25,
      "learning_rate": 1.5455227468172862e-06,
      "loss": 1.64228477,
      "memory(GiB)": 117.38,
      "step": 59575,
      "train_speed(iter/s)": 1.637591
    },
    {
      "acc": 0.6501904,
      "epoch": 1.5114155251141552,
      "grad_norm": 7.125,
      "learning_rate": 1.5447647153928842e-06,
      "loss": 1.65314693,
      "memory(GiB)": 117.38,
      "step": 59580,
      "train_speed(iter/s)": 1.637606
    },
    {
      "acc": 0.65768318,
      "epoch": 1.5115423642820902,
      "grad_norm": 6.4375,
      "learning_rate": 1.5440068359451548e-06,
      "loss": 1.61936302,
      "memory(GiB)": 117.38,
      "step": 59585,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.66429658,
      "epoch": 1.5116692034500254,
      "grad_norm": 9.625,
      "learning_rate": 1.5432491085074381e-06,
      "loss": 1.59712486,
      "memory(GiB)": 117.38,
      "step": 59590,
      "train_speed(iter/s)": 1.637637
    },
    {
      "acc": 0.64664769,
      "epoch": 1.5117960426179604,
      "grad_norm": 4.875,
      "learning_rate": 1.5424915331130568e-06,
      "loss": 1.64883251,
      "memory(GiB)": 117.38,
      "step": 59595,
      "train_speed(iter/s)": 1.637652
    },
    {
      "acc": 0.66840744,
      "epoch": 1.5119228817858956,
      "grad_norm": 8.125,
      "learning_rate": 1.5417341097953332e-06,
      "loss": 1.5181076,
      "memory(GiB)": 117.38,
      "step": 59600,
      "train_speed(iter/s)": 1.637666
    },
    {
      "acc": 0.67108917,
      "epoch": 1.5120497209538306,
      "grad_norm": 7.1875,
      "learning_rate": 1.540976838587585e-06,
      "loss": 1.59098244,
      "memory(GiB)": 117.38,
      "step": 59605,
      "train_speed(iter/s)": 1.637682
    },
    {
      "acc": 0.65572405,
      "epoch": 1.5121765601217656,
      "grad_norm": 5.28125,
      "learning_rate": 1.5402197195231205e-06,
      "loss": 1.60146599,
      "memory(GiB)": 117.38,
      "step": 59610,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.66543159,
      "epoch": 1.5123033992897006,
      "grad_norm": 6.59375,
      "learning_rate": 1.5394627526352379e-06,
      "loss": 1.57943907,
      "memory(GiB)": 117.38,
      "step": 59615,
      "train_speed(iter/s)": 1.637712
    },
    {
      "acc": 0.65606475,
      "epoch": 1.5124302384576356,
      "grad_norm": 5.75,
      "learning_rate": 1.5387059379572322e-06,
      "loss": 1.60311127,
      "memory(GiB)": 117.38,
      "step": 59620,
      "train_speed(iter/s)": 1.637728
    },
    {
      "acc": 0.68065972,
      "epoch": 1.5125570776255708,
      "grad_norm": 6.0,
      "learning_rate": 1.537949275522394e-06,
      "loss": 1.54807396,
      "memory(GiB)": 117.38,
      "step": 59625,
      "train_speed(iter/s)": 1.637743
    },
    {
      "acc": 0.64313092,
      "epoch": 1.512683916793506,
      "grad_norm": 10.1875,
      "learning_rate": 1.5371927653640056e-06,
      "loss": 1.66140862,
      "memory(GiB)": 117.38,
      "step": 59630,
      "train_speed(iter/s)": 1.637759
    },
    {
      "acc": 0.65956144,
      "epoch": 1.512810755961441,
      "grad_norm": 5.1875,
      "learning_rate": 1.5364364075153366e-06,
      "loss": 1.57860355,
      "memory(GiB)": 117.38,
      "step": 59635,
      "train_speed(iter/s)": 1.637774
    },
    {
      "acc": 0.66728215,
      "epoch": 1.512937595129376,
      "grad_norm": 5.875,
      "learning_rate": 1.5356802020096595e-06,
      "loss": 1.5678194,
      "memory(GiB)": 117.38,
      "step": 59640,
      "train_speed(iter/s)": 1.637789
    },
    {
      "acc": 0.66470032,
      "epoch": 1.513064434297311,
      "grad_norm": 5.5625,
      "learning_rate": 1.5349241488802346e-06,
      "loss": 1.57913189,
      "memory(GiB)": 117.38,
      "step": 59645,
      "train_speed(iter/s)": 1.637804
    },
    {
      "acc": 0.65819478,
      "epoch": 1.513191273465246,
      "grad_norm": 5.90625,
      "learning_rate": 1.5341682481603155e-06,
      "loss": 1.63519821,
      "memory(GiB)": 117.38,
      "step": 59650,
      "train_speed(iter/s)": 1.637819
    },
    {
      "acc": 0.66518583,
      "epoch": 1.5133181126331812,
      "grad_norm": 6.125,
      "learning_rate": 1.5334124998831512e-06,
      "loss": 1.56683216,
      "memory(GiB)": 117.38,
      "step": 59655,
      "train_speed(iter/s)": 1.637833
    },
    {
      "acc": 0.63133111,
      "epoch": 1.5134449518011162,
      "grad_norm": 5.03125,
      "learning_rate": 1.532656904081982e-06,
      "loss": 1.64764099,
      "memory(GiB)": 117.38,
      "step": 59660,
      "train_speed(iter/s)": 1.637849
    },
    {
      "acc": 0.66431923,
      "epoch": 1.5135717909690514,
      "grad_norm": 6.0625,
      "learning_rate": 1.5319014607900428e-06,
      "loss": 1.56693516,
      "memory(GiB)": 117.38,
      "step": 59665,
      "train_speed(iter/s)": 1.637863
    },
    {
      "acc": 0.66129799,
      "epoch": 1.5136986301369864,
      "grad_norm": 5.65625,
      "learning_rate": 1.5311461700405617e-06,
      "loss": 1.60601902,
      "memory(GiB)": 117.38,
      "step": 59670,
      "train_speed(iter/s)": 1.637877
    },
    {
      "acc": 0.65645337,
      "epoch": 1.5138254693049213,
      "grad_norm": 6.71875,
      "learning_rate": 1.5303910318667586e-06,
      "loss": 1.6351366,
      "memory(GiB)": 117.38,
      "step": 59675,
      "train_speed(iter/s)": 1.637893
    },
    {
      "acc": 0.6603056,
      "epoch": 1.5139523084728563,
      "grad_norm": 5.625,
      "learning_rate": 1.529636046301849e-06,
      "loss": 1.61880131,
      "memory(GiB)": 117.38,
      "step": 59680,
      "train_speed(iter/s)": 1.637908
    },
    {
      "acc": 0.63773842,
      "epoch": 1.5140791476407913,
      "grad_norm": 6.3125,
      "learning_rate": 1.5288812133790405e-06,
      "loss": 1.68753853,
      "memory(GiB)": 117.38,
      "step": 59685,
      "train_speed(iter/s)": 1.637924
    },
    {
      "acc": 0.64779053,
      "epoch": 1.5142059868087265,
      "grad_norm": 5.21875,
      "learning_rate": 1.5281265331315332e-06,
      "loss": 1.6136404,
      "memory(GiB)": 117.38,
      "step": 59690,
      "train_speed(iter/s)": 1.637938
    },
    {
      "acc": 0.65977907,
      "epoch": 1.5143328259766617,
      "grad_norm": 5.8125,
      "learning_rate": 1.5273720055925217e-06,
      "loss": 1.58422794,
      "memory(GiB)": 117.38,
      "step": 59695,
      "train_speed(iter/s)": 1.637954
    },
    {
      "acc": 0.65920839,
      "epoch": 1.5144596651445967,
      "grad_norm": 5.21875,
      "learning_rate": 1.5266176307951936e-06,
      "loss": 1.61589317,
      "memory(GiB)": 117.38,
      "step": 59700,
      "train_speed(iter/s)": 1.637968
    },
    {
      "acc": 0.62904959,
      "epoch": 1.5145865043125317,
      "grad_norm": 5.28125,
      "learning_rate": 1.5258634087727298e-06,
      "loss": 1.59031744,
      "memory(GiB)": 117.38,
      "step": 59705,
      "train_speed(iter/s)": 1.637983
    },
    {
      "acc": 0.65859184,
      "epoch": 1.5147133434804667,
      "grad_norm": 6.1875,
      "learning_rate": 1.5251093395583045e-06,
      "loss": 1.60083256,
      "memory(GiB)": 117.38,
      "step": 59710,
      "train_speed(iter/s)": 1.637998
    },
    {
      "acc": 0.66071482,
      "epoch": 1.5148401826484017,
      "grad_norm": 5.28125,
      "learning_rate": 1.5243554231850843e-06,
      "loss": 1.57996674,
      "memory(GiB)": 117.38,
      "step": 59715,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.65562344,
      "epoch": 1.514967021816337,
      "grad_norm": 5.9375,
      "learning_rate": 1.5236016596862302e-06,
      "loss": 1.57238712,
      "memory(GiB)": 117.38,
      "step": 59720,
      "train_speed(iter/s)": 1.638027
    },
    {
      "acc": 0.65562716,
      "epoch": 1.515093860984272,
      "grad_norm": 4.90625,
      "learning_rate": 1.5228480490948943e-06,
      "loss": 1.5669693,
      "memory(GiB)": 117.38,
      "step": 59725,
      "train_speed(iter/s)": 1.638042
    },
    {
      "acc": 0.66293459,
      "epoch": 1.5152207001522071,
      "grad_norm": 5.78125,
      "learning_rate": 1.5220945914442292e-06,
      "loss": 1.61664486,
      "memory(GiB)": 117.38,
      "step": 59730,
      "train_speed(iter/s)": 1.638057
    },
    {
      "acc": 0.6535965,
      "epoch": 1.515347539320142,
      "grad_norm": 5.3125,
      "learning_rate": 1.52134128676737e-06,
      "loss": 1.56146193,
      "memory(GiB)": 117.38,
      "step": 59735,
      "train_speed(iter/s)": 1.638072
    },
    {
      "acc": 0.65644436,
      "epoch": 1.515474378488077,
      "grad_norm": 6.4375,
      "learning_rate": 1.5205881350974504e-06,
      "loss": 1.6927681,
      "memory(GiB)": 117.38,
      "step": 59740,
      "train_speed(iter/s)": 1.638086
    },
    {
      "acc": 0.64107752,
      "epoch": 1.515601217656012,
      "grad_norm": 7.25,
      "learning_rate": 1.5198351364676012e-06,
      "loss": 1.61112938,
      "memory(GiB)": 117.38,
      "step": 59745,
      "train_speed(iter/s)": 1.638101
    },
    {
      "acc": 0.6616334,
      "epoch": 1.5157280568239473,
      "grad_norm": 6.4375,
      "learning_rate": 1.5190822909109415e-06,
      "loss": 1.61259651,
      "memory(GiB)": 117.38,
      "step": 59750,
      "train_speed(iter/s)": 1.638115
    },
    {
      "acc": 0.65482903,
      "epoch": 1.5158548959918823,
      "grad_norm": 5.0625,
      "learning_rate": 1.5183295984605824e-06,
      "loss": 1.63801155,
      "memory(GiB)": 117.38,
      "step": 59755,
      "train_speed(iter/s)": 1.638129
    },
    {
      "acc": 0.66039543,
      "epoch": 1.5159817351598175,
      "grad_norm": 5.21875,
      "learning_rate": 1.5175770591496303e-06,
      "loss": 1.58310776,
      "memory(GiB)": 117.38,
      "step": 59760,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.65295467,
      "epoch": 1.5161085743277525,
      "grad_norm": 5.25,
      "learning_rate": 1.5168246730111892e-06,
      "loss": 1.61428986,
      "memory(GiB)": 117.38,
      "step": 59765,
      "train_speed(iter/s)": 1.638158
    },
    {
      "acc": 0.68336835,
      "epoch": 1.5162354134956875,
      "grad_norm": 5.5625,
      "learning_rate": 1.5160724400783511e-06,
      "loss": 1.59299622,
      "memory(GiB)": 117.38,
      "step": 59770,
      "train_speed(iter/s)": 1.638172
    },
    {
      "acc": 0.65406723,
      "epoch": 1.5163622526636225,
      "grad_norm": 5.1875,
      "learning_rate": 1.5153203603841992e-06,
      "loss": 1.59817829,
      "memory(GiB)": 117.38,
      "step": 59775,
      "train_speed(iter/s)": 1.638186
    },
    {
      "acc": 0.65257907,
      "epoch": 1.5164890918315574,
      "grad_norm": 5.84375,
      "learning_rate": 1.5145684339618172e-06,
      "loss": 1.66878014,
      "memory(GiB)": 117.38,
      "step": 59780,
      "train_speed(iter/s)": 1.6382
    },
    {
      "acc": 0.67384071,
      "epoch": 1.5166159309994927,
      "grad_norm": 7.625,
      "learning_rate": 1.5138166608442768e-06,
      "loss": 1.52178497,
      "memory(GiB)": 117.38,
      "step": 59785,
      "train_speed(iter/s)": 1.638215
    },
    {
      "acc": 0.65550747,
      "epoch": 1.5167427701674279,
      "grad_norm": 6.25,
      "learning_rate": 1.5130650410646452e-06,
      "loss": 1.57656002,
      "memory(GiB)": 117.38,
      "step": 59790,
      "train_speed(iter/s)": 1.638229
    },
    {
      "acc": 0.66310744,
      "epoch": 1.5168696093353629,
      "grad_norm": 4.53125,
      "learning_rate": 1.5123135746559792e-06,
      "loss": 1.59146099,
      "memory(GiB)": 117.38,
      "step": 59795,
      "train_speed(iter/s)": 1.638243
    },
    {
      "acc": 0.67225218,
      "epoch": 1.5169964485032978,
      "grad_norm": 6.46875,
      "learning_rate": 1.5115622616513343e-06,
      "loss": 1.52971506,
      "memory(GiB)": 117.38,
      "step": 59800,
      "train_speed(iter/s)": 1.638257
    },
    {
      "acc": 0.64547067,
      "epoch": 1.5171232876712328,
      "grad_norm": 5.28125,
      "learning_rate": 1.5108111020837564e-06,
      "loss": 1.71772346,
      "memory(GiB)": 117.38,
      "step": 59805,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.64823647,
      "epoch": 1.5172501268391678,
      "grad_norm": 5.15625,
      "learning_rate": 1.5100600959862838e-06,
      "loss": 1.58661766,
      "memory(GiB)": 117.38,
      "step": 59810,
      "train_speed(iter/s)": 1.638285
    },
    {
      "acc": 0.65941296,
      "epoch": 1.517376966007103,
      "grad_norm": 5.6875,
      "learning_rate": 1.5093092433919497e-06,
      "loss": 1.59504967,
      "memory(GiB)": 117.38,
      "step": 59815,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.65098925,
      "epoch": 1.517503805175038,
      "grad_norm": 5.8125,
      "learning_rate": 1.5085585443337803e-06,
      "loss": 1.5551198,
      "memory(GiB)": 117.38,
      "step": 59820,
      "train_speed(iter/s)": 1.638314
    },
    {
      "acc": 0.65557141,
      "epoch": 1.5176306443429732,
      "grad_norm": 6.5,
      "learning_rate": 1.507807998844794e-06,
      "loss": 1.68314095,
      "memory(GiB)": 117.38,
      "step": 59825,
      "train_speed(iter/s)": 1.638329
    },
    {
      "acc": 0.65469613,
      "epoch": 1.5177574835109082,
      "grad_norm": 5.875,
      "learning_rate": 1.5070576069580039e-06,
      "loss": 1.57277184,
      "memory(GiB)": 117.38,
      "step": 59830,
      "train_speed(iter/s)": 1.638343
    },
    {
      "acc": 0.65734887,
      "epoch": 1.5178843226788432,
      "grad_norm": 4.75,
      "learning_rate": 1.5063073687064144e-06,
      "loss": 1.63628693,
      "memory(GiB)": 117.38,
      "step": 59835,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.6680016,
      "epoch": 1.5180111618467782,
      "grad_norm": 5.71875,
      "learning_rate": 1.5055572841230253e-06,
      "loss": 1.57862234,
      "memory(GiB)": 117.38,
      "step": 59840,
      "train_speed(iter/s)": 1.638371
    },
    {
      "acc": 0.66577682,
      "epoch": 1.5181380010147132,
      "grad_norm": 6.1875,
      "learning_rate": 1.5048073532408287e-06,
      "loss": 1.5896265,
      "memory(GiB)": 117.38,
      "step": 59845,
      "train_speed(iter/s)": 1.638385
    },
    {
      "acc": 0.66489553,
      "epoch": 1.5182648401826484,
      "grad_norm": 4.84375,
      "learning_rate": 1.5040575760928094e-06,
      "loss": 1.58216515,
      "memory(GiB)": 117.38,
      "step": 59850,
      "train_speed(iter/s)": 1.638399
    },
    {
      "acc": 0.64914498,
      "epoch": 1.5183916793505836,
      "grad_norm": 6.25,
      "learning_rate": 1.5033079527119466e-06,
      "loss": 1.66169624,
      "memory(GiB)": 117.38,
      "step": 59855,
      "train_speed(iter/s)": 1.638413
    },
    {
      "acc": 0.66525979,
      "epoch": 1.5185185185185186,
      "grad_norm": 5.875,
      "learning_rate": 1.5025584831312112e-06,
      "loss": 1.58025627,
      "memory(GiB)": 117.38,
      "step": 59860,
      "train_speed(iter/s)": 1.638427
    },
    {
      "acc": 0.650809,
      "epoch": 1.5186453576864536,
      "grad_norm": 6.6875,
      "learning_rate": 1.5018091673835667e-06,
      "loss": 1.62069435,
      "memory(GiB)": 117.38,
      "step": 59865,
      "train_speed(iter/s)": 1.638442
    },
    {
      "acc": 0.66731596,
      "epoch": 1.5187721968543886,
      "grad_norm": 5.6875,
      "learning_rate": 1.501060005501977e-06,
      "loss": 1.57781525,
      "memory(GiB)": 117.38,
      "step": 59870,
      "train_speed(iter/s)": 1.638456
    },
    {
      "acc": 0.66133699,
      "epoch": 1.5188990360223236,
      "grad_norm": 5.96875,
      "learning_rate": 1.500310997519388e-06,
      "loss": 1.6319046,
      "memory(GiB)": 117.38,
      "step": 59875,
      "train_speed(iter/s)": 1.63847
    },
    {
      "acc": 0.66813779,
      "epoch": 1.5190258751902588,
      "grad_norm": 5.375,
      "learning_rate": 1.4995621434687468e-06,
      "loss": 1.58863544,
      "memory(GiB)": 117.38,
      "step": 59880,
      "train_speed(iter/s)": 1.638483
    },
    {
      "acc": 0.64411836,
      "epoch": 1.5191527143581938,
      "grad_norm": 5.65625,
      "learning_rate": 1.4988134433829892e-06,
      "loss": 1.66836929,
      "memory(GiB)": 117.38,
      "step": 59885,
      "train_speed(iter/s)": 1.638496
    },
    {
      "acc": 0.64596262,
      "epoch": 1.519279553526129,
      "grad_norm": 6.21875,
      "learning_rate": 1.4980648972950507e-06,
      "loss": 1.61121902,
      "memory(GiB)": 117.38,
      "step": 59890,
      "train_speed(iter/s)": 1.63829
    },
    {
      "acc": 0.65286098,
      "epoch": 1.519406392694064,
      "grad_norm": 6.1875,
      "learning_rate": 1.4973165052378518e-06,
      "loss": 1.5737771,
      "memory(GiB)": 117.38,
      "step": 59895,
      "train_speed(iter/s)": 1.638303
    },
    {
      "acc": 0.649895,
      "epoch": 1.519533231861999,
      "grad_norm": 5.9375,
      "learning_rate": 1.49656826724431e-06,
      "loss": 1.61807003,
      "memory(GiB)": 117.38,
      "step": 59900,
      "train_speed(iter/s)": 1.638081
    },
    {
      "acc": 0.66251173,
      "epoch": 1.519660071029934,
      "grad_norm": 5.5625,
      "learning_rate": 1.4958201833473386e-06,
      "loss": 1.55404568,
      "memory(GiB)": 117.38,
      "step": 59905,
      "train_speed(iter/s)": 1.638095
    },
    {
      "acc": 0.63647819,
      "epoch": 1.5197869101978692,
      "grad_norm": 5.90625,
      "learning_rate": 1.4950722535798423e-06,
      "loss": 1.65166512,
      "memory(GiB)": 117.38,
      "step": 59910,
      "train_speed(iter/s)": 1.638109
    },
    {
      "acc": 0.65204973,
      "epoch": 1.5199137493658041,
      "grad_norm": 5.5,
      "learning_rate": 1.4943244779747134e-06,
      "loss": 1.61760178,
      "memory(GiB)": 117.38,
      "step": 59915,
      "train_speed(iter/s)": 1.638123
    },
    {
      "acc": 0.64628162,
      "epoch": 1.5200405885337394,
      "grad_norm": 5.03125,
      "learning_rate": 1.4935768565648478e-06,
      "loss": 1.5935832,
      "memory(GiB)": 117.38,
      "step": 59920,
      "train_speed(iter/s)": 1.638137
    },
    {
      "acc": 0.65135431,
      "epoch": 1.5201674277016743,
      "grad_norm": 7.625,
      "learning_rate": 1.4928293893831265e-06,
      "loss": 1.66151772,
      "memory(GiB)": 117.38,
      "step": 59925,
      "train_speed(iter/s)": 1.638152
    },
    {
      "acc": 0.65062904,
      "epoch": 1.5202942668696093,
      "grad_norm": 6.25,
      "learning_rate": 1.4920820764624288e-06,
      "loss": 1.6067997,
      "memory(GiB)": 117.38,
      "step": 59930,
      "train_speed(iter/s)": 1.638166
    },
    {
      "acc": 0.65350237,
      "epoch": 1.5204211060375443,
      "grad_norm": 5.59375,
      "learning_rate": 1.4913349178356202e-06,
      "loss": 1.63344421,
      "memory(GiB)": 117.38,
      "step": 59935,
      "train_speed(iter/s)": 1.63818
    },
    {
      "acc": 0.64665518,
      "epoch": 1.5205479452054793,
      "grad_norm": 7.0,
      "learning_rate": 1.4905879135355684e-06,
      "loss": 1.66476669,
      "memory(GiB)": 117.38,
      "step": 59940,
      "train_speed(iter/s)": 1.638195
    },
    {
      "acc": 0.64801178,
      "epoch": 1.5206747843734145,
      "grad_norm": 6.625,
      "learning_rate": 1.4898410635951282e-06,
      "loss": 1.57352085,
      "memory(GiB)": 117.38,
      "step": 59945,
      "train_speed(iter/s)": 1.638209
    },
    {
      "acc": 0.65974984,
      "epoch": 1.5208016235413497,
      "grad_norm": 5.9375,
      "learning_rate": 1.4890943680471503e-06,
      "loss": 1.57228327,
      "memory(GiB)": 117.38,
      "step": 59950,
      "train_speed(iter/s)": 1.638223
    },
    {
      "acc": 0.65208964,
      "epoch": 1.5209284627092847,
      "grad_norm": 5.53125,
      "learning_rate": 1.4883478269244766e-06,
      "loss": 1.58871241,
      "memory(GiB)": 117.38,
      "step": 59955,
      "train_speed(iter/s)": 1.638238
    },
    {
      "acc": 0.65905828,
      "epoch": 1.5210553018772197,
      "grad_norm": 5.40625,
      "learning_rate": 1.4876014402599443e-06,
      "loss": 1.65592327,
      "memory(GiB)": 117.38,
      "step": 59960,
      "train_speed(iter/s)": 1.638253
    },
    {
      "acc": 0.64793797,
      "epoch": 1.5211821410451547,
      "grad_norm": 5.65625,
      "learning_rate": 1.4868552080863824e-06,
      "loss": 1.66602058,
      "memory(GiB)": 117.38,
      "step": 59965,
      "train_speed(iter/s)": 1.638267
    },
    {
      "acc": 0.67543235,
      "epoch": 1.5213089802130897,
      "grad_norm": 7.125,
      "learning_rate": 1.4861091304366139e-06,
      "loss": 1.5205225,
      "memory(GiB)": 117.38,
      "step": 59970,
      "train_speed(iter/s)": 1.638281
    },
    {
      "acc": 0.66942892,
      "epoch": 1.521435819381025,
      "grad_norm": 5.96875,
      "learning_rate": 1.4853632073434533e-06,
      "loss": 1.50150871,
      "memory(GiB)": 117.38,
      "step": 59975,
      "train_speed(iter/s)": 1.638295
    },
    {
      "acc": 0.64819031,
      "epoch": 1.52156265854896,
      "grad_norm": 10.0,
      "learning_rate": 1.484617438839711e-06,
      "loss": 1.66951733,
      "memory(GiB)": 117.38,
      "step": 59980,
      "train_speed(iter/s)": 1.63831
    },
    {
      "acc": 0.66218333,
      "epoch": 1.521689497716895,
      "grad_norm": 8.0,
      "learning_rate": 1.483871824958189e-06,
      "loss": 1.55853748,
      "memory(GiB)": 117.38,
      "step": 59985,
      "train_speed(iter/s)": 1.638324
    },
    {
      "acc": 0.64892235,
      "epoch": 1.52181633688483,
      "grad_norm": 6.3125,
      "learning_rate": 1.483126365731682e-06,
      "loss": 1.59185219,
      "memory(GiB)": 117.38,
      "step": 59990,
      "train_speed(iter/s)": 1.638339
    },
    {
      "acc": 0.66940746,
      "epoch": 1.521943176052765,
      "grad_norm": 5.375,
      "learning_rate": 1.4823810611929795e-06,
      "loss": 1.52065325,
      "memory(GiB)": 117.38,
      "step": 59995,
      "train_speed(iter/s)": 1.638354
    },
    {
      "acc": 0.67647648,
      "epoch": 1.5220700152207,
      "grad_norm": 5.34375,
      "learning_rate": 1.481635911374863e-06,
      "loss": 1.50588226,
      "memory(GiB)": 117.38,
      "step": 60000,
      "train_speed(iter/s)": 1.638368
    },
    {
      "epoch": 1.5220700152207,
      "eval_acc": 0.6462238204725002,
      "eval_loss": 1.5732791423797607,
      "eval_runtime": 58.3904,
      "eval_samples_per_second": 109.093,
      "eval_steps_per_second": 27.282,
      "step": 60000
    },
    {
      "acc": 0.65408506,
      "epoch": 1.522196854388635,
      "grad_norm": 7.3125,
      "learning_rate": 1.480890916310106e-06,
      "loss": 1.63356438,
      "memory(GiB)": 117.38,
      "step": 60005,
      "train_speed(iter/s)": 1.635583
    },
    {
      "acc": 0.67194986,
      "epoch": 1.5223236935565703,
      "grad_norm": 6.25,
      "learning_rate": 1.4801460760314811e-06,
      "loss": 1.54215813,
      "memory(GiB)": 117.38,
      "step": 60010,
      "train_speed(iter/s)": 1.635595
    },
    {
      "acc": 0.64792438,
      "epoch": 1.5224505327245055,
      "grad_norm": 7.03125,
      "learning_rate": 1.4794013905717453e-06,
      "loss": 1.64431,
      "memory(GiB)": 117.38,
      "step": 60015,
      "train_speed(iter/s)": 1.635609
    },
    {
      "acc": 0.65688086,
      "epoch": 1.5225773718924405,
      "grad_norm": 5.25,
      "learning_rate": 1.4786568599636548e-06,
      "loss": 1.59861679,
      "memory(GiB)": 117.38,
      "step": 60020,
      "train_speed(iter/s)": 1.635623
    },
    {
      "acc": 0.64923539,
      "epoch": 1.5227042110603755,
      "grad_norm": 5.21875,
      "learning_rate": 1.4779124842399556e-06,
      "loss": 1.66376095,
      "memory(GiB)": 117.38,
      "step": 60025,
      "train_speed(iter/s)": 1.635637
    },
    {
      "acc": 0.65651307,
      "epoch": 1.5228310502283104,
      "grad_norm": 5.9375,
      "learning_rate": 1.4771682634333933e-06,
      "loss": 1.54960289,
      "memory(GiB)": 117.38,
      "step": 60030,
      "train_speed(iter/s)": 1.635651
    },
    {
      "acc": 0.66246595,
      "epoch": 1.5229578893962454,
      "grad_norm": 6.9375,
      "learning_rate": 1.4764241975766975e-06,
      "loss": 1.54694185,
      "memory(GiB)": 117.38,
      "step": 60035,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.65754814,
      "epoch": 1.5230847285641806,
      "grad_norm": 6.0625,
      "learning_rate": 1.475680286702596e-06,
      "loss": 1.57554607,
      "memory(GiB)": 117.38,
      "step": 60040,
      "train_speed(iter/s)": 1.635679
    },
    {
      "acc": 0.65941486,
      "epoch": 1.5232115677321156,
      "grad_norm": 5.28125,
      "learning_rate": 1.474936530843812e-06,
      "loss": 1.53971472,
      "memory(GiB)": 117.38,
      "step": 60045,
      "train_speed(iter/s)": 1.635691
    },
    {
      "acc": 0.6630558,
      "epoch": 1.5233384069000508,
      "grad_norm": 5.8125,
      "learning_rate": 1.4741929300330588e-06,
      "loss": 1.57260914,
      "memory(GiB)": 117.38,
      "step": 60050,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.65574389,
      "epoch": 1.5234652460679858,
      "grad_norm": 5.71875,
      "learning_rate": 1.4734494843030405e-06,
      "loss": 1.60111217,
      "memory(GiB)": 117.38,
      "step": 60055,
      "train_speed(iter/s)": 1.63572
    },
    {
      "acc": 0.65624018,
      "epoch": 1.5235920852359208,
      "grad_norm": 5.65625,
      "learning_rate": 1.4727061936864573e-06,
      "loss": 1.61748276,
      "memory(GiB)": 117.38,
      "step": 60060,
      "train_speed(iter/s)": 1.635733
    },
    {
      "acc": 0.64636483,
      "epoch": 1.5237189244038558,
      "grad_norm": 5.71875,
      "learning_rate": 1.4719630582160056e-06,
      "loss": 1.66449432,
      "memory(GiB)": 117.38,
      "step": 60065,
      "train_speed(iter/s)": 1.635747
    },
    {
      "acc": 0.65765944,
      "epoch": 1.523845763571791,
      "grad_norm": 5.5,
      "learning_rate": 1.4712200779243718e-06,
      "loss": 1.6092598,
      "memory(GiB)": 117.38,
      "step": 60070,
      "train_speed(iter/s)": 1.635762
    },
    {
      "acc": 0.65735688,
      "epoch": 1.523972602739726,
      "grad_norm": 5.8125,
      "learning_rate": 1.4704772528442308e-06,
      "loss": 1.63104286,
      "memory(GiB)": 117.38,
      "step": 60075,
      "train_speed(iter/s)": 1.635776
    },
    {
      "acc": 0.66829715,
      "epoch": 1.5240994419076612,
      "grad_norm": 5.03125,
      "learning_rate": 1.46973458300826e-06,
      "loss": 1.50231695,
      "memory(GiB)": 117.38,
      "step": 60080,
      "train_speed(iter/s)": 1.63579
    },
    {
      "acc": 0.65415154,
      "epoch": 1.5242262810755962,
      "grad_norm": 6.375,
      "learning_rate": 1.4689920684491232e-06,
      "loss": 1.59666777,
      "memory(GiB)": 117.38,
      "step": 60085,
      "train_speed(iter/s)": 1.635804
    },
    {
      "acc": 0.66662707,
      "epoch": 1.5243531202435312,
      "grad_norm": 5.78125,
      "learning_rate": 1.4682497091994807e-06,
      "loss": 1.59601307,
      "memory(GiB)": 117.38,
      "step": 60090,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.66017933,
      "epoch": 1.5244799594114662,
      "grad_norm": 5.40625,
      "learning_rate": 1.467507505291984e-06,
      "loss": 1.58230801,
      "memory(GiB)": 117.38,
      "step": 60095,
      "train_speed(iter/s)": 1.635832
    },
    {
      "acc": 0.64228773,
      "epoch": 1.5246067985794012,
      "grad_norm": 5.84375,
      "learning_rate": 1.4667654567592781e-06,
      "loss": 1.68955727,
      "memory(GiB)": 117.38,
      "step": 60100,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.64893932,
      "epoch": 1.5247336377473364,
      "grad_norm": 5.9375,
      "learning_rate": 1.4660235636340025e-06,
      "loss": 1.62020645,
      "memory(GiB)": 117.38,
      "step": 60105,
      "train_speed(iter/s)": 1.635859
    },
    {
      "acc": 0.65086508,
      "epoch": 1.5248604769152716,
      "grad_norm": 6.5,
      "learning_rate": 1.465281825948789e-06,
      "loss": 1.62364807,
      "memory(GiB)": 117.38,
      "step": 60110,
      "train_speed(iter/s)": 1.635873
    },
    {
      "acc": 0.64010787,
      "epoch": 1.5249873160832066,
      "grad_norm": 5.625,
      "learning_rate": 1.464540243736262e-06,
      "loss": 1.63286953,
      "memory(GiB)": 117.38,
      "step": 60115,
      "train_speed(iter/s)": 1.635887
    },
    {
      "acc": 0.65779972,
      "epoch": 1.5251141552511416,
      "grad_norm": 5.875,
      "learning_rate": 1.4637988170290396e-06,
      "loss": 1.5491478,
      "memory(GiB)": 117.38,
      "step": 60120,
      "train_speed(iter/s)": 1.635902
    },
    {
      "acc": 0.66906114,
      "epoch": 1.5252409944190766,
      "grad_norm": 5.53125,
      "learning_rate": 1.4630575458597334e-06,
      "loss": 1.56732073,
      "memory(GiB)": 117.38,
      "step": 60125,
      "train_speed(iter/s)": 1.635916
    },
    {
      "acc": 0.67282534,
      "epoch": 1.5253678335870116,
      "grad_norm": 6.75,
      "learning_rate": 1.4623164302609472e-06,
      "loss": 1.51577368,
      "memory(GiB)": 117.38,
      "step": 60130,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.64581718,
      "epoch": 1.5254946727549468,
      "grad_norm": 5.9375,
      "learning_rate": 1.4615754702652796e-06,
      "loss": 1.68193436,
      "memory(GiB)": 117.38,
      "step": 60135,
      "train_speed(iter/s)": 1.635943
    },
    {
      "acc": 0.66856804,
      "epoch": 1.5256215119228818,
      "grad_norm": 5.375,
      "learning_rate": 1.4608346659053208e-06,
      "loss": 1.53394117,
      "memory(GiB)": 117.38,
      "step": 60140,
      "train_speed(iter/s)": 1.635958
    },
    {
      "acc": 0.65752859,
      "epoch": 1.525748351090817,
      "grad_norm": 6.03125,
      "learning_rate": 1.4600940172136541e-06,
      "loss": 1.58509798,
      "memory(GiB)": 117.38,
      "step": 60145,
      "train_speed(iter/s)": 1.635972
    },
    {
      "acc": 0.66464953,
      "epoch": 1.525875190258752,
      "grad_norm": 6.46875,
      "learning_rate": 1.4593535242228575e-06,
      "loss": 1.55644188,
      "memory(GiB)": 117.38,
      "step": 60150,
      "train_speed(iter/s)": 1.635985
    },
    {
      "acc": 0.65431881,
      "epoch": 1.526002029426687,
      "grad_norm": 6.9375,
      "learning_rate": 1.4586131869655001e-06,
      "loss": 1.59162655,
      "memory(GiB)": 117.38,
      "step": 60155,
      "train_speed(iter/s)": 1.636
    },
    {
      "acc": 0.64266405,
      "epoch": 1.526128868594622,
      "grad_norm": 6.21875,
      "learning_rate": 1.4578730054741462e-06,
      "loss": 1.65194492,
      "memory(GiB)": 117.38,
      "step": 60160,
      "train_speed(iter/s)": 1.636013
    },
    {
      "acc": 0.65512137,
      "epoch": 1.526255707762557,
      "grad_norm": 5.5625,
      "learning_rate": 1.4571329797813511e-06,
      "loss": 1.56337051,
      "memory(GiB)": 117.38,
      "step": 60165,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.66711531,
      "epoch": 1.5263825469304921,
      "grad_norm": 6.4375,
      "learning_rate": 1.4563931099196678e-06,
      "loss": 1.54521618,
      "memory(GiB)": 117.38,
      "step": 60170,
      "train_speed(iter/s)": 1.636043
    },
    {
      "acc": 0.67029066,
      "epoch": 1.5265093860984273,
      "grad_norm": 6.0625,
      "learning_rate": 1.455653395921635e-06,
      "loss": 1.5556139,
      "memory(GiB)": 117.38,
      "step": 60175,
      "train_speed(iter/s)": 1.636057
    },
    {
      "acc": 0.66554408,
      "epoch": 1.5266362252663623,
      "grad_norm": 7.8125,
      "learning_rate": 1.4549138378197891e-06,
      "loss": 1.61882439,
      "memory(GiB)": 117.38,
      "step": 60180,
      "train_speed(iter/s)": 1.636071
    },
    {
      "acc": 0.65113807,
      "epoch": 1.5267630644342973,
      "grad_norm": 5.4375,
      "learning_rate": 1.4541744356466615e-06,
      "loss": 1.63148766,
      "memory(GiB)": 117.38,
      "step": 60185,
      "train_speed(iter/s)": 1.636085
    },
    {
      "acc": 0.65669012,
      "epoch": 1.5268899036022323,
      "grad_norm": 5.03125,
      "learning_rate": 1.4534351894347748e-06,
      "loss": 1.53870411,
      "memory(GiB)": 117.38,
      "step": 60190,
      "train_speed(iter/s)": 1.636099
    },
    {
      "acc": 0.66183615,
      "epoch": 1.5270167427701673,
      "grad_norm": 5.78125,
      "learning_rate": 1.4526960992166412e-06,
      "loss": 1.60380211,
      "memory(GiB)": 117.38,
      "step": 60195,
      "train_speed(iter/s)": 1.636113
    },
    {
      "acc": 0.65956669,
      "epoch": 1.5271435819381025,
      "grad_norm": 5.28125,
      "learning_rate": 1.4519571650247687e-06,
      "loss": 1.57495747,
      "memory(GiB)": 117.38,
      "step": 60200,
      "train_speed(iter/s)": 1.636127
    },
    {
      "acc": 0.66309075,
      "epoch": 1.5272704211060375,
      "grad_norm": 5.40625,
      "learning_rate": 1.4512183868916629e-06,
      "loss": 1.5950633,
      "memory(GiB)": 117.38,
      "step": 60205,
      "train_speed(iter/s)": 1.636139
    },
    {
      "acc": 0.6613862,
      "epoch": 1.5273972602739727,
      "grad_norm": 6.15625,
      "learning_rate": 1.4504797648498186e-06,
      "loss": 1.60185127,
      "memory(GiB)": 117.38,
      "step": 60210,
      "train_speed(iter/s)": 1.636153
    },
    {
      "acc": 0.6566926,
      "epoch": 1.5275240994419077,
      "grad_norm": 5.375,
      "learning_rate": 1.4497412989317184e-06,
      "loss": 1.57675056,
      "memory(GiB)": 117.38,
      "step": 60215,
      "train_speed(iter/s)": 1.636168
    },
    {
      "acc": 0.66068482,
      "epoch": 1.5276509386098427,
      "grad_norm": 5.65625,
      "learning_rate": 1.4490029891698476e-06,
      "loss": 1.6750309,
      "memory(GiB)": 117.38,
      "step": 60220,
      "train_speed(iter/s)": 1.636182
    },
    {
      "acc": 0.65352163,
      "epoch": 1.5277777777777777,
      "grad_norm": 5.84375,
      "learning_rate": 1.44826483559668e-06,
      "loss": 1.54409981,
      "memory(GiB)": 117.38,
      "step": 60225,
      "train_speed(iter/s)": 1.636196
    },
    {
      "acc": 0.66548691,
      "epoch": 1.527904616945713,
      "grad_norm": 6.03125,
      "learning_rate": 1.4475268382446833e-06,
      "loss": 1.52457895,
      "memory(GiB)": 117.38,
      "step": 60230,
      "train_speed(iter/s)": 1.63621
    },
    {
      "acc": 0.6594141,
      "epoch": 1.5280314561136479,
      "grad_norm": 6.125,
      "learning_rate": 1.4467889971463144e-06,
      "loss": 1.59770603,
      "memory(GiB)": 117.38,
      "step": 60235,
      "train_speed(iter/s)": 1.636226
    },
    {
      "acc": 0.66068048,
      "epoch": 1.528158295281583,
      "grad_norm": 4.875,
      "learning_rate": 1.4460513123340308e-06,
      "loss": 1.59796419,
      "memory(GiB)": 117.38,
      "step": 60240,
      "train_speed(iter/s)": 1.63624
    },
    {
      "acc": 0.66386471,
      "epoch": 1.528285134449518,
      "grad_norm": 6.75,
      "learning_rate": 1.4453137838402775e-06,
      "loss": 1.59874687,
      "memory(GiB)": 117.38,
      "step": 60245,
      "train_speed(iter/s)": 1.636255
    },
    {
      "acc": 0.65998678,
      "epoch": 1.528411973617453,
      "grad_norm": 5.6875,
      "learning_rate": 1.4445764116974948e-06,
      "loss": 1.54967842,
      "memory(GiB)": 117.38,
      "step": 60250,
      "train_speed(iter/s)": 1.636271
    },
    {
      "acc": 0.66433616,
      "epoch": 1.528538812785388,
      "grad_norm": 5.5625,
      "learning_rate": 1.4438391959381149e-06,
      "loss": 1.56788483,
      "memory(GiB)": 117.38,
      "step": 60255,
      "train_speed(iter/s)": 1.636285
    },
    {
      "acc": 0.65703015,
      "epoch": 1.528665651953323,
      "grad_norm": 6.0625,
      "learning_rate": 1.4431021365945647e-06,
      "loss": 1.58435555,
      "memory(GiB)": 117.38,
      "step": 60260,
      "train_speed(iter/s)": 1.636299
    },
    {
      "acc": 0.64913206,
      "epoch": 1.5287924911212583,
      "grad_norm": 6.375,
      "learning_rate": 1.4423652336992627e-06,
      "loss": 1.64434853,
      "memory(GiB)": 117.38,
      "step": 60265,
      "train_speed(iter/s)": 1.636313
    },
    {
      "acc": 0.65893993,
      "epoch": 1.5289193302891935,
      "grad_norm": 6.59375,
      "learning_rate": 1.4416284872846215e-06,
      "loss": 1.55997715,
      "memory(GiB)": 117.38,
      "step": 60270,
      "train_speed(iter/s)": 1.636328
    },
    {
      "acc": 0.67031555,
      "epoch": 1.5290461694571285,
      "grad_norm": 5.78125,
      "learning_rate": 1.440891897383046e-06,
      "loss": 1.48104639,
      "memory(GiB)": 117.38,
      "step": 60275,
      "train_speed(iter/s)": 1.636343
    },
    {
      "acc": 0.67872453,
      "epoch": 1.5291730086250634,
      "grad_norm": 7.15625,
      "learning_rate": 1.4401554640269354e-06,
      "loss": 1.45684309,
      "memory(GiB)": 117.38,
      "step": 60280,
      "train_speed(iter/s)": 1.636357
    },
    {
      "acc": 0.64625325,
      "epoch": 1.5292998477929984,
      "grad_norm": 5.9375,
      "learning_rate": 1.4394191872486812e-06,
      "loss": 1.64098358,
      "memory(GiB)": 117.38,
      "step": 60285,
      "train_speed(iter/s)": 1.63637
    },
    {
      "acc": 0.65403628,
      "epoch": 1.5294266869609334,
      "grad_norm": 6.125,
      "learning_rate": 1.4386830670806684e-06,
      "loss": 1.66142616,
      "memory(GiB)": 117.38,
      "step": 60290,
      "train_speed(iter/s)": 1.636384
    },
    {
      "acc": 0.66217656,
      "epoch": 1.5295535261288686,
      "grad_norm": 5.71875,
      "learning_rate": 1.4379471035552738e-06,
      "loss": 1.56906013,
      "memory(GiB)": 117.38,
      "step": 60295,
      "train_speed(iter/s)": 1.636398
    },
    {
      "acc": 0.65460219,
      "epoch": 1.5296803652968036,
      "grad_norm": 6.53125,
      "learning_rate": 1.437211296704869e-06,
      "loss": 1.63268814,
      "memory(GiB)": 117.38,
      "step": 60300,
      "train_speed(iter/s)": 1.636413
    },
    {
      "acc": 0.65590053,
      "epoch": 1.5298072044647388,
      "grad_norm": 4.71875,
      "learning_rate": 1.4364756465618167e-06,
      "loss": 1.62609615,
      "memory(GiB)": 117.38,
      "step": 60305,
      "train_speed(iter/s)": 1.636427
    },
    {
      "acc": 0.67777815,
      "epoch": 1.5299340436326738,
      "grad_norm": 5.9375,
      "learning_rate": 1.4357401531584792e-06,
      "loss": 1.53188763,
      "memory(GiB)": 117.38,
      "step": 60310,
      "train_speed(iter/s)": 1.636441
    },
    {
      "acc": 0.65823317,
      "epoch": 1.5300608828006088,
      "grad_norm": 6.59375,
      "learning_rate": 1.4350048165272006e-06,
      "loss": 1.5599124,
      "memory(GiB)": 117.38,
      "step": 60315,
      "train_speed(iter/s)": 1.636455
    },
    {
      "acc": 0.6543376,
      "epoch": 1.5301877219685438,
      "grad_norm": 6.5,
      "learning_rate": 1.4342696367003272e-06,
      "loss": 1.66451702,
      "memory(GiB)": 117.38,
      "step": 60320,
      "train_speed(iter/s)": 1.636468
    },
    {
      "acc": 0.64522972,
      "epoch": 1.5303145611364788,
      "grad_norm": 6.65625,
      "learning_rate": 1.433534613710193e-06,
      "loss": 1.55607014,
      "memory(GiB)": 117.38,
      "step": 60325,
      "train_speed(iter/s)": 1.636482
    },
    {
      "acc": 0.66271977,
      "epoch": 1.530441400304414,
      "grad_norm": 5.625,
      "learning_rate": 1.4327997475891331e-06,
      "loss": 1.57936897,
      "memory(GiB)": 117.38,
      "step": 60330,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.65755644,
      "epoch": 1.5305682394723492,
      "grad_norm": 6.75,
      "learning_rate": 1.432065038369465e-06,
      "loss": 1.58260489,
      "memory(GiB)": 117.38,
      "step": 60335,
      "train_speed(iter/s)": 1.63651
    },
    {
      "acc": 0.66864429,
      "epoch": 1.5306950786402842,
      "grad_norm": 5.5,
      "learning_rate": 1.4313304860835048e-06,
      "loss": 1.54122305,
      "memory(GiB)": 117.38,
      "step": 60340,
      "train_speed(iter/s)": 1.636523
    },
    {
      "acc": 0.65223794,
      "epoch": 1.5308219178082192,
      "grad_norm": 5.0625,
      "learning_rate": 1.4305960907635641e-06,
      "loss": 1.62357731,
      "memory(GiB)": 117.38,
      "step": 60345,
      "train_speed(iter/s)": 1.636537
    },
    {
      "acc": 0.66769967,
      "epoch": 1.5309487569761542,
      "grad_norm": 5.25,
      "learning_rate": 1.4298618524419455e-06,
      "loss": 1.58717089,
      "memory(GiB)": 117.38,
      "step": 60350,
      "train_speed(iter/s)": 1.636552
    },
    {
      "acc": 0.65245843,
      "epoch": 1.5310755961440892,
      "grad_norm": 5.1875,
      "learning_rate": 1.4291277711509388e-06,
      "loss": 1.64225197,
      "memory(GiB)": 117.38,
      "step": 60355,
      "train_speed(iter/s)": 1.636566
    },
    {
      "acc": 0.6647758,
      "epoch": 1.5312024353120244,
      "grad_norm": 5.71875,
      "learning_rate": 1.428393846922837e-06,
      "loss": 1.48412437,
      "memory(GiB)": 117.38,
      "step": 60360,
      "train_speed(iter/s)": 1.636581
    },
    {
      "acc": 0.67687397,
      "epoch": 1.5313292744799594,
      "grad_norm": 5.0625,
      "learning_rate": 1.4276600797899199e-06,
      "loss": 1.50028172,
      "memory(GiB)": 117.38,
      "step": 60365,
      "train_speed(iter/s)": 1.636595
    },
    {
      "acc": 0.66287665,
      "epoch": 1.5314561136478946,
      "grad_norm": 5.46875,
      "learning_rate": 1.426926469784463e-06,
      "loss": 1.56222439,
      "memory(GiB)": 117.38,
      "step": 60370,
      "train_speed(iter/s)": 1.636609
    },
    {
      "acc": 0.65942359,
      "epoch": 1.5315829528158296,
      "grad_norm": 7.375,
      "learning_rate": 1.42619301693873e-06,
      "loss": 1.66098785,
      "memory(GiB)": 117.38,
      "step": 60375,
      "train_speed(iter/s)": 1.636624
    },
    {
      "acc": 0.66916437,
      "epoch": 1.5317097919837646,
      "grad_norm": 5.375,
      "learning_rate": 1.4254597212849858e-06,
      "loss": 1.54763126,
      "memory(GiB)": 117.38,
      "step": 60380,
      "train_speed(iter/s)": 1.636638
    },
    {
      "acc": 0.65305176,
      "epoch": 1.5318366311516995,
      "grad_norm": 6.0625,
      "learning_rate": 1.4247265828554819e-06,
      "loss": 1.60305157,
      "memory(GiB)": 117.38,
      "step": 60385,
      "train_speed(iter/s)": 1.636652
    },
    {
      "acc": 0.64701223,
      "epoch": 1.5319634703196348,
      "grad_norm": 6.4375,
      "learning_rate": 1.423993601682465e-06,
      "loss": 1.64670525,
      "memory(GiB)": 117.38,
      "step": 60390,
      "train_speed(iter/s)": 1.636666
    },
    {
      "acc": 0.6553968,
      "epoch": 1.5320903094875697,
      "grad_norm": 5.46875,
      "learning_rate": 1.423260777798176e-06,
      "loss": 1.58378267,
      "memory(GiB)": 117.38,
      "step": 60395,
      "train_speed(iter/s)": 1.636681
    },
    {
      "acc": 0.65751104,
      "epoch": 1.532217148655505,
      "grad_norm": 6.78125,
      "learning_rate": 1.4225281112348466e-06,
      "loss": 1.61481094,
      "memory(GiB)": 117.38,
      "step": 60400,
      "train_speed(iter/s)": 1.636695
    },
    {
      "acc": 0.63155832,
      "epoch": 1.53234398782344,
      "grad_norm": 6.84375,
      "learning_rate": 1.421795602024703e-06,
      "loss": 1.69914551,
      "memory(GiB)": 117.38,
      "step": 60405,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.66281881,
      "epoch": 1.532470826991375,
      "grad_norm": 6.25,
      "learning_rate": 1.4210632501999643e-06,
      "loss": 1.60844517,
      "memory(GiB)": 117.38,
      "step": 60410,
      "train_speed(iter/s)": 1.636723
    },
    {
      "acc": 0.65548601,
      "epoch": 1.53259766615931,
      "grad_norm": 5.28125,
      "learning_rate": 1.4203310557928428e-06,
      "loss": 1.58384342,
      "memory(GiB)": 117.38,
      "step": 60415,
      "train_speed(iter/s)": 1.636737
    },
    {
      "acc": 0.6365365,
      "epoch": 1.532724505327245,
      "grad_norm": 5.78125,
      "learning_rate": 1.4195990188355435e-06,
      "loss": 1.71370163,
      "memory(GiB)": 117.38,
      "step": 60420,
      "train_speed(iter/s)": 1.636752
    },
    {
      "acc": 0.65308132,
      "epoch": 1.5328513444951801,
      "grad_norm": 7.625,
      "learning_rate": 1.418867139360265e-06,
      "loss": 1.57925014,
      "memory(GiB)": 117.38,
      "step": 60425,
      "train_speed(iter/s)": 1.636765
    },
    {
      "acc": 0.6641264,
      "epoch": 1.5329781836631153,
      "grad_norm": 7.0625,
      "learning_rate": 1.418135417399198e-06,
      "loss": 1.58189831,
      "memory(GiB)": 117.38,
      "step": 60430,
      "train_speed(iter/s)": 1.63678
    },
    {
      "acc": 0.64957228,
      "epoch": 1.5331050228310503,
      "grad_norm": 7.4375,
      "learning_rate": 1.4174038529845273e-06,
      "loss": 1.67629395,
      "memory(GiB)": 117.38,
      "step": 60435,
      "train_speed(iter/s)": 1.636795
    },
    {
      "acc": 0.64387145,
      "epoch": 1.5332318619989853,
      "grad_norm": 6.625,
      "learning_rate": 1.4166724461484304e-06,
      "loss": 1.64619179,
      "memory(GiB)": 117.38,
      "step": 60440,
      "train_speed(iter/s)": 1.63681
    },
    {
      "acc": 0.64260406,
      "epoch": 1.5333587011669203,
      "grad_norm": 4.8125,
      "learning_rate": 1.4159411969230758e-06,
      "loss": 1.6511631,
      "memory(GiB)": 117.38,
      "step": 60445,
      "train_speed(iter/s)": 1.636824
    },
    {
      "acc": 0.65834703,
      "epoch": 1.5334855403348553,
      "grad_norm": 5.1875,
      "learning_rate": 1.4152101053406325e-06,
      "loss": 1.5497858,
      "memory(GiB)": 117.38,
      "step": 60450,
      "train_speed(iter/s)": 1.636838
    },
    {
      "acc": 0.65695348,
      "epoch": 1.5336123795027905,
      "grad_norm": 5.59375,
      "learning_rate": 1.4144791714332517e-06,
      "loss": 1.56492538,
      "memory(GiB)": 117.38,
      "step": 60455,
      "train_speed(iter/s)": 1.636852
    },
    {
      "acc": 0.65435057,
      "epoch": 1.5337392186707255,
      "grad_norm": 5.375,
      "learning_rate": 1.4137483952330855e-06,
      "loss": 1.60241508,
      "memory(GiB)": 117.38,
      "step": 60460,
      "train_speed(iter/s)": 1.636867
    },
    {
      "acc": 0.66194444,
      "epoch": 1.5338660578386607,
      "grad_norm": 6.96875,
      "learning_rate": 1.4130177767722753e-06,
      "loss": 1.59549732,
      "memory(GiB)": 117.38,
      "step": 60465,
      "train_speed(iter/s)": 1.636881
    },
    {
      "acc": 0.66424203,
      "epoch": 1.5339928970065957,
      "grad_norm": 5.75,
      "learning_rate": 1.4122873160829603e-06,
      "loss": 1.5887455,
      "memory(GiB)": 117.38,
      "step": 60470,
      "train_speed(iter/s)": 1.636894
    },
    {
      "acc": 0.64726505,
      "epoch": 1.5341197361745307,
      "grad_norm": 4.65625,
      "learning_rate": 1.4115570131972655e-06,
      "loss": 1.62380295,
      "memory(GiB)": 117.38,
      "step": 60475,
      "train_speed(iter/s)": 1.636909
    },
    {
      "acc": 0.66210413,
      "epoch": 1.5342465753424657,
      "grad_norm": 5.21875,
      "learning_rate": 1.4108268681473136e-06,
      "loss": 1.61364632,
      "memory(GiB)": 117.38,
      "step": 60480,
      "train_speed(iter/s)": 1.636923
    },
    {
      "acc": 0.66844954,
      "epoch": 1.5343734145104007,
      "grad_norm": 9.0,
      "learning_rate": 1.4100968809652215e-06,
      "loss": 1.62208824,
      "memory(GiB)": 117.38,
      "step": 60485,
      "train_speed(iter/s)": 1.636936
    },
    {
      "acc": 0.66008363,
      "epoch": 1.5345002536783359,
      "grad_norm": 6.9375,
      "learning_rate": 1.4093670516830982e-06,
      "loss": 1.60053558,
      "memory(GiB)": 117.38,
      "step": 60490,
      "train_speed(iter/s)": 1.636951
    },
    {
      "acc": 0.6450129,
      "epoch": 1.534627092846271,
      "grad_norm": 6.28125,
      "learning_rate": 1.4086373803330417e-06,
      "loss": 1.60832767,
      "memory(GiB)": 117.38,
      "step": 60495,
      "train_speed(iter/s)": 1.636965
    },
    {
      "acc": 0.65314512,
      "epoch": 1.534753932014206,
      "grad_norm": 6.15625,
      "learning_rate": 1.4079078669471457e-06,
      "loss": 1.64581146,
      "memory(GiB)": 117.38,
      "step": 60500,
      "train_speed(iter/s)": 1.636979
    },
    {
      "acc": 0.65411048,
      "epoch": 1.534880771182141,
      "grad_norm": 6.15625,
      "learning_rate": 1.4071785115575005e-06,
      "loss": 1.56350098,
      "memory(GiB)": 117.38,
      "step": 60505,
      "train_speed(iter/s)": 1.636993
    },
    {
      "acc": 0.67226439,
      "epoch": 1.535007610350076,
      "grad_norm": 7.125,
      "learning_rate": 1.4064493141961872e-06,
      "loss": 1.5487051,
      "memory(GiB)": 117.38,
      "step": 60510,
      "train_speed(iter/s)": 1.637007
    },
    {
      "acc": 0.67400651,
      "epoch": 1.535134449518011,
      "grad_norm": 6.96875,
      "learning_rate": 1.4057202748952736e-06,
      "loss": 1.58081856,
      "memory(GiB)": 117.38,
      "step": 60515,
      "train_speed(iter/s)": 1.637022
    },
    {
      "acc": 0.66180763,
      "epoch": 1.5352612886859462,
      "grad_norm": 6.90625,
      "learning_rate": 1.4049913936868314e-06,
      "loss": 1.55781975,
      "memory(GiB)": 117.38,
      "step": 60520,
      "train_speed(iter/s)": 1.637036
    },
    {
      "acc": 0.64300485,
      "epoch": 1.5353881278538812,
      "grad_norm": 5.8125,
      "learning_rate": 1.4042626706029184e-06,
      "loss": 1.63914547,
      "memory(GiB)": 117.38,
      "step": 60525,
      "train_speed(iter/s)": 1.63705
    },
    {
      "acc": 0.6416573,
      "epoch": 1.5355149670218164,
      "grad_norm": 6.03125,
      "learning_rate": 1.4035341056755864e-06,
      "loss": 1.64014778,
      "memory(GiB)": 117.38,
      "step": 60530,
      "train_speed(iter/s)": 1.637064
    },
    {
      "acc": 0.67381473,
      "epoch": 1.5356418061897514,
      "grad_norm": 6.4375,
      "learning_rate": 1.402805698936882e-06,
      "loss": 1.56713543,
      "memory(GiB)": 117.38,
      "step": 60535,
      "train_speed(iter/s)": 1.637078
    },
    {
      "acc": 0.65821733,
      "epoch": 1.5357686453576864,
      "grad_norm": 5.1875,
      "learning_rate": 1.4020774504188428e-06,
      "loss": 1.6099659,
      "memory(GiB)": 117.38,
      "step": 60540,
      "train_speed(iter/s)": 1.637093
    },
    {
      "acc": 0.65900126,
      "epoch": 1.5358954845256214,
      "grad_norm": 5.625,
      "learning_rate": 1.4013493601535016e-06,
      "loss": 1.60410881,
      "memory(GiB)": 117.38,
      "step": 60545,
      "train_speed(iter/s)": 1.637107
    },
    {
      "acc": 0.6433095,
      "epoch": 1.5360223236935566,
      "grad_norm": 9.6875,
      "learning_rate": 1.400621428172882e-06,
      "loss": 1.68636627,
      "memory(GiB)": 117.38,
      "step": 60550,
      "train_speed(iter/s)": 1.637121
    },
    {
      "acc": 0.65696101,
      "epoch": 1.5361491628614916,
      "grad_norm": 6.875,
      "learning_rate": 1.399893654509002e-06,
      "loss": 1.64220657,
      "memory(GiB)": 117.38,
      "step": 60555,
      "train_speed(iter/s)": 1.637136
    },
    {
      "acc": 0.64825425,
      "epoch": 1.5362760020294268,
      "grad_norm": 5.40625,
      "learning_rate": 1.3991660391938721e-06,
      "loss": 1.62121353,
      "memory(GiB)": 117.38,
      "step": 60560,
      "train_speed(iter/s)": 1.63715
    },
    {
      "acc": 0.65843449,
      "epoch": 1.5364028411973618,
      "grad_norm": 7.1875,
      "learning_rate": 1.398438582259497e-06,
      "loss": 1.61310921,
      "memory(GiB)": 117.38,
      "step": 60565,
      "train_speed(iter/s)": 1.637165
    },
    {
      "acc": 0.65139036,
      "epoch": 1.5365296803652968,
      "grad_norm": 6.375,
      "learning_rate": 1.3977112837378726e-06,
      "loss": 1.60211563,
      "memory(GiB)": 117.38,
      "step": 60570,
      "train_speed(iter/s)": 1.63718
    },
    {
      "acc": 0.64712067,
      "epoch": 1.5366565195332318,
      "grad_norm": 5.59375,
      "learning_rate": 1.3969841436609888e-06,
      "loss": 1.6357235,
      "memory(GiB)": 117.38,
      "step": 60575,
      "train_speed(iter/s)": 1.637194
    },
    {
      "acc": 0.6533452,
      "epoch": 1.5367833587011668,
      "grad_norm": 4.90625,
      "learning_rate": 1.396257162060829e-06,
      "loss": 1.58391037,
      "memory(GiB)": 117.38,
      "step": 60580,
      "train_speed(iter/s)": 1.637208
    },
    {
      "acc": 0.63941755,
      "epoch": 1.536910197869102,
      "grad_norm": 4.625,
      "learning_rate": 1.395530338969367e-06,
      "loss": 1.67452774,
      "memory(GiB)": 117.38,
      "step": 60585,
      "train_speed(iter/s)": 1.637222
    },
    {
      "acc": 0.64957657,
      "epoch": 1.5370370370370372,
      "grad_norm": 5.5,
      "learning_rate": 1.3948036744185767e-06,
      "loss": 1.59090881,
      "memory(GiB)": 117.38,
      "step": 60590,
      "train_speed(iter/s)": 1.637237
    },
    {
      "acc": 0.64799085,
      "epoch": 1.5371638762049722,
      "grad_norm": 6.46875,
      "learning_rate": 1.3940771684404153e-06,
      "loss": 1.56088428,
      "memory(GiB)": 117.38,
      "step": 60595,
      "train_speed(iter/s)": 1.637251
    },
    {
      "acc": 0.66535153,
      "epoch": 1.5372907153729072,
      "grad_norm": 5.5,
      "learning_rate": 1.393350821066839e-06,
      "loss": 1.53039045,
      "memory(GiB)": 117.38,
      "step": 60600,
      "train_speed(iter/s)": 1.637264
    },
    {
      "acc": 0.67648168,
      "epoch": 1.5374175545408422,
      "grad_norm": 5.40625,
      "learning_rate": 1.3926246323297948e-06,
      "loss": 1.56906986,
      "memory(GiB)": 117.38,
      "step": 60605,
      "train_speed(iter/s)": 1.637278
    },
    {
      "acc": 0.6554822,
      "epoch": 1.5375443937087772,
      "grad_norm": 5.71875,
      "learning_rate": 1.3918986022612285e-06,
      "loss": 1.53117218,
      "memory(GiB)": 117.38,
      "step": 60610,
      "train_speed(iter/s)": 1.637293
    },
    {
      "acc": 0.66043124,
      "epoch": 1.5376712328767124,
      "grad_norm": 6.78125,
      "learning_rate": 1.3911727308930684e-06,
      "loss": 1.61840401,
      "memory(GiB)": 117.38,
      "step": 60615,
      "train_speed(iter/s)": 1.637306
    },
    {
      "acc": 0.64953861,
      "epoch": 1.5377980720446474,
      "grad_norm": 6.65625,
      "learning_rate": 1.3904470182572428e-06,
      "loss": 1.60523891,
      "memory(GiB)": 117.38,
      "step": 60620,
      "train_speed(iter/s)": 1.637319
    },
    {
      "acc": 0.6532434,
      "epoch": 1.5379249112125826,
      "grad_norm": 5.9375,
      "learning_rate": 1.3897214643856744e-06,
      "loss": 1.67258987,
      "memory(GiB)": 117.38,
      "step": 60625,
      "train_speed(iter/s)": 1.637333
    },
    {
      "acc": 0.67551403,
      "epoch": 1.5380517503805176,
      "grad_norm": 7.625,
      "learning_rate": 1.388996069310276e-06,
      "loss": 1.53150196,
      "memory(GiB)": 117.38,
      "step": 60630,
      "train_speed(iter/s)": 1.637345
    },
    {
      "acc": 0.65537319,
      "epoch": 1.5381785895484525,
      "grad_norm": 5.34375,
      "learning_rate": 1.3882708330629514e-06,
      "loss": 1.6425642,
      "memory(GiB)": 117.38,
      "step": 60635,
      "train_speed(iter/s)": 1.63736
    },
    {
      "acc": 0.65168991,
      "epoch": 1.5383054287163875,
      "grad_norm": 5.21875,
      "learning_rate": 1.3875457556755989e-06,
      "loss": 1.61624603,
      "memory(GiB)": 117.38,
      "step": 60640,
      "train_speed(iter/s)": 1.637375
    },
    {
      "acc": 0.65625429,
      "epoch": 1.5384322678843225,
      "grad_norm": 6.0,
      "learning_rate": 1.386820837180114e-06,
      "loss": 1.56737709,
      "memory(GiB)": 117.38,
      "step": 60645,
      "train_speed(iter/s)": 1.637388
    },
    {
      "acc": 0.66664801,
      "epoch": 1.5385591070522577,
      "grad_norm": 6.0,
      "learning_rate": 1.386096077608382e-06,
      "loss": 1.56442862,
      "memory(GiB)": 117.38,
      "step": 60650,
      "train_speed(iter/s)": 1.637401
    },
    {
      "acc": 0.65851536,
      "epoch": 1.538685946220193,
      "grad_norm": 5.3125,
      "learning_rate": 1.385371476992276e-06,
      "loss": 1.5890173,
      "memory(GiB)": 117.38,
      "step": 60655,
      "train_speed(iter/s)": 1.637415
    },
    {
      "acc": 0.6478879,
      "epoch": 1.538812785388128,
      "grad_norm": 5.8125,
      "learning_rate": 1.3846470353636726e-06,
      "loss": 1.63984299,
      "memory(GiB)": 117.38,
      "step": 60660,
      "train_speed(iter/s)": 1.637429
    },
    {
      "acc": 0.6499064,
      "epoch": 1.538939624556063,
      "grad_norm": 4.84375,
      "learning_rate": 1.3839227527544336e-06,
      "loss": 1.61538315,
      "memory(GiB)": 117.38,
      "step": 60665,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.6608439,
      "epoch": 1.539066463723998,
      "grad_norm": 5.4375,
      "learning_rate": 1.3831986291964184e-06,
      "loss": 1.5654541,
      "memory(GiB)": 117.38,
      "step": 60670,
      "train_speed(iter/s)": 1.637457
    },
    {
      "acc": 0.65856328,
      "epoch": 1.539193302891933,
      "grad_norm": 6.5,
      "learning_rate": 1.382474664721472e-06,
      "loss": 1.61815567,
      "memory(GiB)": 117.38,
      "step": 60675,
      "train_speed(iter/s)": 1.63747
    },
    {
      "acc": 0.65965481,
      "epoch": 1.5393201420598681,
      "grad_norm": 5.96875,
      "learning_rate": 1.3817508593614425e-06,
      "loss": 1.5282938,
      "memory(GiB)": 117.38,
      "step": 60680,
      "train_speed(iter/s)": 1.637484
    },
    {
      "acc": 0.65463648,
      "epoch": 1.539446981227803,
      "grad_norm": 5.28125,
      "learning_rate": 1.381027213148165e-06,
      "loss": 1.63583755,
      "memory(GiB)": 117.38,
      "step": 60685,
      "train_speed(iter/s)": 1.637498
    },
    {
      "acc": 0.64780493,
      "epoch": 1.5395738203957383,
      "grad_norm": 6.9375,
      "learning_rate": 1.3803037261134678e-06,
      "loss": 1.61269207,
      "memory(GiB)": 117.38,
      "step": 60690,
      "train_speed(iter/s)": 1.637513
    },
    {
      "acc": 0.65708666,
      "epoch": 1.5397006595636733,
      "grad_norm": 6.0625,
      "learning_rate": 1.3795803982891736e-06,
      "loss": 1.61737595,
      "memory(GiB)": 117.38,
      "step": 60695,
      "train_speed(iter/s)": 1.637525
    },
    {
      "acc": 0.65703225,
      "epoch": 1.5398274987316083,
      "grad_norm": 5.4375,
      "learning_rate": 1.3788572297070974e-06,
      "loss": 1.56723289,
      "memory(GiB)": 117.38,
      "step": 60700,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.66658363,
      "epoch": 1.5399543378995433,
      "grad_norm": 4.8125,
      "learning_rate": 1.3781342203990478e-06,
      "loss": 1.57033339,
      "memory(GiB)": 117.38,
      "step": 60705,
      "train_speed(iter/s)": 1.637553
    },
    {
      "acc": 0.65937982,
      "epoch": 1.5400811770674785,
      "grad_norm": 5.84375,
      "learning_rate": 1.3774113703968255e-06,
      "loss": 1.62580185,
      "memory(GiB)": 117.38,
      "step": 60710,
      "train_speed(iter/s)": 1.637567
    },
    {
      "acc": 0.6533318,
      "epoch": 1.5402080162354135,
      "grad_norm": 5.9375,
      "learning_rate": 1.3766886797322248e-06,
      "loss": 1.56688538,
      "memory(GiB)": 117.38,
      "step": 60715,
      "train_speed(iter/s)": 1.637581
    },
    {
      "acc": 0.67758055,
      "epoch": 1.5403348554033487,
      "grad_norm": 4.65625,
      "learning_rate": 1.3759661484370324e-06,
      "loss": 1.50598564,
      "memory(GiB)": 117.38,
      "step": 60720,
      "train_speed(iter/s)": 1.637594
    },
    {
      "acc": 0.66571622,
      "epoch": 1.5404616945712837,
      "grad_norm": 6.65625,
      "learning_rate": 1.3752437765430294e-06,
      "loss": 1.55234203,
      "memory(GiB)": 117.38,
      "step": 60725,
      "train_speed(iter/s)": 1.637609
    },
    {
      "acc": 0.66015892,
      "epoch": 1.5405885337392187,
      "grad_norm": 8.0625,
      "learning_rate": 1.3745215640819886e-06,
      "loss": 1.59730568,
      "memory(GiB)": 117.38,
      "step": 60730,
      "train_speed(iter/s)": 1.637623
    },
    {
      "acc": 0.67173948,
      "epoch": 1.5407153729071537,
      "grad_norm": 6.03125,
      "learning_rate": 1.373799511085676e-06,
      "loss": 1.5125762,
      "memory(GiB)": 117.38,
      "step": 60735,
      "train_speed(iter/s)": 1.637637
    },
    {
      "acc": 0.65377264,
      "epoch": 1.5408422120750886,
      "grad_norm": 5.34375,
      "learning_rate": 1.3730776175858506e-06,
      "loss": 1.63221302,
      "memory(GiB)": 117.38,
      "step": 60740,
      "train_speed(iter/s)": 1.63765
    },
    {
      "acc": 0.66715221,
      "epoch": 1.5409690512430239,
      "grad_norm": 5.78125,
      "learning_rate": 1.3723558836142631e-06,
      "loss": 1.58224907,
      "memory(GiB)": 117.38,
      "step": 60745,
      "train_speed(iter/s)": 1.637664
    },
    {
      "acc": 0.6688014,
      "epoch": 1.541095890410959,
      "grad_norm": 4.8125,
      "learning_rate": 1.371634309202663e-06,
      "loss": 1.55853424,
      "memory(GiB)": 117.38,
      "step": 60750,
      "train_speed(iter/s)": 1.637678
    },
    {
      "acc": 0.65355935,
      "epoch": 1.541222729578894,
      "grad_norm": 6.28125,
      "learning_rate": 1.3709128943827842e-06,
      "loss": 1.62020283,
      "memory(GiB)": 117.38,
      "step": 60755,
      "train_speed(iter/s)": 1.637691
    },
    {
      "acc": 0.63813562,
      "epoch": 1.541349568746829,
      "grad_norm": 6.03125,
      "learning_rate": 1.3701916391863573e-06,
      "loss": 1.64351025,
      "memory(GiB)": 117.38,
      "step": 60760,
      "train_speed(iter/s)": 1.637706
    },
    {
      "acc": 0.64319868,
      "epoch": 1.541476407914764,
      "grad_norm": 6.09375,
      "learning_rate": 1.3694705436451093e-06,
      "loss": 1.62536964,
      "memory(GiB)": 117.38,
      "step": 60765,
      "train_speed(iter/s)": 1.63772
    },
    {
      "acc": 0.6715888,
      "epoch": 1.541603247082699,
      "grad_norm": 5.84375,
      "learning_rate": 1.368749607790758e-06,
      "loss": 1.51326818,
      "memory(GiB)": 117.38,
      "step": 60770,
      "train_speed(iter/s)": 1.637733
    },
    {
      "acc": 0.66085086,
      "epoch": 1.5417300862506342,
      "grad_norm": 5.625,
      "learning_rate": 1.3680288316550095e-06,
      "loss": 1.55030251,
      "memory(GiB)": 117.38,
      "step": 60775,
      "train_speed(iter/s)": 1.637747
    },
    {
      "acc": 0.64841723,
      "epoch": 1.5418569254185692,
      "grad_norm": 6.90625,
      "learning_rate": 1.3673082152695672e-06,
      "loss": 1.66748238,
      "memory(GiB)": 117.38,
      "step": 60780,
      "train_speed(iter/s)": 1.637761
    },
    {
      "acc": 0.63058891,
      "epoch": 1.5419837645865044,
      "grad_norm": 4.96875,
      "learning_rate": 1.3665877586661296e-06,
      "loss": 1.67498512,
      "memory(GiB)": 117.38,
      "step": 60785,
      "train_speed(iter/s)": 1.637775
    },
    {
      "acc": 0.66198859,
      "epoch": 1.5421106037544394,
      "grad_norm": 5.5,
      "learning_rate": 1.3658674618763862e-06,
      "loss": 1.57020464,
      "memory(GiB)": 117.38,
      "step": 60790,
      "train_speed(iter/s)": 1.637788
    },
    {
      "acc": 0.64411054,
      "epoch": 1.5422374429223744,
      "grad_norm": 6.65625,
      "learning_rate": 1.365147324932014e-06,
      "loss": 1.59758244,
      "memory(GiB)": 117.38,
      "step": 60795,
      "train_speed(iter/s)": 1.637802
    },
    {
      "acc": 0.64423852,
      "epoch": 1.5423642820903094,
      "grad_norm": 6.53125,
      "learning_rate": 1.3644273478646925e-06,
      "loss": 1.68765011,
      "memory(GiB)": 117.38,
      "step": 60800,
      "train_speed(iter/s)": 1.637816
    },
    {
      "acc": 0.65495815,
      "epoch": 1.5424911212582444,
      "grad_norm": 5.75,
      "learning_rate": 1.3637075307060877e-06,
      "loss": 1.62945271,
      "memory(GiB)": 117.38,
      "step": 60805,
      "train_speed(iter/s)": 1.63783
    },
    {
      "acc": 0.66216636,
      "epoch": 1.5426179604261796,
      "grad_norm": 5.1875,
      "learning_rate": 1.362987873487862e-06,
      "loss": 1.5814949,
      "memory(GiB)": 117.38,
      "step": 60810,
      "train_speed(iter/s)": 1.637845
    },
    {
      "acc": 0.6472908,
      "epoch": 1.5427447995941148,
      "grad_norm": 5.4375,
      "learning_rate": 1.362268376241665e-06,
      "loss": 1.59065418,
      "memory(GiB)": 117.38,
      "step": 60815,
      "train_speed(iter/s)": 1.637858
    },
    {
      "acc": 0.63248982,
      "epoch": 1.5428716387620498,
      "grad_norm": 7.28125,
      "learning_rate": 1.3615490389991476e-06,
      "loss": 1.71146088,
      "memory(GiB)": 117.38,
      "step": 60820,
      "train_speed(iter/s)": 1.637872
    },
    {
      "acc": 0.65516539,
      "epoch": 1.5429984779299848,
      "grad_norm": 5.34375,
      "learning_rate": 1.360829861791948e-06,
      "loss": 1.57450037,
      "memory(GiB)": 117.38,
      "step": 60825,
      "train_speed(iter/s)": 1.637886
    },
    {
      "acc": 0.6589664,
      "epoch": 1.5431253170979198,
      "grad_norm": 6.46875,
      "learning_rate": 1.3601108446516985e-06,
      "loss": 1.6026823,
      "memory(GiB)": 117.38,
      "step": 60830,
      "train_speed(iter/s)": 1.6379
    },
    {
      "acc": 0.67016191,
      "epoch": 1.5432521562658548,
      "grad_norm": 5.875,
      "learning_rate": 1.3593919876100254e-06,
      "loss": 1.54794121,
      "memory(GiB)": 117.38,
      "step": 60835,
      "train_speed(iter/s)": 1.637914
    },
    {
      "acc": 0.64960527,
      "epoch": 1.54337899543379,
      "grad_norm": 6.15625,
      "learning_rate": 1.3586732906985467e-06,
      "loss": 1.63107777,
      "memory(GiB)": 117.38,
      "step": 60840,
      "train_speed(iter/s)": 1.637928
    },
    {
      "acc": 0.66171756,
      "epoch": 1.543505834601725,
      "grad_norm": 6.15625,
      "learning_rate": 1.357954753948874e-06,
      "loss": 1.58175297,
      "memory(GiB)": 117.38,
      "step": 60845,
      "train_speed(iter/s)": 1.637942
    },
    {
      "acc": 0.68012829,
      "epoch": 1.5436326737696602,
      "grad_norm": 5.375,
      "learning_rate": 1.3572363773926117e-06,
      "loss": 1.52260113,
      "memory(GiB)": 117.38,
      "step": 60850,
      "train_speed(iter/s)": 1.637955
    },
    {
      "acc": 0.66428514,
      "epoch": 1.5437595129375952,
      "grad_norm": 7.0,
      "learning_rate": 1.3565181610613571e-06,
      "loss": 1.61247501,
      "memory(GiB)": 117.38,
      "step": 60855,
      "train_speed(iter/s)": 1.637968
    },
    {
      "acc": 0.65341148,
      "epoch": 1.5438863521055302,
      "grad_norm": 6.3125,
      "learning_rate": 1.3558001049867008e-06,
      "loss": 1.58205357,
      "memory(GiB)": 117.38,
      "step": 60860,
      "train_speed(iter/s)": 1.637981
    },
    {
      "acc": 0.66341696,
      "epoch": 1.5440131912734651,
      "grad_norm": 5.375,
      "learning_rate": 1.3550822092002264e-06,
      "loss": 1.57977371,
      "memory(GiB)": 117.38,
      "step": 60865,
      "train_speed(iter/s)": 1.637995
    },
    {
      "acc": 0.66312504,
      "epoch": 1.5441400304414004,
      "grad_norm": 4.6875,
      "learning_rate": 1.3543644737335099e-06,
      "loss": 1.50969276,
      "memory(GiB)": 117.38,
      "step": 60870,
      "train_speed(iter/s)": 1.638008
    },
    {
      "acc": 0.64081087,
      "epoch": 1.5442668696093353,
      "grad_norm": 6.28125,
      "learning_rate": 1.35364689861812e-06,
      "loss": 1.63472366,
      "memory(GiB)": 117.38,
      "step": 60875,
      "train_speed(iter/s)": 1.638023
    },
    {
      "acc": 0.67057085,
      "epoch": 1.5443937087772706,
      "grad_norm": 6.53125,
      "learning_rate": 1.3529294838856194e-06,
      "loss": 1.56202316,
      "memory(GiB)": 117.38,
      "step": 60880,
      "train_speed(iter/s)": 1.638037
    },
    {
      "acc": 0.64279742,
      "epoch": 1.5445205479452055,
      "grad_norm": 6.25,
      "learning_rate": 1.3522122295675616e-06,
      "loss": 1.59809437,
      "memory(GiB)": 117.38,
      "step": 60885,
      "train_speed(iter/s)": 1.638051
    },
    {
      "acc": 0.65396233,
      "epoch": 1.5446473871131405,
      "grad_norm": 7.0625,
      "learning_rate": 1.351495135695499e-06,
      "loss": 1.61423817,
      "memory(GiB)": 117.38,
      "step": 60890,
      "train_speed(iter/s)": 1.638065
    },
    {
      "acc": 0.6627183,
      "epoch": 1.5447742262810755,
      "grad_norm": 5.8125,
      "learning_rate": 1.3507782023009692e-06,
      "loss": 1.58792992,
      "memory(GiB)": 117.38,
      "step": 60895,
      "train_speed(iter/s)": 1.638079
    },
    {
      "acc": 0.66423535,
      "epoch": 1.5449010654490105,
      "grad_norm": 5.15625,
      "learning_rate": 1.3500614294155056e-06,
      "loss": 1.57492504,
      "memory(GiB)": 117.38,
      "step": 60900,
      "train_speed(iter/s)": 1.638092
    },
    {
      "acc": 0.67106886,
      "epoch": 1.5450279046169457,
      "grad_norm": 5.625,
      "learning_rate": 1.3493448170706347e-06,
      "loss": 1.60399418,
      "memory(GiB)": 117.38,
      "step": 60905,
      "train_speed(iter/s)": 1.638105
    },
    {
      "acc": 0.65734005,
      "epoch": 1.545154743784881,
      "grad_norm": 5.25,
      "learning_rate": 1.348628365297881e-06,
      "loss": 1.59561119,
      "memory(GiB)": 117.38,
      "step": 60910,
      "train_speed(iter/s)": 1.63812
    },
    {
      "acc": 0.65490484,
      "epoch": 1.545281582952816,
      "grad_norm": 6.03125,
      "learning_rate": 1.3479120741287526e-06,
      "loss": 1.59032516,
      "memory(GiB)": 117.38,
      "step": 60915,
      "train_speed(iter/s)": 1.638134
    },
    {
      "acc": 0.65446835,
      "epoch": 1.545408422120751,
      "grad_norm": 5.40625,
      "learning_rate": 1.3471959435947552e-06,
      "loss": 1.66792221,
      "memory(GiB)": 117.38,
      "step": 60920,
      "train_speed(iter/s)": 1.638147
    },
    {
      "acc": 0.64600048,
      "epoch": 1.545535261288686,
      "grad_norm": 5.1875,
      "learning_rate": 1.3464799737273898e-06,
      "loss": 1.63425903,
      "memory(GiB)": 117.38,
      "step": 60925,
      "train_speed(iter/s)": 1.63816
    },
    {
      "acc": 0.65153418,
      "epoch": 1.545662100456621,
      "grad_norm": 6.1875,
      "learning_rate": 1.3457641645581487e-06,
      "loss": 1.60337391,
      "memory(GiB)": 117.38,
      "step": 60930,
      "train_speed(iter/s)": 1.638174
    },
    {
      "acc": 0.64624863,
      "epoch": 1.545788939624556,
      "grad_norm": 5.84375,
      "learning_rate": 1.3450485161185133e-06,
      "loss": 1.64083881,
      "memory(GiB)": 117.38,
      "step": 60935,
      "train_speed(iter/s)": 1.638187
    },
    {
      "acc": 0.65853739,
      "epoch": 1.545915778792491,
      "grad_norm": 7.6875,
      "learning_rate": 1.344333028439961e-06,
      "loss": 1.56394987,
      "memory(GiB)": 117.38,
      "step": 60940,
      "train_speed(iter/s)": 1.638201
    },
    {
      "acc": 0.67052164,
      "epoch": 1.5460426179604263,
      "grad_norm": 5.8125,
      "learning_rate": 1.3436177015539647e-06,
      "loss": 1.54976892,
      "memory(GiB)": 117.38,
      "step": 60945,
      "train_speed(iter/s)": 1.638216
    },
    {
      "acc": 0.67395668,
      "epoch": 1.5461694571283613,
      "grad_norm": 6.15625,
      "learning_rate": 1.3429025354919877e-06,
      "loss": 1.548804,
      "memory(GiB)": 117.38,
      "step": 60950,
      "train_speed(iter/s)": 1.638229
    },
    {
      "acc": 0.64560823,
      "epoch": 1.5462962962962963,
      "grad_norm": 6.09375,
      "learning_rate": 1.3421875302854826e-06,
      "loss": 1.62121315,
      "memory(GiB)": 117.38,
      "step": 60955,
      "train_speed(iter/s)": 1.638243
    },
    {
      "acc": 0.64463673,
      "epoch": 1.5464231354642313,
      "grad_norm": 5.625,
      "learning_rate": 1.3414726859659016e-06,
      "loss": 1.62692986,
      "memory(GiB)": 117.38,
      "step": 60960,
      "train_speed(iter/s)": 1.638257
    },
    {
      "acc": 0.63233428,
      "epoch": 1.5465499746321663,
      "grad_norm": 5.5625,
      "learning_rate": 1.3407580025646866e-06,
      "loss": 1.6654583,
      "memory(GiB)": 117.38,
      "step": 60965,
      "train_speed(iter/s)": 1.638271
    },
    {
      "acc": 0.64950213,
      "epoch": 1.5466768138001015,
      "grad_norm": 7.8125,
      "learning_rate": 1.3400434801132716e-06,
      "loss": 1.63694344,
      "memory(GiB)": 117.38,
      "step": 60970,
      "train_speed(iter/s)": 1.638286
    },
    {
      "acc": 0.65419989,
      "epoch": 1.5468036529680367,
      "grad_norm": 5.375,
      "learning_rate": 1.3393291186430852e-06,
      "loss": 1.61325874,
      "memory(GiB)": 117.38,
      "step": 60975,
      "train_speed(iter/s)": 1.638299
    },
    {
      "acc": 0.64392223,
      "epoch": 1.5469304921359717,
      "grad_norm": 10.625,
      "learning_rate": 1.338614918185548e-06,
      "loss": 1.6487709,
      "memory(GiB)": 117.38,
      "step": 60980,
      "train_speed(iter/s)": 1.638313
    },
    {
      "acc": 0.66089287,
      "epoch": 1.5470573313039067,
      "grad_norm": 4.90625,
      "learning_rate": 1.3379008787720732e-06,
      "loss": 1.59338989,
      "memory(GiB)": 117.38,
      "step": 60985,
      "train_speed(iter/s)": 1.638326
    },
    {
      "acc": 0.66049089,
      "epoch": 1.5471841704718416,
      "grad_norm": 6.15625,
      "learning_rate": 1.3371870004340681e-06,
      "loss": 1.61071091,
      "memory(GiB)": 117.38,
      "step": 60990,
      "train_speed(iter/s)": 1.638341
    },
    {
      "acc": 0.66481338,
      "epoch": 1.5473110096397766,
      "grad_norm": 5.90625,
      "learning_rate": 1.3364732832029315e-06,
      "loss": 1.57785244,
      "memory(GiB)": 117.38,
      "step": 60995,
      "train_speed(iter/s)": 1.638355
    },
    {
      "acc": 0.66319914,
      "epoch": 1.5474378488077118,
      "grad_norm": 5.6875,
      "learning_rate": 1.335759727110057e-06,
      "loss": 1.56670542,
      "memory(GiB)": 117.38,
      "step": 61000,
      "train_speed(iter/s)": 1.63837
    },
    {
      "epoch": 1.5474378488077118,
      "eval_acc": 0.6462568157090739,
      "eval_loss": 1.5732755661010742,
      "eval_runtime": 58.4487,
      "eval_samples_per_second": 108.985,
      "eval_steps_per_second": 27.255,
      "step": 61000
    },
    {
      "acc": 0.65179029,
      "epoch": 1.5475646879756468,
      "grad_norm": 6.8125,
      "learning_rate": 1.335046332186829e-06,
      "loss": 1.62582741,
      "memory(GiB)": 117.38,
      "step": 61005,
      "train_speed(iter/s)": 1.635626
    },
    {
      "acc": 0.66695957,
      "epoch": 1.547691527143582,
      "grad_norm": 5.8125,
      "learning_rate": 1.3343330984646262e-06,
      "loss": 1.54848003,
      "memory(GiB)": 117.38,
      "step": 61010,
      "train_speed(iter/s)": 1.63564
    },
    {
      "acc": 0.67391047,
      "epoch": 1.547818366311517,
      "grad_norm": 5.09375,
      "learning_rate": 1.33362002597482e-06,
      "loss": 1.59511576,
      "memory(GiB)": 117.38,
      "step": 61015,
      "train_speed(iter/s)": 1.635653
    },
    {
      "acc": 0.64972095,
      "epoch": 1.547945205479452,
      "grad_norm": 5.53125,
      "learning_rate": 1.3329071147487743e-06,
      "loss": 1.62496414,
      "memory(GiB)": 117.38,
      "step": 61020,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.65210514,
      "epoch": 1.548072044647387,
      "grad_norm": 5.75,
      "learning_rate": 1.3321943648178442e-06,
      "loss": 1.5740551,
      "memory(GiB)": 117.38,
      "step": 61025,
      "train_speed(iter/s)": 1.63568
    },
    {
      "acc": 0.64864464,
      "epoch": 1.5481988838153222,
      "grad_norm": 5.625,
      "learning_rate": 1.3314817762133848e-06,
      "loss": 1.63391151,
      "memory(GiB)": 117.38,
      "step": 61030,
      "train_speed(iter/s)": 1.635692
    },
    {
      "acc": 0.66745377,
      "epoch": 1.5483257229832572,
      "grad_norm": 5.84375,
      "learning_rate": 1.330769348966734e-06,
      "loss": 1.61501598,
      "memory(GiB)": 117.38,
      "step": 61035,
      "train_speed(iter/s)": 1.635705
    },
    {
      "acc": 0.65837078,
      "epoch": 1.5484525621511924,
      "grad_norm": 5.8125,
      "learning_rate": 1.3300570831092292e-06,
      "loss": 1.61278992,
      "memory(GiB)": 117.38,
      "step": 61040,
      "train_speed(iter/s)": 1.635718
    },
    {
      "acc": 0.67180367,
      "epoch": 1.5485794013191274,
      "grad_norm": 8.125,
      "learning_rate": 1.3293449786721973e-06,
      "loss": 1.49866285,
      "memory(GiB)": 117.38,
      "step": 61045,
      "train_speed(iter/s)": 1.635732
    },
    {
      "acc": 0.65392499,
      "epoch": 1.5487062404870624,
      "grad_norm": 6.03125,
      "learning_rate": 1.3286330356869648e-06,
      "loss": 1.63412418,
      "memory(GiB)": 117.38,
      "step": 61050,
      "train_speed(iter/s)": 1.635746
    },
    {
      "acc": 0.64719868,
      "epoch": 1.5488330796549974,
      "grad_norm": 6.28125,
      "learning_rate": 1.3279212541848413e-06,
      "loss": 1.63076973,
      "memory(GiB)": 117.38,
      "step": 61055,
      "train_speed(iter/s)": 1.63576
    },
    {
      "acc": 0.64747467,
      "epoch": 1.5489599188229324,
      "grad_norm": 7.25,
      "learning_rate": 1.3272096341971342e-06,
      "loss": 1.60763969,
      "memory(GiB)": 117.38,
      "step": 61060,
      "train_speed(iter/s)": 1.635774
    },
    {
      "acc": 0.65634375,
      "epoch": 1.5490867579908676,
      "grad_norm": 7.59375,
      "learning_rate": 1.326498175755147e-06,
      "loss": 1.64439659,
      "memory(GiB)": 117.38,
      "step": 61065,
      "train_speed(iter/s)": 1.635788
    },
    {
      "acc": 0.66577773,
      "epoch": 1.5492135971588028,
      "grad_norm": 7.71875,
      "learning_rate": 1.3257868788901722e-06,
      "loss": 1.61822376,
      "memory(GiB)": 117.38,
      "step": 61070,
      "train_speed(iter/s)": 1.635801
    },
    {
      "acc": 0.65768957,
      "epoch": 1.5493404363267378,
      "grad_norm": 5.3125,
      "learning_rate": 1.3250757436334932e-06,
      "loss": 1.57822151,
      "memory(GiB)": 117.38,
      "step": 61075,
      "train_speed(iter/s)": 1.635814
    },
    {
      "acc": 0.66250062,
      "epoch": 1.5494672754946728,
      "grad_norm": 6.4375,
      "learning_rate": 1.3243647700163887e-06,
      "loss": 1.56040421,
      "memory(GiB)": 117.38,
      "step": 61080,
      "train_speed(iter/s)": 1.635829
    },
    {
      "acc": 0.65652628,
      "epoch": 1.5495941146626078,
      "grad_norm": 5.28125,
      "learning_rate": 1.323653958070134e-06,
      "loss": 1.61688766,
      "memory(GiB)": 117.38,
      "step": 61085,
      "train_speed(iter/s)": 1.635841
    },
    {
      "acc": 0.65840831,
      "epoch": 1.5497209538305428,
      "grad_norm": 6.90625,
      "learning_rate": 1.3229433078259928e-06,
      "loss": 1.57728958,
      "memory(GiB)": 117.38,
      "step": 61090,
      "train_speed(iter/s)": 1.635856
    },
    {
      "acc": 0.64791031,
      "epoch": 1.549847792998478,
      "grad_norm": 5.125,
      "learning_rate": 1.3222328193152195e-06,
      "loss": 1.56909704,
      "memory(GiB)": 117.38,
      "step": 61095,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.66923327,
      "epoch": 1.549974632166413,
      "grad_norm": 5.84375,
      "learning_rate": 1.3215224925690683e-06,
      "loss": 1.63195572,
      "memory(GiB)": 117.38,
      "step": 61100,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.67461915,
      "epoch": 1.5501014713343482,
      "grad_norm": 5.15625,
      "learning_rate": 1.3208123276187807e-06,
      "loss": 1.54255314,
      "memory(GiB)": 117.38,
      "step": 61105,
      "train_speed(iter/s)": 1.635898
    },
    {
      "acc": 0.65144835,
      "epoch": 1.5502283105022832,
      "grad_norm": 4.9375,
      "learning_rate": 1.3201023244955952e-06,
      "loss": 1.61530056,
      "memory(GiB)": 117.38,
      "step": 61110,
      "train_speed(iter/s)": 1.63591
    },
    {
      "acc": 0.66974187,
      "epoch": 1.5503551496702181,
      "grad_norm": 4.9375,
      "learning_rate": 1.319392483230736e-06,
      "loss": 1.56306629,
      "memory(GiB)": 117.38,
      "step": 61115,
      "train_speed(iter/s)": 1.635925
    },
    {
      "acc": 0.64651308,
      "epoch": 1.5504819888381531,
      "grad_norm": 5.625,
      "learning_rate": 1.318682803855429e-06,
      "loss": 1.67715492,
      "memory(GiB)": 117.38,
      "step": 61120,
      "train_speed(iter/s)": 1.635939
    },
    {
      "acc": 0.65884929,
      "epoch": 1.5506088280060881,
      "grad_norm": 6.375,
      "learning_rate": 1.3179732864008888e-06,
      "loss": 1.62272549,
      "memory(GiB)": 117.38,
      "step": 61125,
      "train_speed(iter/s)": 1.635953
    },
    {
      "acc": 0.6555727,
      "epoch": 1.5507356671740233,
      "grad_norm": 5.96875,
      "learning_rate": 1.3172639308983226e-06,
      "loss": 1.5814868,
      "memory(GiB)": 117.38,
      "step": 61130,
      "train_speed(iter/s)": 1.635967
    },
    {
      "acc": 0.6675652,
      "epoch": 1.5508625063419585,
      "grad_norm": 5.0,
      "learning_rate": 1.3165547373789306e-06,
      "loss": 1.55184765,
      "memory(GiB)": 117.38,
      "step": 61135,
      "train_speed(iter/s)": 1.63598
    },
    {
      "acc": 0.65238214,
      "epoch": 1.5509893455098935,
      "grad_norm": 5.75,
      "learning_rate": 1.3158457058739066e-06,
      "loss": 1.63189926,
      "memory(GiB)": 117.38,
      "step": 61140,
      "train_speed(iter/s)": 1.635994
    },
    {
      "acc": 0.66230302,
      "epoch": 1.5511161846778285,
      "grad_norm": 5.9375,
      "learning_rate": 1.3151368364144373e-06,
      "loss": 1.64738388,
      "memory(GiB)": 117.38,
      "step": 61145,
      "train_speed(iter/s)": 1.636009
    },
    {
      "acc": 0.65345531,
      "epoch": 1.5512430238457635,
      "grad_norm": 5.96875,
      "learning_rate": 1.3144281290317012e-06,
      "loss": 1.60982533,
      "memory(GiB)": 117.38,
      "step": 61150,
      "train_speed(iter/s)": 1.636023
    },
    {
      "acc": 0.66240487,
      "epoch": 1.5513698630136985,
      "grad_norm": 6.125,
      "learning_rate": 1.3137195837568716e-06,
      "loss": 1.51676483,
      "memory(GiB)": 117.38,
      "step": 61155,
      "train_speed(iter/s)": 1.636036
    },
    {
      "acc": 0.66346321,
      "epoch": 1.5514967021816337,
      "grad_norm": 5.71875,
      "learning_rate": 1.313011200621112e-06,
      "loss": 1.57237358,
      "memory(GiB)": 117.38,
      "step": 61160,
      "train_speed(iter/s)": 1.636051
    },
    {
      "acc": 0.65693192,
      "epoch": 1.5516235413495687,
      "grad_norm": 4.375,
      "learning_rate": 1.312302979655582e-06,
      "loss": 1.58477821,
      "memory(GiB)": 117.38,
      "step": 61165,
      "train_speed(iter/s)": 1.636065
    },
    {
      "acc": 0.65312538,
      "epoch": 1.551750380517504,
      "grad_norm": 6.40625,
      "learning_rate": 1.3115949208914302e-06,
      "loss": 1.59472466,
      "memory(GiB)": 117.38,
      "step": 61170,
      "train_speed(iter/s)": 1.636079
    },
    {
      "acc": 0.65050321,
      "epoch": 1.551877219685439,
      "grad_norm": 5.40625,
      "learning_rate": 1.3108870243598022e-06,
      "loss": 1.61206379,
      "memory(GiB)": 117.38,
      "step": 61175,
      "train_speed(iter/s)": 1.636093
    },
    {
      "acc": 0.65264826,
      "epoch": 1.552004058853374,
      "grad_norm": 5.75,
      "learning_rate": 1.310179290091833e-06,
      "loss": 1.65162907,
      "memory(GiB)": 117.38,
      "step": 61180,
      "train_speed(iter/s)": 1.636107
    },
    {
      "acc": 0.66704984,
      "epoch": 1.5521308980213089,
      "grad_norm": 5.90625,
      "learning_rate": 1.3094717181186518e-06,
      "loss": 1.55495434,
      "memory(GiB)": 117.38,
      "step": 61185,
      "train_speed(iter/s)": 1.636122
    },
    {
      "acc": 0.64566684,
      "epoch": 1.552257737189244,
      "grad_norm": 7.84375,
      "learning_rate": 1.3087643084713836e-06,
      "loss": 1.59781446,
      "memory(GiB)": 117.38,
      "step": 61190,
      "train_speed(iter/s)": 1.636137
    },
    {
      "acc": 0.66397443,
      "epoch": 1.552384576357179,
      "grad_norm": 6.53125,
      "learning_rate": 1.30805706118114e-06,
      "loss": 1.58068399,
      "memory(GiB)": 117.38,
      "step": 61195,
      "train_speed(iter/s)": 1.636151
    },
    {
      "acc": 0.65234904,
      "epoch": 1.5525114155251143,
      "grad_norm": 6.21875,
      "learning_rate": 1.3073499762790287e-06,
      "loss": 1.62983627,
      "memory(GiB)": 117.38,
      "step": 61200,
      "train_speed(iter/s)": 1.636165
    },
    {
      "acc": 0.65697637,
      "epoch": 1.5526382546930493,
      "grad_norm": 5.03125,
      "learning_rate": 1.306643053796154e-06,
      "loss": 1.59063282,
      "memory(GiB)": 117.38,
      "step": 61205,
      "train_speed(iter/s)": 1.63618
    },
    {
      "acc": 0.66361923,
      "epoch": 1.5527650938609843,
      "grad_norm": 7.1875,
      "learning_rate": 1.3059362937636084e-06,
      "loss": 1.61873455,
      "memory(GiB)": 117.38,
      "step": 61210,
      "train_speed(iter/s)": 1.636195
    },
    {
      "acc": 0.66971402,
      "epoch": 1.5528919330289193,
      "grad_norm": 5.875,
      "learning_rate": 1.3052296962124756e-06,
      "loss": 1.52288589,
      "memory(GiB)": 117.38,
      "step": 61215,
      "train_speed(iter/s)": 1.636208
    },
    {
      "acc": 0.6382885,
      "epoch": 1.5530187721968542,
      "grad_norm": 6.125,
      "learning_rate": 1.3045232611738357e-06,
      "loss": 1.63298988,
      "memory(GiB)": 117.38,
      "step": 61220,
      "train_speed(iter/s)": 1.636222
    },
    {
      "acc": 0.65036759,
      "epoch": 1.5531456113647895,
      "grad_norm": 5.84375,
      "learning_rate": 1.3038169886787632e-06,
      "loss": 1.60665169,
      "memory(GiB)": 117.38,
      "step": 61225,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.68696127,
      "epoch": 1.5532724505327247,
      "grad_norm": 8.875,
      "learning_rate": 1.3031108787583235e-06,
      "loss": 1.53828192,
      "memory(GiB)": 117.38,
      "step": 61230,
      "train_speed(iter/s)": 1.63625
    },
    {
      "acc": 0.64480381,
      "epoch": 1.5533992897006597,
      "grad_norm": 5.375,
      "learning_rate": 1.3024049314435694e-06,
      "loss": 1.60862217,
      "memory(GiB)": 117.38,
      "step": 61235,
      "train_speed(iter/s)": 1.636264
    },
    {
      "acc": 0.64851785,
      "epoch": 1.5535261288685946,
      "grad_norm": 5.75,
      "learning_rate": 1.301699146765557e-06,
      "loss": 1.62237988,
      "memory(GiB)": 117.38,
      "step": 61240,
      "train_speed(iter/s)": 1.636278
    },
    {
      "acc": 0.67204733,
      "epoch": 1.5536529680365296,
      "grad_norm": 5.8125,
      "learning_rate": 1.3009935247553274e-06,
      "loss": 1.60659752,
      "memory(GiB)": 117.38,
      "step": 61245,
      "train_speed(iter/s)": 1.636292
    },
    {
      "acc": 0.64988775,
      "epoch": 1.5537798072044646,
      "grad_norm": 5.65625,
      "learning_rate": 1.3002880654439192e-06,
      "loss": 1.67673302,
      "memory(GiB)": 117.38,
      "step": 61250,
      "train_speed(iter/s)": 1.636307
    },
    {
      "acc": 0.64734554,
      "epoch": 1.5539066463723998,
      "grad_norm": 4.90625,
      "learning_rate": 1.2995827688623568e-06,
      "loss": 1.66474113,
      "memory(GiB)": 117.38,
      "step": 61255,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.66390038,
      "epoch": 1.5540334855403348,
      "grad_norm": 5.28125,
      "learning_rate": 1.298877635041667e-06,
      "loss": 1.51859856,
      "memory(GiB)": 117.38,
      "step": 61260,
      "train_speed(iter/s)": 1.636335
    },
    {
      "acc": 0.64104948,
      "epoch": 1.55416032470827,
      "grad_norm": 5.125,
      "learning_rate": 1.2981726640128633e-06,
      "loss": 1.63175583,
      "memory(GiB)": 117.38,
      "step": 61265,
      "train_speed(iter/s)": 1.636349
    },
    {
      "acc": 0.65757418,
      "epoch": 1.554287163876205,
      "grad_norm": 4.71875,
      "learning_rate": 1.297467855806953e-06,
      "loss": 1.53976688,
      "memory(GiB)": 117.38,
      "step": 61270,
      "train_speed(iter/s)": 1.636363
    },
    {
      "acc": 0.66194124,
      "epoch": 1.55441400304414,
      "grad_norm": 9.875,
      "learning_rate": 1.2967632104549371e-06,
      "loss": 1.62684193,
      "memory(GiB)": 117.38,
      "step": 61275,
      "train_speed(iter/s)": 1.636377
    },
    {
      "acc": 0.65334568,
      "epoch": 1.554540842212075,
      "grad_norm": 5.28125,
      "learning_rate": 1.296058727987809e-06,
      "loss": 1.58629646,
      "memory(GiB)": 117.38,
      "step": 61280,
      "train_speed(iter/s)": 1.636392
    },
    {
      "acc": 0.65941591,
      "epoch": 1.55466768138001,
      "grad_norm": 4.96875,
      "learning_rate": 1.295354408436555e-06,
      "loss": 1.64778557,
      "memory(GiB)": 117.38,
      "step": 61285,
      "train_speed(iter/s)": 1.636406
    },
    {
      "acc": 0.66195574,
      "epoch": 1.5547945205479452,
      "grad_norm": 6.5625,
      "learning_rate": 1.294650251832154e-06,
      "loss": 1.56548967,
      "memory(GiB)": 117.38,
      "step": 61290,
      "train_speed(iter/s)": 1.63642
    },
    {
      "acc": 0.6524641,
      "epoch": 1.5549213597158804,
      "grad_norm": 6.25,
      "learning_rate": 1.2939462582055784e-06,
      "loss": 1.5599987,
      "memory(GiB)": 117.38,
      "step": 61295,
      "train_speed(iter/s)": 1.636434
    },
    {
      "acc": 0.66586266,
      "epoch": 1.5550481988838154,
      "grad_norm": 5.625,
      "learning_rate": 1.2932424275877926e-06,
      "loss": 1.5513875,
      "memory(GiB)": 117.38,
      "step": 61300,
      "train_speed(iter/s)": 1.636448
    },
    {
      "acc": 0.66176119,
      "epoch": 1.5551750380517504,
      "grad_norm": 6.59375,
      "learning_rate": 1.2925387600097543e-06,
      "loss": 1.57261553,
      "memory(GiB)": 117.38,
      "step": 61305,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.64702597,
      "epoch": 1.5553018772196854,
      "grad_norm": 4.375,
      "learning_rate": 1.291835255502414e-06,
      "loss": 1.562537,
      "memory(GiB)": 117.38,
      "step": 61310,
      "train_speed(iter/s)": 1.636477
    },
    {
      "acc": 0.64694924,
      "epoch": 1.5554287163876204,
      "grad_norm": 4.84375,
      "learning_rate": 1.2911319140967148e-06,
      "loss": 1.61142235,
      "memory(GiB)": 117.38,
      "step": 61315,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.64774475,
      "epoch": 1.5555555555555556,
      "grad_norm": 5.375,
      "learning_rate": 1.290428735823593e-06,
      "loss": 1.64048691,
      "memory(GiB)": 117.38,
      "step": 61320,
      "train_speed(iter/s)": 1.636505
    },
    {
      "acc": 0.64523745,
      "epoch": 1.5556823947234906,
      "grad_norm": 5.5625,
      "learning_rate": 1.2897257207139758e-06,
      "loss": 1.6832386,
      "memory(GiB)": 117.38,
      "step": 61325,
      "train_speed(iter/s)": 1.636519
    },
    {
      "acc": 0.65186181,
      "epoch": 1.5558092338914258,
      "grad_norm": 5.40625,
      "learning_rate": 1.28902286879879e-06,
      "loss": 1.58911114,
      "memory(GiB)": 117.38,
      "step": 61330,
      "train_speed(iter/s)": 1.636534
    },
    {
      "acc": 0.64767332,
      "epoch": 1.5559360730593608,
      "grad_norm": 6.125,
      "learning_rate": 1.2883201801089445e-06,
      "loss": 1.6280632,
      "memory(GiB)": 117.38,
      "step": 61335,
      "train_speed(iter/s)": 1.636548
    },
    {
      "acc": 0.64430685,
      "epoch": 1.5560629122272958,
      "grad_norm": 5.5625,
      "learning_rate": 1.2876176546753494e-06,
      "loss": 1.66660042,
      "memory(GiB)": 117.38,
      "step": 61340,
      "train_speed(iter/s)": 1.636563
    },
    {
      "acc": 0.65408363,
      "epoch": 1.5561897513952307,
      "grad_norm": 5.9375,
      "learning_rate": 1.286915292528903e-06,
      "loss": 1.65121346,
      "memory(GiB)": 117.38,
      "step": 61345,
      "train_speed(iter/s)": 1.636577
    },
    {
      "acc": 0.65273867,
      "epoch": 1.556316590563166,
      "grad_norm": 5.96875,
      "learning_rate": 1.286213093700503e-06,
      "loss": 1.57195454,
      "memory(GiB)": 117.38,
      "step": 61350,
      "train_speed(iter/s)": 1.636592
    },
    {
      "acc": 0.66110177,
      "epoch": 1.556443429731101,
      "grad_norm": 6.6875,
      "learning_rate": 1.28551105822103e-06,
      "loss": 1.551651,
      "memory(GiB)": 117.38,
      "step": 61355,
      "train_speed(iter/s)": 1.636608
    },
    {
      "acc": 0.65804434,
      "epoch": 1.5565702688990362,
      "grad_norm": 6.65625,
      "learning_rate": 1.2848091861213636e-06,
      "loss": 1.59232464,
      "memory(GiB)": 117.38,
      "step": 61360,
      "train_speed(iter/s)": 1.636623
    },
    {
      "acc": 0.64280934,
      "epoch": 1.5566971080669711,
      "grad_norm": 5.71875,
      "learning_rate": 1.2841074774323775e-06,
      "loss": 1.69599457,
      "memory(GiB)": 117.38,
      "step": 61365,
      "train_speed(iter/s)": 1.636636
    },
    {
      "acc": 0.6431088,
      "epoch": 1.5568239472349061,
      "grad_norm": 5.4375,
      "learning_rate": 1.2834059321849363e-06,
      "loss": 1.64311657,
      "memory(GiB)": 117.38,
      "step": 61370,
      "train_speed(iter/s)": 1.636651
    },
    {
      "acc": 0.68124781,
      "epoch": 1.5569507864028411,
      "grad_norm": 7.09375,
      "learning_rate": 1.2827045504098928e-06,
      "loss": 1.5941206,
      "memory(GiB)": 117.38,
      "step": 61375,
      "train_speed(iter/s)": 1.636665
    },
    {
      "acc": 0.65773225,
      "epoch": 1.5570776255707761,
      "grad_norm": 6.0,
      "learning_rate": 1.2820033321381009e-06,
      "loss": 1.56766624,
      "memory(GiB)": 117.38,
      "step": 61380,
      "train_speed(iter/s)": 1.636678
    },
    {
      "acc": 0.67315111,
      "epoch": 1.5572044647387113,
      "grad_norm": 5.8125,
      "learning_rate": 1.2813022774004024e-06,
      "loss": 1.54263153,
      "memory(GiB)": 117.38,
      "step": 61385,
      "train_speed(iter/s)": 1.636691
    },
    {
      "acc": 0.65441117,
      "epoch": 1.5573313039066465,
      "grad_norm": 5.5625,
      "learning_rate": 1.280601386227634e-06,
      "loss": 1.62208328,
      "memory(GiB)": 117.38,
      "step": 61390,
      "train_speed(iter/s)": 1.636706
    },
    {
      "acc": 0.65167103,
      "epoch": 1.5574581430745815,
      "grad_norm": 5.625,
      "learning_rate": 1.279900658650619e-06,
      "loss": 1.64166241,
      "memory(GiB)": 117.38,
      "step": 61395,
      "train_speed(iter/s)": 1.63672
    },
    {
      "acc": 0.64311442,
      "epoch": 1.5575849822425165,
      "grad_norm": 7.0,
      "learning_rate": 1.2792000947001842e-06,
      "loss": 1.65377274,
      "memory(GiB)": 117.38,
      "step": 61400,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.66006284,
      "epoch": 1.5577118214104515,
      "grad_norm": 5.125,
      "learning_rate": 1.2784996944071415e-06,
      "loss": 1.5988925,
      "memory(GiB)": 117.38,
      "step": 61405,
      "train_speed(iter/s)": 1.636747
    },
    {
      "acc": 0.65563731,
      "epoch": 1.5578386605783865,
      "grad_norm": 5.84375,
      "learning_rate": 1.2777994578022972e-06,
      "loss": 1.66416016,
      "memory(GiB)": 117.38,
      "step": 61410,
      "train_speed(iter/s)": 1.636761
    },
    {
      "acc": 0.65468316,
      "epoch": 1.5579654997463217,
      "grad_norm": 6.1875,
      "learning_rate": 1.2770993849164514e-06,
      "loss": 1.63350372,
      "memory(GiB)": 117.38,
      "step": 61415,
      "train_speed(iter/s)": 1.636775
    },
    {
      "acc": 0.64138761,
      "epoch": 1.5580923389142567,
      "grad_norm": 6.125,
      "learning_rate": 1.276399475780396e-06,
      "loss": 1.7182621,
      "memory(GiB)": 117.38,
      "step": 61420,
      "train_speed(iter/s)": 1.636789
    },
    {
      "acc": 0.66096191,
      "epoch": 1.558219178082192,
      "grad_norm": 5.09375,
      "learning_rate": 1.2756997304249164e-06,
      "loss": 1.5984087,
      "memory(GiB)": 117.38,
      "step": 61425,
      "train_speed(iter/s)": 1.636802
    },
    {
      "acc": 0.6529911,
      "epoch": 1.558346017250127,
      "grad_norm": 5.78125,
      "learning_rate": 1.2750001488807906e-06,
      "loss": 1.56206484,
      "memory(GiB)": 117.38,
      "step": 61430,
      "train_speed(iter/s)": 1.636816
    },
    {
      "acc": 0.66399221,
      "epoch": 1.5584728564180619,
      "grad_norm": 6.53125,
      "learning_rate": 1.2743007311787892e-06,
      "loss": 1.61260223,
      "memory(GiB)": 117.38,
      "step": 61435,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.66387944,
      "epoch": 1.5585996955859969,
      "grad_norm": 6.34375,
      "learning_rate": 1.2736014773496757e-06,
      "loss": 1.54993162,
      "memory(GiB)": 117.38,
      "step": 61440,
      "train_speed(iter/s)": 1.636846
    },
    {
      "acc": 0.66723323,
      "epoch": 1.5587265347539319,
      "grad_norm": 6.125,
      "learning_rate": 1.2729023874242064e-06,
      "loss": 1.52294521,
      "memory(GiB)": 117.38,
      "step": 61445,
      "train_speed(iter/s)": 1.636859
    },
    {
      "acc": 0.64380026,
      "epoch": 1.558853373921867,
      "grad_norm": 6.59375,
      "learning_rate": 1.2722034614331303e-06,
      "loss": 1.71321545,
      "memory(GiB)": 117.38,
      "step": 61450,
      "train_speed(iter/s)": 1.636874
    },
    {
      "acc": 0.64076867,
      "epoch": 1.5589802130898023,
      "grad_norm": 6.0,
      "learning_rate": 1.2715046994071889e-06,
      "loss": 1.66914959,
      "memory(GiB)": 117.38,
      "step": 61455,
      "train_speed(iter/s)": 1.636888
    },
    {
      "acc": 0.64837737,
      "epoch": 1.5591070522577373,
      "grad_norm": 5.78125,
      "learning_rate": 1.2708061013771179e-06,
      "loss": 1.60034485,
      "memory(GiB)": 117.38,
      "step": 61460,
      "train_speed(iter/s)": 1.636902
    },
    {
      "acc": 0.65644279,
      "epoch": 1.5592338914256723,
      "grad_norm": 4.75,
      "learning_rate": 1.2701076673736428e-06,
      "loss": 1.56525068,
      "memory(GiB)": 117.38,
      "step": 61465,
      "train_speed(iter/s)": 1.636915
    },
    {
      "acc": 0.65083151,
      "epoch": 1.5593607305936072,
      "grad_norm": 5.6875,
      "learning_rate": 1.269409397427488e-06,
      "loss": 1.62524376,
      "memory(GiB)": 117.38,
      "step": 61470,
      "train_speed(iter/s)": 1.636928
    },
    {
      "acc": 0.66408896,
      "epoch": 1.5594875697615422,
      "grad_norm": 5.875,
      "learning_rate": 1.2687112915693622e-06,
      "loss": 1.52112322,
      "memory(GiB)": 117.38,
      "step": 61475,
      "train_speed(iter/s)": 1.636941
    },
    {
      "acc": 0.64434872,
      "epoch": 1.5596144089294774,
      "grad_norm": 5.34375,
      "learning_rate": 1.2680133498299729e-06,
      "loss": 1.68178368,
      "memory(GiB)": 117.38,
      "step": 61480,
      "train_speed(iter/s)": 1.636955
    },
    {
      "acc": 0.6533885,
      "epoch": 1.5597412480974124,
      "grad_norm": 5.4375,
      "learning_rate": 1.2673155722400177e-06,
      "loss": 1.58763123,
      "memory(GiB)": 117.38,
      "step": 61485,
      "train_speed(iter/s)": 1.63697
    },
    {
      "acc": 0.64297004,
      "epoch": 1.5598680872653476,
      "grad_norm": 5.6875,
      "learning_rate": 1.2666179588301908e-06,
      "loss": 1.69776306,
      "memory(GiB)": 117.38,
      "step": 61490,
      "train_speed(iter/s)": 1.636984
    },
    {
      "acc": 0.66218548,
      "epoch": 1.5599949264332826,
      "grad_norm": 7.28125,
      "learning_rate": 1.2659205096311738e-06,
      "loss": 1.59236755,
      "memory(GiB)": 117.38,
      "step": 61495,
      "train_speed(iter/s)": 1.636999
    },
    {
      "acc": 0.64790192,
      "epoch": 1.5601217656012176,
      "grad_norm": 5.875,
      "learning_rate": 1.2652232246736423e-06,
      "loss": 1.60135746,
      "memory(GiB)": 117.38,
      "step": 61500,
      "train_speed(iter/s)": 1.637013
    },
    {
      "acc": 0.6695004,
      "epoch": 1.5602486047691526,
      "grad_norm": 5.78125,
      "learning_rate": 1.2645261039882694e-06,
      "loss": 1.65159931,
      "memory(GiB)": 117.38,
      "step": 61505,
      "train_speed(iter/s)": 1.637027
    },
    {
      "acc": 0.65085125,
      "epoch": 1.5603754439370878,
      "grad_norm": 6.03125,
      "learning_rate": 1.263829147605718e-06,
      "loss": 1.62512245,
      "memory(GiB)": 117.38,
      "step": 61510,
      "train_speed(iter/s)": 1.637041
    },
    {
      "acc": 0.65961914,
      "epoch": 1.5605022831050228,
      "grad_norm": 6.34375,
      "learning_rate": 1.26313235555664e-06,
      "loss": 1.61042614,
      "memory(GiB)": 117.38,
      "step": 61515,
      "train_speed(iter/s)": 1.637055
    },
    {
      "acc": 0.6738863,
      "epoch": 1.560629122272958,
      "grad_norm": 5.625,
      "learning_rate": 1.2624357278716832e-06,
      "loss": 1.5503334,
      "memory(GiB)": 117.38,
      "step": 61520,
      "train_speed(iter/s)": 1.63707
    },
    {
      "acc": 0.64084396,
      "epoch": 1.560755961440893,
      "grad_norm": 5.46875,
      "learning_rate": 1.2617392645814913e-06,
      "loss": 1.5916008,
      "memory(GiB)": 117.38,
      "step": 61525,
      "train_speed(iter/s)": 1.637083
    },
    {
      "acc": 0.6529985,
      "epoch": 1.560882800608828,
      "grad_norm": 5.4375,
      "learning_rate": 1.2610429657166983e-06,
      "loss": 1.67311039,
      "memory(GiB)": 117.38,
      "step": 61530,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.67050619,
      "epoch": 1.561009639776763,
      "grad_norm": 5.21875,
      "learning_rate": 1.2603468313079265e-06,
      "loss": 1.52844963,
      "memory(GiB)": 117.38,
      "step": 61535,
      "train_speed(iter/s)": 1.637112
    },
    {
      "acc": 0.66273241,
      "epoch": 1.561136478944698,
      "grad_norm": 5.34375,
      "learning_rate": 1.2596508613857982e-06,
      "loss": 1.54820318,
      "memory(GiB)": 117.38,
      "step": 61540,
      "train_speed(iter/s)": 1.637126
    },
    {
      "acc": 0.64780221,
      "epoch": 1.5612633181126332,
      "grad_norm": 5.34375,
      "learning_rate": 1.258955055980925e-06,
      "loss": 1.59088593,
      "memory(GiB)": 117.38,
      "step": 61545,
      "train_speed(iter/s)": 1.63714
    },
    {
      "acc": 0.64865084,
      "epoch": 1.5613901572805684,
      "grad_norm": 5.125,
      "learning_rate": 1.258259415123911e-06,
      "loss": 1.60054665,
      "memory(GiB)": 117.38,
      "step": 61550,
      "train_speed(iter/s)": 1.637154
    },
    {
      "acc": 0.6744133,
      "epoch": 1.5615169964485034,
      "grad_norm": 5.125,
      "learning_rate": 1.2575639388453532e-06,
      "loss": 1.49141979,
      "memory(GiB)": 117.38,
      "step": 61555,
      "train_speed(iter/s)": 1.637169
    },
    {
      "acc": 0.65708084,
      "epoch": 1.5616438356164384,
      "grad_norm": 6.125,
      "learning_rate": 1.2568686271758423e-06,
      "loss": 1.55554161,
      "memory(GiB)": 117.38,
      "step": 61560,
      "train_speed(iter/s)": 1.637183
    },
    {
      "acc": 0.64792876,
      "epoch": 1.5617706747843734,
      "grad_norm": 5.21875,
      "learning_rate": 1.2561734801459612e-06,
      "loss": 1.64314842,
      "memory(GiB)": 117.38,
      "step": 61565,
      "train_speed(iter/s)": 1.637197
    },
    {
      "acc": 0.66256962,
      "epoch": 1.5618975139523084,
      "grad_norm": 6.0,
      "learning_rate": 1.2554784977862856e-06,
      "loss": 1.5920846,
      "memory(GiB)": 117.38,
      "step": 61570,
      "train_speed(iter/s)": 1.637212
    },
    {
      "acc": 0.65574789,
      "epoch": 1.5620243531202436,
      "grad_norm": 5.625,
      "learning_rate": 1.2547836801273833e-06,
      "loss": 1.60377159,
      "memory(GiB)": 117.38,
      "step": 61575,
      "train_speed(iter/s)": 1.637226
    },
    {
      "acc": 0.65416446,
      "epoch": 1.5621511922881786,
      "grad_norm": 5.875,
      "learning_rate": 1.2540890271998162e-06,
      "loss": 1.64567146,
      "memory(GiB)": 117.38,
      "step": 61580,
      "train_speed(iter/s)": 1.637241
    },
    {
      "acc": 0.64260716,
      "epoch": 1.5622780314561138,
      "grad_norm": 4.8125,
      "learning_rate": 1.2533945390341379e-06,
      "loss": 1.68209915,
      "memory(GiB)": 117.38,
      "step": 61585,
      "train_speed(iter/s)": 1.637254
    },
    {
      "acc": 0.65278797,
      "epoch": 1.5624048706240488,
      "grad_norm": 5.75,
      "learning_rate": 1.2527002156608946e-06,
      "loss": 1.610886,
      "memory(GiB)": 117.38,
      "step": 61590,
      "train_speed(iter/s)": 1.637269
    },
    {
      "acc": 0.66167889,
      "epoch": 1.5625317097919837,
      "grad_norm": 6.1875,
      "learning_rate": 1.2520060571106275e-06,
      "loss": 1.57451191,
      "memory(GiB)": 117.38,
      "step": 61595,
      "train_speed(iter/s)": 1.637284
    },
    {
      "acc": 0.64473963,
      "epoch": 1.5626585489599187,
      "grad_norm": 5.90625,
      "learning_rate": 1.2513120634138665e-06,
      "loss": 1.6647213,
      "memory(GiB)": 117.38,
      "step": 61600,
      "train_speed(iter/s)": 1.637297
    },
    {
      "acc": 0.65826559,
      "epoch": 1.5627853881278537,
      "grad_norm": 5.875,
      "learning_rate": 1.250618234601138e-06,
      "loss": 1.54982643,
      "memory(GiB)": 117.38,
      "step": 61605,
      "train_speed(iter/s)": 1.63731
    },
    {
      "acc": 0.66492996,
      "epoch": 1.562912227295789,
      "grad_norm": 6.4375,
      "learning_rate": 1.2499245707029595e-06,
      "loss": 1.54981127,
      "memory(GiB)": 117.38,
      "step": 61610,
      "train_speed(iter/s)": 1.637325
    },
    {
      "acc": 0.64957094,
      "epoch": 1.5630390664637241,
      "grad_norm": 6.0,
      "learning_rate": 1.2492310717498412e-06,
      "loss": 1.59807701,
      "memory(GiB)": 117.38,
      "step": 61615,
      "train_speed(iter/s)": 1.637339
    },
    {
      "acc": 0.67292881,
      "epoch": 1.5631659056316591,
      "grad_norm": 6.25,
      "learning_rate": 1.2485377377722863e-06,
      "loss": 1.60729065,
      "memory(GiB)": 117.38,
      "step": 61620,
      "train_speed(iter/s)": 1.637354
    },
    {
      "acc": 0.67249317,
      "epoch": 1.5632927447995941,
      "grad_norm": 10.375,
      "learning_rate": 1.2478445688007894e-06,
      "loss": 1.5835084,
      "memory(GiB)": 117.38,
      "step": 61625,
      "train_speed(iter/s)": 1.637368
    },
    {
      "acc": 0.66082482,
      "epoch": 1.5634195839675291,
      "grad_norm": 5.9375,
      "learning_rate": 1.2471515648658434e-06,
      "loss": 1.56413441,
      "memory(GiB)": 117.38,
      "step": 61630,
      "train_speed(iter/s)": 1.637383
    },
    {
      "acc": 0.6674428,
      "epoch": 1.563546423135464,
      "grad_norm": 6.1875,
      "learning_rate": 1.2464587259979254e-06,
      "loss": 1.58378563,
      "memory(GiB)": 117.38,
      "step": 61635,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.65258722,
      "epoch": 1.5636732623033993,
      "grad_norm": 9.125,
      "learning_rate": 1.2457660522275095e-06,
      "loss": 1.60452137,
      "memory(GiB)": 117.38,
      "step": 61640,
      "train_speed(iter/s)": 1.637412
    },
    {
      "acc": 0.65679646,
      "epoch": 1.5638001014713343,
      "grad_norm": 8.875,
      "learning_rate": 1.2450735435850654e-06,
      "loss": 1.57998753,
      "memory(GiB)": 117.38,
      "step": 61645,
      "train_speed(iter/s)": 1.637427
    },
    {
      "acc": 0.65105896,
      "epoch": 1.5639269406392695,
      "grad_norm": 6.375,
      "learning_rate": 1.244381200101053e-06,
      "loss": 1.63169594,
      "memory(GiB)": 117.38,
      "step": 61650,
      "train_speed(iter/s)": 1.637441
    },
    {
      "acc": 0.65601263,
      "epoch": 1.5640537798072045,
      "grad_norm": 5.6875,
      "learning_rate": 1.2436890218059217e-06,
      "loss": 1.6355278,
      "memory(GiB)": 117.38,
      "step": 61655,
      "train_speed(iter/s)": 1.637455
    },
    {
      "acc": 0.65831928,
      "epoch": 1.5641806189751395,
      "grad_norm": 5.4375,
      "learning_rate": 1.2429970087301163e-06,
      "loss": 1.57492008,
      "memory(GiB)": 117.38,
      "step": 61660,
      "train_speed(iter/s)": 1.637469
    },
    {
      "acc": 0.650877,
      "epoch": 1.5643074581430745,
      "grad_norm": 5.9375,
      "learning_rate": 1.2423051609040777e-06,
      "loss": 1.65746803,
      "memory(GiB)": 117.38,
      "step": 61665,
      "train_speed(iter/s)": 1.637483
    },
    {
      "acc": 0.64344234,
      "epoch": 1.5644342973110097,
      "grad_norm": 6.09375,
      "learning_rate": 1.2416134783582368e-06,
      "loss": 1.63480892,
      "memory(GiB)": 117.38,
      "step": 61670,
      "train_speed(iter/s)": 1.637497
    },
    {
      "acc": 0.65756078,
      "epoch": 1.5645611364789447,
      "grad_norm": 5.15625,
      "learning_rate": 1.2409219611230116e-06,
      "loss": 1.6118103,
      "memory(GiB)": 117.38,
      "step": 61675,
      "train_speed(iter/s)": 1.637511
    },
    {
      "acc": 0.66029873,
      "epoch": 1.56468797564688,
      "grad_norm": 5.65625,
      "learning_rate": 1.2402306092288236e-06,
      "loss": 1.56962442,
      "memory(GiB)": 117.38,
      "step": 61680,
      "train_speed(iter/s)": 1.637525
    },
    {
      "acc": 0.66273575,
      "epoch": 1.5648148148148149,
      "grad_norm": 6.625,
      "learning_rate": 1.2395394227060793e-06,
      "loss": 1.62700691,
      "memory(GiB)": 117.38,
      "step": 61685,
      "train_speed(iter/s)": 1.637539
    },
    {
      "acc": 0.65293407,
      "epoch": 1.5649416539827499,
      "grad_norm": 6.875,
      "learning_rate": 1.238848401585182e-06,
      "loss": 1.61073303,
      "memory(GiB)": 117.38,
      "step": 61690,
      "train_speed(iter/s)": 1.637553
    },
    {
      "acc": 0.64637957,
      "epoch": 1.5650684931506849,
      "grad_norm": 5.75,
      "learning_rate": 1.2381575458965218e-06,
      "loss": 1.56928968,
      "memory(GiB)": 117.38,
      "step": 61695,
      "train_speed(iter/s)": 1.637567
    },
    {
      "acc": 0.65601873,
      "epoch": 1.5651953323186198,
      "grad_norm": 5.96875,
      "learning_rate": 1.2374668556704888e-06,
      "loss": 1.61248283,
      "memory(GiB)": 117.38,
      "step": 61700,
      "train_speed(iter/s)": 1.637581
    },
    {
      "acc": 0.66395602,
      "epoch": 1.565322171486555,
      "grad_norm": 6.5625,
      "learning_rate": 1.2367763309374625e-06,
      "loss": 1.56700306,
      "memory(GiB)": 117.38,
      "step": 61705,
      "train_speed(iter/s)": 1.637596
    },
    {
      "acc": 0.66082716,
      "epoch": 1.5654490106544903,
      "grad_norm": 5.1875,
      "learning_rate": 1.2360859717278145e-06,
      "loss": 1.59073658,
      "memory(GiB)": 117.38,
      "step": 61710,
      "train_speed(iter/s)": 1.63761
    },
    {
      "acc": 0.64166784,
      "epoch": 1.5655758498224253,
      "grad_norm": 5.875,
      "learning_rate": 1.2353957780719106e-06,
      "loss": 1.60281105,
      "memory(GiB)": 117.38,
      "step": 61715,
      "train_speed(iter/s)": 1.637624
    },
    {
      "acc": 0.64489803,
      "epoch": 1.5657026889903602,
      "grad_norm": 5.90625,
      "learning_rate": 1.2347057500001075e-06,
      "loss": 1.69349117,
      "memory(GiB)": 117.38,
      "step": 61720,
      "train_speed(iter/s)": 1.637638
    },
    {
      "acc": 0.65718765,
      "epoch": 1.5658295281582952,
      "grad_norm": 5.15625,
      "learning_rate": 1.2340158875427566e-06,
      "loss": 1.59059925,
      "memory(GiB)": 117.38,
      "step": 61725,
      "train_speed(iter/s)": 1.637653
    },
    {
      "acc": 0.64590435,
      "epoch": 1.5659563673262302,
      "grad_norm": 6.40625,
      "learning_rate": 1.2333261907302013e-06,
      "loss": 1.58984556,
      "memory(GiB)": 117.38,
      "step": 61730,
      "train_speed(iter/s)": 1.637667
    },
    {
      "acc": 0.67564478,
      "epoch": 1.5660832064941654,
      "grad_norm": 5.3125,
      "learning_rate": 1.2326366595927763e-06,
      "loss": 1.56249037,
      "memory(GiB)": 117.38,
      "step": 61735,
      "train_speed(iter/s)": 1.637682
    },
    {
      "acc": 0.65448256,
      "epoch": 1.5662100456621004,
      "grad_norm": 5.0625,
      "learning_rate": 1.2319472941608118e-06,
      "loss": 1.6299675,
      "memory(GiB)": 117.38,
      "step": 61740,
      "train_speed(iter/s)": 1.637696
    },
    {
      "acc": 0.65279875,
      "epoch": 1.5663368848300356,
      "grad_norm": 6.21875,
      "learning_rate": 1.231258094464628e-06,
      "loss": 1.61363525,
      "memory(GiB)": 117.38,
      "step": 61745,
      "train_speed(iter/s)": 1.637711
    },
    {
      "acc": 0.6548912,
      "epoch": 1.5664637239979706,
      "grad_norm": 6.84375,
      "learning_rate": 1.23056906053454e-06,
      "loss": 1.62874794,
      "memory(GiB)": 117.38,
      "step": 61750,
      "train_speed(iter/s)": 1.637726
    },
    {
      "acc": 0.64261389,
      "epoch": 1.5665905631659056,
      "grad_norm": 5.9375,
      "learning_rate": 1.2298801924008535e-06,
      "loss": 1.65138493,
      "memory(GiB)": 117.38,
      "step": 61755,
      "train_speed(iter/s)": 1.63774
    },
    {
      "acc": 0.64777794,
      "epoch": 1.5667174023338406,
      "grad_norm": 5.03125,
      "learning_rate": 1.2291914900938685e-06,
      "loss": 1.60229836,
      "memory(GiB)": 117.38,
      "step": 61760,
      "train_speed(iter/s)": 1.637755
    },
    {
      "acc": 0.65367351,
      "epoch": 1.5668442415017756,
      "grad_norm": 6.4375,
      "learning_rate": 1.2285029536438759e-06,
      "loss": 1.62405052,
      "memory(GiB)": 117.38,
      "step": 61765,
      "train_speed(iter/s)": 1.63777
    },
    {
      "acc": 0.6409234,
      "epoch": 1.5669710806697108,
      "grad_norm": 4.6875,
      "learning_rate": 1.227814583081165e-06,
      "loss": 1.62026291,
      "memory(GiB)": 117.38,
      "step": 61770,
      "train_speed(iter/s)": 1.637784
    },
    {
      "acc": 0.65885797,
      "epoch": 1.567097919837646,
      "grad_norm": 8.25,
      "learning_rate": 1.2271263784360088e-06,
      "loss": 1.58672199,
      "memory(GiB)": 117.38,
      "step": 61775,
      "train_speed(iter/s)": 1.637798
    },
    {
      "acc": 0.65834951,
      "epoch": 1.567224759005581,
      "grad_norm": 5.75,
      "learning_rate": 1.2264383397386787e-06,
      "loss": 1.64564991,
      "memory(GiB)": 117.38,
      "step": 61780,
      "train_speed(iter/s)": 1.637813
    },
    {
      "acc": 0.66006937,
      "epoch": 1.567351598173516,
      "grad_norm": 6.625,
      "learning_rate": 1.225750467019437e-06,
      "loss": 1.62769165,
      "memory(GiB)": 117.38,
      "step": 61785,
      "train_speed(iter/s)": 1.637827
    },
    {
      "acc": 0.6515975,
      "epoch": 1.567478437341451,
      "grad_norm": 5.53125,
      "learning_rate": 1.2250627603085435e-06,
      "loss": 1.65951691,
      "memory(GiB)": 117.38,
      "step": 61790,
      "train_speed(iter/s)": 1.63784
    },
    {
      "acc": 0.66303587,
      "epoch": 1.567605276509386,
      "grad_norm": 6.84375,
      "learning_rate": 1.2243752196362423e-06,
      "loss": 1.61594296,
      "memory(GiB)": 117.38,
      "step": 61795,
      "train_speed(iter/s)": 1.637854
    },
    {
      "acc": 0.65061016,
      "epoch": 1.5677321156773212,
      "grad_norm": 5.28125,
      "learning_rate": 1.2236878450327743e-06,
      "loss": 1.63613167,
      "memory(GiB)": 117.38,
      "step": 61800,
      "train_speed(iter/s)": 1.637867
    },
    {
      "acc": 0.65407839,
      "epoch": 1.5678589548452562,
      "grad_norm": 7.5,
      "learning_rate": 1.223000636528377e-06,
      "loss": 1.64742126,
      "memory(GiB)": 117.38,
      "step": 61805,
      "train_speed(iter/s)": 1.637881
    },
    {
      "acc": 0.67551012,
      "epoch": 1.5679857940131914,
      "grad_norm": 5.34375,
      "learning_rate": 1.2223135941532754e-06,
      "loss": 1.50869789,
      "memory(GiB)": 117.38,
      "step": 61810,
      "train_speed(iter/s)": 1.637896
    },
    {
      "acc": 0.6596117,
      "epoch": 1.5681126331811264,
      "grad_norm": 6.53125,
      "learning_rate": 1.2216267179376857e-06,
      "loss": 1.61112366,
      "memory(GiB)": 117.38,
      "step": 61815,
      "train_speed(iter/s)": 1.637909
    },
    {
      "acc": 0.66198697,
      "epoch": 1.5682394723490614,
      "grad_norm": 5.71875,
      "learning_rate": 1.2209400079118233e-06,
      "loss": 1.61957493,
      "memory(GiB)": 117.38,
      "step": 61820,
      "train_speed(iter/s)": 1.637923
    },
    {
      "acc": 0.66062198,
      "epoch": 1.5683663115169963,
      "grad_norm": 4.6875,
      "learning_rate": 1.2202534641058916e-06,
      "loss": 1.57511215,
      "memory(GiB)": 117.38,
      "step": 61825,
      "train_speed(iter/s)": 1.637937
    },
    {
      "acc": 0.68316174,
      "epoch": 1.5684931506849316,
      "grad_norm": 5.34375,
      "learning_rate": 1.2195670865500896e-06,
      "loss": 1.45706577,
      "memory(GiB)": 117.38,
      "step": 61830,
      "train_speed(iter/s)": 1.63795
    },
    {
      "acc": 0.64681873,
      "epoch": 1.5686199898528665,
      "grad_norm": 5.375,
      "learning_rate": 1.2188808752746022e-06,
      "loss": 1.65325012,
      "memory(GiB)": 117.38,
      "step": 61835,
      "train_speed(iter/s)": 1.637964
    },
    {
      "acc": 0.6711802,
      "epoch": 1.5687468290208018,
      "grad_norm": 6.6875,
      "learning_rate": 1.2181948303096176e-06,
      "loss": 1.57406139,
      "memory(GiB)": 117.38,
      "step": 61840,
      "train_speed(iter/s)": 1.637978
    },
    {
      "acc": 0.65753489,
      "epoch": 1.5688736681887367,
      "grad_norm": 5.625,
      "learning_rate": 1.2175089516853083e-06,
      "loss": 1.5384573,
      "memory(GiB)": 117.38,
      "step": 61845,
      "train_speed(iter/s)": 1.637992
    },
    {
      "acc": 0.67108984,
      "epoch": 1.5690005073566717,
      "grad_norm": 5.15625,
      "learning_rate": 1.216823239431843e-06,
      "loss": 1.54368925,
      "memory(GiB)": 117.38,
      "step": 61850,
      "train_speed(iter/s)": 1.638006
    },
    {
      "acc": 0.66490846,
      "epoch": 1.5691273465246067,
      "grad_norm": 7.6875,
      "learning_rate": 1.2161376935793827e-06,
      "loss": 1.57292805,
      "memory(GiB)": 117.38,
      "step": 61855,
      "train_speed(iter/s)": 1.63802
    },
    {
      "acc": 0.64931173,
      "epoch": 1.5692541856925417,
      "grad_norm": 5.90625,
      "learning_rate": 1.21545231415808e-06,
      "loss": 1.618433,
      "memory(GiB)": 117.38,
      "step": 61860,
      "train_speed(iter/s)": 1.638033
    },
    {
      "acc": 0.64918823,
      "epoch": 1.569381024860477,
      "grad_norm": 5.75,
      "learning_rate": 1.2147671011980816e-06,
      "loss": 1.65103722,
      "memory(GiB)": 117.38,
      "step": 61865,
      "train_speed(iter/s)": 1.638048
    },
    {
      "acc": 0.66030388,
      "epoch": 1.5695078640284121,
      "grad_norm": 5.96875,
      "learning_rate": 1.2140820547295256e-06,
      "loss": 1.55695314,
      "memory(GiB)": 117.38,
      "step": 61870,
      "train_speed(iter/s)": 1.638063
    },
    {
      "acc": 0.64868951,
      "epoch": 1.5696347031963471,
      "grad_norm": 5.4375,
      "learning_rate": 1.2133971747825435e-06,
      "loss": 1.53055067,
      "memory(GiB)": 117.38,
      "step": 61875,
      "train_speed(iter/s)": 1.638076
    },
    {
      "acc": 0.66051111,
      "epoch": 1.5697615423642821,
      "grad_norm": 9.0625,
      "learning_rate": 1.2127124613872603e-06,
      "loss": 1.60274849,
      "memory(GiB)": 117.38,
      "step": 61880,
      "train_speed(iter/s)": 1.638091
    },
    {
      "acc": 0.65472326,
      "epoch": 1.569888381532217,
      "grad_norm": 6.78125,
      "learning_rate": 1.2120279145737918e-06,
      "loss": 1.61206284,
      "memory(GiB)": 117.38,
      "step": 61885,
      "train_speed(iter/s)": 1.638104
    },
    {
      "acc": 0.67270355,
      "epoch": 1.570015220700152,
      "grad_norm": 5.75,
      "learning_rate": 1.2113435343722474e-06,
      "loss": 1.51774015,
      "memory(GiB)": 117.38,
      "step": 61890,
      "train_speed(iter/s)": 1.638118
    },
    {
      "acc": 0.64952636,
      "epoch": 1.5701420598680873,
      "grad_norm": 5.59375,
      "learning_rate": 1.21065932081273e-06,
      "loss": 1.60222054,
      "memory(GiB)": 117.38,
      "step": 61895,
      "train_speed(iter/s)": 1.638131
    },
    {
      "acc": 0.65741158,
      "epoch": 1.5702688990360223,
      "grad_norm": 5.34375,
      "learning_rate": 1.2099752739253334e-06,
      "loss": 1.61414833,
      "memory(GiB)": 117.38,
      "step": 61900,
      "train_speed(iter/s)": 1.638144
    },
    {
      "acc": 0.67015448,
      "epoch": 1.5703957382039575,
      "grad_norm": 6.125,
      "learning_rate": 1.209291393740144e-06,
      "loss": 1.56904812,
      "memory(GiB)": 117.38,
      "step": 61905,
      "train_speed(iter/s)": 1.638159
    },
    {
      "acc": 0.64554729,
      "epoch": 1.5705225773718925,
      "grad_norm": 5.65625,
      "learning_rate": 1.2086076802872472e-06,
      "loss": 1.63488846,
      "memory(GiB)": 117.38,
      "step": 61910,
      "train_speed(iter/s)": 1.638172
    },
    {
      "acc": 0.65189953,
      "epoch": 1.5706494165398275,
      "grad_norm": 5.53125,
      "learning_rate": 1.2079241335967096e-06,
      "loss": 1.6647541,
      "memory(GiB)": 117.38,
      "step": 61915,
      "train_speed(iter/s)": 1.638186
    },
    {
      "acc": 0.65260954,
      "epoch": 1.5707762557077625,
      "grad_norm": 5.75,
      "learning_rate": 1.207240753698599e-06,
      "loss": 1.63380394,
      "memory(GiB)": 117.38,
      "step": 61920,
      "train_speed(iter/s)": 1.6382
    },
    {
      "acc": 0.68474879,
      "epoch": 1.5709030948756975,
      "grad_norm": 6.34375,
      "learning_rate": 1.2065575406229723e-06,
      "loss": 1.44651051,
      "memory(GiB)": 117.38,
      "step": 61925,
      "train_speed(iter/s)": 1.638213
    },
    {
      "acc": 0.67783751,
      "epoch": 1.5710299340436327,
      "grad_norm": 5.53125,
      "learning_rate": 1.2058744943998847e-06,
      "loss": 1.54204235,
      "memory(GiB)": 117.38,
      "step": 61930,
      "train_speed(iter/s)": 1.638227
    },
    {
      "acc": 0.66343856,
      "epoch": 1.5711567732115679,
      "grad_norm": 6.84375,
      "learning_rate": 1.2051916150593746e-06,
      "loss": 1.588445,
      "memory(GiB)": 117.38,
      "step": 61935,
      "train_speed(iter/s)": 1.638241
    },
    {
      "acc": 0.64701262,
      "epoch": 1.5712836123795029,
      "grad_norm": 6.125,
      "learning_rate": 1.2045089026314783e-06,
      "loss": 1.61286488,
      "memory(GiB)": 117.38,
      "step": 61940,
      "train_speed(iter/s)": 1.638254
    },
    {
      "acc": 0.64579959,
      "epoch": 1.5714104515474379,
      "grad_norm": 6.34375,
      "learning_rate": 1.2038263571462278e-06,
      "loss": 1.59338303,
      "memory(GiB)": 117.38,
      "step": 61945,
      "train_speed(iter/s)": 1.638268
    },
    {
      "acc": 0.66114888,
      "epoch": 1.5715372907153728,
      "grad_norm": 8.625,
      "learning_rate": 1.203143978633644e-06,
      "loss": 1.62083321,
      "memory(GiB)": 117.38,
      "step": 61950,
      "train_speed(iter/s)": 1.638282
    },
    {
      "acc": 0.650282,
      "epoch": 1.5716641298833078,
      "grad_norm": 6.5625,
      "learning_rate": 1.2024617671237388e-06,
      "loss": 1.56989927,
      "memory(GiB)": 117.38,
      "step": 61955,
      "train_speed(iter/s)": 1.638294
    },
    {
      "acc": 0.64680548,
      "epoch": 1.571790969051243,
      "grad_norm": 4.09375,
      "learning_rate": 1.2017797226465178e-06,
      "loss": 1.622472,
      "memory(GiB)": 117.38,
      "step": 61960,
      "train_speed(iter/s)": 1.638307
    },
    {
      "acc": 0.63211393,
      "epoch": 1.571917808219178,
      "grad_norm": 5.125,
      "learning_rate": 1.2010978452319843e-06,
      "loss": 1.63500671,
      "memory(GiB)": 117.38,
      "step": 61965,
      "train_speed(iter/s)": 1.63832
    },
    {
      "acc": 0.64652376,
      "epoch": 1.5720446473871132,
      "grad_norm": 6.0625,
      "learning_rate": 1.2004161349101295e-06,
      "loss": 1.61188011,
      "memory(GiB)": 117.38,
      "step": 61970,
      "train_speed(iter/s)": 1.638335
    },
    {
      "acc": 0.66792326,
      "epoch": 1.5721714865550482,
      "grad_norm": 7.78125,
      "learning_rate": 1.1997345917109348e-06,
      "loss": 1.57882843,
      "memory(GiB)": 117.38,
      "step": 61975,
      "train_speed(iter/s)": 1.638349
    },
    {
      "acc": 0.65533686,
      "epoch": 1.5722983257229832,
      "grad_norm": 6.15625,
      "learning_rate": 1.1990532156643808e-06,
      "loss": 1.63751373,
      "memory(GiB)": 117.38,
      "step": 61980,
      "train_speed(iter/s)": 1.638363
    },
    {
      "acc": 0.65759177,
      "epoch": 1.5724251648909182,
      "grad_norm": 5.625,
      "learning_rate": 1.198372006800436e-06,
      "loss": 1.59721909,
      "memory(GiB)": 117.38,
      "step": 61985,
      "train_speed(iter/s)": 1.638378
    },
    {
      "acc": 0.64423833,
      "epoch": 1.5725520040588534,
      "grad_norm": 6.3125,
      "learning_rate": 1.1976909651490637e-06,
      "loss": 1.61672058,
      "memory(GiB)": 117.38,
      "step": 61990,
      "train_speed(iter/s)": 1.638392
    },
    {
      "acc": 0.66801519,
      "epoch": 1.5726788432267884,
      "grad_norm": 4.5,
      "learning_rate": 1.1970100907402188e-06,
      "loss": 1.56915073,
      "memory(GiB)": 117.38,
      "step": 61995,
      "train_speed(iter/s)": 1.638406
    },
    {
      "acc": 0.661198,
      "epoch": 1.5728056823947236,
      "grad_norm": 7.5,
      "learning_rate": 1.1963293836038492e-06,
      "loss": 1.62549477,
      "memory(GiB)": 117.38,
      "step": 62000,
      "train_speed(iter/s)": 1.638419
    },
    {
      "epoch": 1.5728056823947236,
      "eval_acc": 0.6462472095009575,
      "eval_loss": 1.5733178853988647,
      "eval_runtime": 58.6077,
      "eval_samples_per_second": 108.689,
      "eval_steps_per_second": 27.181,
      "step": 62000
    },
    {
      "acc": 0.6578105,
      "epoch": 1.5729325215626586,
      "grad_norm": 6.34375,
      "learning_rate": 1.195648843769896e-06,
      "loss": 1.51423578,
      "memory(GiB)": 117.38,
      "step": 62005,
      "train_speed(iter/s)": 1.635712
    },
    {
      "acc": 0.6574604,
      "epoch": 1.5730593607305936,
      "grad_norm": 6.78125,
      "learning_rate": 1.1949684712682912e-06,
      "loss": 1.66762466,
      "memory(GiB)": 117.38,
      "step": 62010,
      "train_speed(iter/s)": 1.635725
    },
    {
      "acc": 0.66303735,
      "epoch": 1.5731861998985286,
      "grad_norm": 6.03125,
      "learning_rate": 1.1942882661289618e-06,
      "loss": 1.5493082,
      "memory(GiB)": 117.38,
      "step": 62015,
      "train_speed(iter/s)": 1.635738
    },
    {
      "acc": 0.66413956,
      "epoch": 1.5733130390664636,
      "grad_norm": 7.875,
      "learning_rate": 1.1936082283818252e-06,
      "loss": 1.57665606,
      "memory(GiB)": 117.38,
      "step": 62020,
      "train_speed(iter/s)": 1.635752
    },
    {
      "acc": 0.67713356,
      "epoch": 1.5734398782343988,
      "grad_norm": 5.34375,
      "learning_rate": 1.1929283580567936e-06,
      "loss": 1.51045151,
      "memory(GiB)": 117.38,
      "step": 62025,
      "train_speed(iter/s)": 1.635765
    },
    {
      "acc": 0.67225881,
      "epoch": 1.573566717402334,
      "grad_norm": 6.09375,
      "learning_rate": 1.1922486551837697e-06,
      "loss": 1.58824205,
      "memory(GiB)": 117.38,
      "step": 62030,
      "train_speed(iter/s)": 1.635778
    },
    {
      "acc": 0.65589252,
      "epoch": 1.573693556570269,
      "grad_norm": 6.0625,
      "learning_rate": 1.1915691197926505e-06,
      "loss": 1.65517998,
      "memory(GiB)": 117.38,
      "step": 62035,
      "train_speed(iter/s)": 1.635791
    },
    {
      "acc": 0.65546012,
      "epoch": 1.573820395738204,
      "grad_norm": 6.09375,
      "learning_rate": 1.1908897519133244e-06,
      "loss": 1.5825655,
      "memory(GiB)": 117.38,
      "step": 62040,
      "train_speed(iter/s)": 1.635804
    },
    {
      "acc": 0.66267967,
      "epoch": 1.573947234906139,
      "grad_norm": 4.71875,
      "learning_rate": 1.190210551575674e-06,
      "loss": 1.56688929,
      "memory(GiB)": 117.38,
      "step": 62045,
      "train_speed(iter/s)": 1.635817
    },
    {
      "acc": 0.66432581,
      "epoch": 1.574074074074074,
      "grad_norm": 5.71875,
      "learning_rate": 1.189531518809573e-06,
      "loss": 1.5060647,
      "memory(GiB)": 117.38,
      "step": 62050,
      "train_speed(iter/s)": 1.63583
    },
    {
      "acc": 0.63940964,
      "epoch": 1.5742009132420092,
      "grad_norm": 6.28125,
      "learning_rate": 1.188852653644888e-06,
      "loss": 1.72135315,
      "memory(GiB)": 117.38,
      "step": 62055,
      "train_speed(iter/s)": 1.635843
    },
    {
      "acc": 0.65729837,
      "epoch": 1.5743277524099442,
      "grad_norm": 6.5625,
      "learning_rate": 1.1881739561114792e-06,
      "loss": 1.55113869,
      "memory(GiB)": 117.38,
      "step": 62060,
      "train_speed(iter/s)": 1.635857
    },
    {
      "acc": 0.65653839,
      "epoch": 1.5744545915778794,
      "grad_norm": 6.15625,
      "learning_rate": 1.1874954262391968e-06,
      "loss": 1.62202301,
      "memory(GiB)": 117.38,
      "step": 62065,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.65237069,
      "epoch": 1.5745814307458144,
      "grad_norm": 6.65625,
      "learning_rate": 1.1868170640578901e-06,
      "loss": 1.67198792,
      "memory(GiB)": 117.38,
      "step": 62070,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.6636961,
      "epoch": 1.5747082699137493,
      "grad_norm": 4.84375,
      "learning_rate": 1.1861388695973918e-06,
      "loss": 1.56774473,
      "memory(GiB)": 117.38,
      "step": 62075,
      "train_speed(iter/s)": 1.635897
    },
    {
      "acc": 0.6428288,
      "epoch": 1.5748351090816843,
      "grad_norm": 5.8125,
      "learning_rate": 1.1854608428875332e-06,
      "loss": 1.62966423,
      "memory(GiB)": 117.38,
      "step": 62080,
      "train_speed(iter/s)": 1.635911
    },
    {
      "acc": 0.66092014,
      "epoch": 1.5749619482496193,
      "grad_norm": 5.90625,
      "learning_rate": 1.1847829839581377e-06,
      "loss": 1.58559055,
      "memory(GiB)": 117.38,
      "step": 62085,
      "train_speed(iter/s)": 1.635924
    },
    {
      "acc": 0.6670011,
      "epoch": 1.5750887874175545,
      "grad_norm": 4.4375,
      "learning_rate": 1.1841052928390223e-06,
      "loss": 1.5324194,
      "memory(GiB)": 117.38,
      "step": 62090,
      "train_speed(iter/s)": 1.635938
    },
    {
      "acc": 0.65734129,
      "epoch": 1.5752156265854897,
      "grad_norm": 5.53125,
      "learning_rate": 1.183427769559991e-06,
      "loss": 1.54112549,
      "memory(GiB)": 117.38,
      "step": 62095,
      "train_speed(iter/s)": 1.635952
    },
    {
      "acc": 0.66071706,
      "epoch": 1.5753424657534247,
      "grad_norm": 5.90625,
      "learning_rate": 1.1827504141508456e-06,
      "loss": 1.53575726,
      "memory(GiB)": 117.38,
      "step": 62100,
      "train_speed(iter/s)": 1.635966
    },
    {
      "acc": 0.64720974,
      "epoch": 1.5754693049213597,
      "grad_norm": 5.875,
      "learning_rate": 1.1820732266413803e-06,
      "loss": 1.64022408,
      "memory(GiB)": 117.38,
      "step": 62105,
      "train_speed(iter/s)": 1.63598
    },
    {
      "acc": 0.6631947,
      "epoch": 1.5755961440892947,
      "grad_norm": 6.6875,
      "learning_rate": 1.181396207061382e-06,
      "loss": 1.59497671,
      "memory(GiB)": 117.38,
      "step": 62110,
      "train_speed(iter/s)": 1.635994
    },
    {
      "acc": 0.66114702,
      "epoch": 1.5757229832572297,
      "grad_norm": 5.71875,
      "learning_rate": 1.1807193554406248e-06,
      "loss": 1.60235004,
      "memory(GiB)": 117.38,
      "step": 62115,
      "train_speed(iter/s)": 1.636008
    },
    {
      "acc": 0.64758363,
      "epoch": 1.575849822425165,
      "grad_norm": 6.28125,
      "learning_rate": 1.1800426718088837e-06,
      "loss": 1.62359772,
      "memory(GiB)": 117.38,
      "step": 62120,
      "train_speed(iter/s)": 1.636021
    },
    {
      "acc": 0.66658449,
      "epoch": 1.5759766615931,
      "grad_norm": 5.25,
      "learning_rate": 1.1793661561959201e-06,
      "loss": 1.6198698,
      "memory(GiB)": 117.38,
      "step": 62125,
      "train_speed(iter/s)": 1.636036
    },
    {
      "acc": 0.66156578,
      "epoch": 1.5761035007610351,
      "grad_norm": 8.3125,
      "learning_rate": 1.178689808631493e-06,
      "loss": 1.54968681,
      "memory(GiB)": 117.38,
      "step": 62130,
      "train_speed(iter/s)": 1.636051
    },
    {
      "acc": 0.65869455,
      "epoch": 1.57623033992897,
      "grad_norm": 5.90625,
      "learning_rate": 1.178013629145346e-06,
      "loss": 1.51563187,
      "memory(GiB)": 117.38,
      "step": 62135,
      "train_speed(iter/s)": 1.636066
    },
    {
      "acc": 0.6438611,
      "epoch": 1.576357179096905,
      "grad_norm": 6.40625,
      "learning_rate": 1.1773376177672246e-06,
      "loss": 1.61444626,
      "memory(GiB)": 117.38,
      "step": 62140,
      "train_speed(iter/s)": 1.63608
    },
    {
      "acc": 0.66838465,
      "epoch": 1.57648401826484,
      "grad_norm": 5.34375,
      "learning_rate": 1.176661774526862e-06,
      "loss": 1.56084623,
      "memory(GiB)": 117.38,
      "step": 62145,
      "train_speed(iter/s)": 1.636094
    },
    {
      "acc": 0.65313196,
      "epoch": 1.5766108574327753,
      "grad_norm": 5.25,
      "learning_rate": 1.1759860994539846e-06,
      "loss": 1.61813583,
      "memory(GiB)": 117.38,
      "step": 62150,
      "train_speed(iter/s)": 1.636107
    },
    {
      "acc": 0.67126603,
      "epoch": 1.5767376966007103,
      "grad_norm": 6.21875,
      "learning_rate": 1.1753105925783114e-06,
      "loss": 1.59181824,
      "memory(GiB)": 117.38,
      "step": 62155,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.65228877,
      "epoch": 1.5768645357686455,
      "grad_norm": 5.125,
      "learning_rate": 1.174635253929554e-06,
      "loss": 1.61933212,
      "memory(GiB)": 117.38,
      "step": 62160,
      "train_speed(iter/s)": 1.636135
    },
    {
      "acc": 0.6476645,
      "epoch": 1.5769913749365805,
      "grad_norm": 7.03125,
      "learning_rate": 1.1739600835374177e-06,
      "loss": 1.61889229,
      "memory(GiB)": 117.38,
      "step": 62165,
      "train_speed(iter/s)": 1.636148
    },
    {
      "acc": 0.65815201,
      "epoch": 1.5771182141045155,
      "grad_norm": 5.71875,
      "learning_rate": 1.173285081431599e-06,
      "loss": 1.55051842,
      "memory(GiB)": 117.38,
      "step": 62170,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.65677757,
      "epoch": 1.5772450532724505,
      "grad_norm": 4.75,
      "learning_rate": 1.1726102476417871e-06,
      "loss": 1.62125092,
      "memory(GiB)": 117.38,
      "step": 62175,
      "train_speed(iter/s)": 1.636175
    },
    {
      "acc": 0.66700225,
      "epoch": 1.5773718924403854,
      "grad_norm": 6.03125,
      "learning_rate": 1.1719355821976647e-06,
      "loss": 1.57801123,
      "memory(GiB)": 117.38,
      "step": 62180,
      "train_speed(iter/s)": 1.636188
    },
    {
      "acc": 0.64894843,
      "epoch": 1.5774987316083207,
      "grad_norm": 5.71875,
      "learning_rate": 1.1712610851289069e-06,
      "loss": 1.66683464,
      "memory(GiB)": 117.38,
      "step": 62185,
      "train_speed(iter/s)": 1.636201
    },
    {
      "acc": 0.64769721,
      "epoch": 1.5776255707762559,
      "grad_norm": 6.125,
      "learning_rate": 1.1705867564651802e-06,
      "loss": 1.61999207,
      "memory(GiB)": 117.38,
      "step": 62190,
      "train_speed(iter/s)": 1.636215
    },
    {
      "acc": 0.65408702,
      "epoch": 1.5777524099441909,
      "grad_norm": 5.34375,
      "learning_rate": 1.1699125962361451e-06,
      "loss": 1.67037792,
      "memory(GiB)": 117.38,
      "step": 62195,
      "train_speed(iter/s)": 1.636229
    },
    {
      "acc": 0.66943359,
      "epoch": 1.5778792491121258,
      "grad_norm": 5.8125,
      "learning_rate": 1.1692386044714543e-06,
      "loss": 1.51186943,
      "memory(GiB)": 117.38,
      "step": 62200,
      "train_speed(iter/s)": 1.636242
    },
    {
      "acc": 0.63414979,
      "epoch": 1.5780060882800608,
      "grad_norm": 6.1875,
      "learning_rate": 1.1685647812007512e-06,
      "loss": 1.66729527,
      "memory(GiB)": 117.38,
      "step": 62205,
      "train_speed(iter/s)": 1.636256
    },
    {
      "acc": 0.64741068,
      "epoch": 1.5781329274479958,
      "grad_norm": 4.96875,
      "learning_rate": 1.167891126453678e-06,
      "loss": 1.58616371,
      "memory(GiB)": 117.38,
      "step": 62210,
      "train_speed(iter/s)": 1.636269
    },
    {
      "acc": 0.65222759,
      "epoch": 1.578259766615931,
      "grad_norm": 5.5625,
      "learning_rate": 1.1672176402598607e-06,
      "loss": 1.64116974,
      "memory(GiB)": 117.38,
      "step": 62215,
      "train_speed(iter/s)": 1.636284
    },
    {
      "acc": 0.66092558,
      "epoch": 1.578386605783866,
      "grad_norm": 5.875,
      "learning_rate": 1.1665443226489236e-06,
      "loss": 1.50364008,
      "memory(GiB)": 117.38,
      "step": 62220,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.65456295,
      "epoch": 1.5785134449518012,
      "grad_norm": 5.6875,
      "learning_rate": 1.1658711736504814e-06,
      "loss": 1.5665453,
      "memory(GiB)": 117.38,
      "step": 62225,
      "train_speed(iter/s)": 1.636308
    },
    {
      "acc": 0.65833483,
      "epoch": 1.5786402841197362,
      "grad_norm": 5.1875,
      "learning_rate": 1.165198193294146e-06,
      "loss": 1.58607111,
      "memory(GiB)": 117.38,
      "step": 62230,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.67806826,
      "epoch": 1.5787671232876712,
      "grad_norm": 7.34375,
      "learning_rate": 1.1645253816095131e-06,
      "loss": 1.49834433,
      "memory(GiB)": 117.38,
      "step": 62235,
      "train_speed(iter/s)": 1.636336
    },
    {
      "acc": 0.66620808,
      "epoch": 1.5788939624556062,
      "grad_norm": 5.0,
      "learning_rate": 1.1638527386261772e-06,
      "loss": 1.63065815,
      "memory(GiB)": 117.38,
      "step": 62240,
      "train_speed(iter/s)": 1.63635
    },
    {
      "acc": 0.65975437,
      "epoch": 1.5790208016235412,
      "grad_norm": 5.5,
      "learning_rate": 1.163180264373726e-06,
      "loss": 1.54425268,
      "memory(GiB)": 117.38,
      "step": 62245,
      "train_speed(iter/s)": 1.636364
    },
    {
      "acc": 0.65600834,
      "epoch": 1.5791476407914764,
      "grad_norm": 6.75,
      "learning_rate": 1.162507958881739e-06,
      "loss": 1.61289635,
      "memory(GiB)": 117.38,
      "step": 62250,
      "train_speed(iter/s)": 1.636378
    },
    {
      "acc": 0.68163147,
      "epoch": 1.5792744799594116,
      "grad_norm": 4.90625,
      "learning_rate": 1.161835822179782e-06,
      "loss": 1.46188602,
      "memory(GiB)": 117.38,
      "step": 62255,
      "train_speed(iter/s)": 1.636392
    },
    {
      "acc": 0.64814558,
      "epoch": 1.5794013191273466,
      "grad_norm": 5.25,
      "learning_rate": 1.1611638542974229e-06,
      "loss": 1.63704948,
      "memory(GiB)": 117.38,
      "step": 62260,
      "train_speed(iter/s)": 1.636405
    },
    {
      "acc": 0.6573988,
      "epoch": 1.5795281582952816,
      "grad_norm": 7.4375,
      "learning_rate": 1.160492055264217e-06,
      "loss": 1.63067589,
      "memory(GiB)": 117.38,
      "step": 62265,
      "train_speed(iter/s)": 1.636419
    },
    {
      "acc": 0.65740538,
      "epoch": 1.5796549974632166,
      "grad_norm": 5.03125,
      "learning_rate": 1.159820425109714e-06,
      "loss": 1.57271214,
      "memory(GiB)": 117.38,
      "step": 62270,
      "train_speed(iter/s)": 1.636433
    },
    {
      "acc": 0.65450792,
      "epoch": 1.5797818366311516,
      "grad_norm": 5.28125,
      "learning_rate": 1.1591489638634513e-06,
      "loss": 1.61792564,
      "memory(GiB)": 117.38,
      "step": 62275,
      "train_speed(iter/s)": 1.636448
    },
    {
      "acc": 0.66762323,
      "epoch": 1.5799086757990868,
      "grad_norm": 5.21875,
      "learning_rate": 1.1584776715549662e-06,
      "loss": 1.61795654,
      "memory(GiB)": 117.38,
      "step": 62280,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.659761,
      "epoch": 1.5800355149670218,
      "grad_norm": 5.46875,
      "learning_rate": 1.1578065482137845e-06,
      "loss": 1.5779583,
      "memory(GiB)": 117.38,
      "step": 62285,
      "train_speed(iter/s)": 1.636475
    },
    {
      "acc": 0.63942657,
      "epoch": 1.580162354134957,
      "grad_norm": 5.78125,
      "learning_rate": 1.157135593869425e-06,
      "loss": 1.66861153,
      "memory(GiB)": 117.38,
      "step": 62290,
      "train_speed(iter/s)": 1.636489
    },
    {
      "acc": 0.65551019,
      "epoch": 1.580289193302892,
      "grad_norm": 5.21875,
      "learning_rate": 1.156464808551399e-06,
      "loss": 1.59630804,
      "memory(GiB)": 117.38,
      "step": 62295,
      "train_speed(iter/s)": 1.636502
    },
    {
      "acc": 0.65974436,
      "epoch": 1.580416032470827,
      "grad_norm": 5.71875,
      "learning_rate": 1.1557941922892113e-06,
      "loss": 1.59582405,
      "memory(GiB)": 117.38,
      "step": 62300,
      "train_speed(iter/s)": 1.636516
    },
    {
      "acc": 0.65655079,
      "epoch": 1.580542871638762,
      "grad_norm": 5.15625,
      "learning_rate": 1.155123745112358e-06,
      "loss": 1.52191238,
      "memory(GiB)": 117.38,
      "step": 62305,
      "train_speed(iter/s)": 1.63653
    },
    {
      "acc": 0.68731256,
      "epoch": 1.5806697108066972,
      "grad_norm": 6.96875,
      "learning_rate": 1.1544534670503282e-06,
      "loss": 1.47692757,
      "memory(GiB)": 117.38,
      "step": 62310,
      "train_speed(iter/s)": 1.636544
    },
    {
      "acc": 0.65880308,
      "epoch": 1.5807965499746321,
      "grad_norm": 6.125,
      "learning_rate": 1.1537833581326048e-06,
      "loss": 1.54738388,
      "memory(GiB)": 117.38,
      "step": 62315,
      "train_speed(iter/s)": 1.636558
    },
    {
      "acc": 0.66302571,
      "epoch": 1.5809233891425674,
      "grad_norm": 7.125,
      "learning_rate": 1.1531134183886606e-06,
      "loss": 1.57727394,
      "memory(GiB)": 117.38,
      "step": 62320,
      "train_speed(iter/s)": 1.636571
    },
    {
      "acc": 0.66358538,
      "epoch": 1.5810502283105023,
      "grad_norm": 5.875,
      "learning_rate": 1.1524436478479638e-06,
      "loss": 1.59886723,
      "memory(GiB)": 117.38,
      "step": 62325,
      "train_speed(iter/s)": 1.636585
    },
    {
      "acc": 0.65415287,
      "epoch": 1.5811770674784373,
      "grad_norm": 7.3125,
      "learning_rate": 1.1517740465399736e-06,
      "loss": 1.62806416,
      "memory(GiB)": 117.38,
      "step": 62330,
      "train_speed(iter/s)": 1.636599
    },
    {
      "acc": 0.68105745,
      "epoch": 1.5813039066463723,
      "grad_norm": 6.15625,
      "learning_rate": 1.1511046144941417e-06,
      "loss": 1.56159506,
      "memory(GiB)": 117.38,
      "step": 62335,
      "train_speed(iter/s)": 1.636612
    },
    {
      "acc": 0.66029892,
      "epoch": 1.5814307458143073,
      "grad_norm": 5.9375,
      "learning_rate": 1.150435351739913e-06,
      "loss": 1.5606842,
      "memory(GiB)": 117.38,
      "step": 62340,
      "train_speed(iter/s)": 1.636626
    },
    {
      "acc": 0.66902351,
      "epoch": 1.5815575849822425,
      "grad_norm": 4.90625,
      "learning_rate": 1.1497662583067231e-06,
      "loss": 1.54241104,
      "memory(GiB)": 117.38,
      "step": 62345,
      "train_speed(iter/s)": 1.636638
    },
    {
      "acc": 0.67156844,
      "epoch": 1.5816844241501777,
      "grad_norm": 6.0625,
      "learning_rate": 1.1490973342240063e-06,
      "loss": 1.56138258,
      "memory(GiB)": 117.38,
      "step": 62350,
      "train_speed(iter/s)": 1.636652
    },
    {
      "acc": 0.64750681,
      "epoch": 1.5818112633181127,
      "grad_norm": 5.90625,
      "learning_rate": 1.1484285795211803e-06,
      "loss": 1.66835213,
      "memory(GiB)": 117.38,
      "step": 62355,
      "train_speed(iter/s)": 1.636666
    },
    {
      "acc": 0.66828451,
      "epoch": 1.5819381024860477,
      "grad_norm": 7.0625,
      "learning_rate": 1.1477599942276613e-06,
      "loss": 1.60028343,
      "memory(GiB)": 117.38,
      "step": 62360,
      "train_speed(iter/s)": 1.63668
    },
    {
      "acc": 0.6553627,
      "epoch": 1.5820649416539827,
      "grad_norm": 5.65625,
      "learning_rate": 1.1470915783728547e-06,
      "loss": 1.54887571,
      "memory(GiB)": 117.38,
      "step": 62365,
      "train_speed(iter/s)": 1.636693
    },
    {
      "acc": 0.66102877,
      "epoch": 1.5821917808219177,
      "grad_norm": 5.09375,
      "learning_rate": 1.1464233319861661e-06,
      "loss": 1.55981331,
      "memory(GiB)": 117.38,
      "step": 62370,
      "train_speed(iter/s)": 1.636706
    },
    {
      "acc": 0.6541666,
      "epoch": 1.582318619989853,
      "grad_norm": 5.5625,
      "learning_rate": 1.145755255096983e-06,
      "loss": 1.55644989,
      "memory(GiB)": 117.38,
      "step": 62375,
      "train_speed(iter/s)": 1.63672
    },
    {
      "acc": 0.64556084,
      "epoch": 1.582445459157788,
      "grad_norm": 5.5625,
      "learning_rate": 1.1450873477346901e-06,
      "loss": 1.60299301,
      "memory(GiB)": 117.38,
      "step": 62380,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.6577282,
      "epoch": 1.582572298325723,
      "grad_norm": 5.90625,
      "learning_rate": 1.1444196099286682e-06,
      "loss": 1.55571156,
      "memory(GiB)": 117.38,
      "step": 62385,
      "train_speed(iter/s)": 1.636747
    },
    {
      "acc": 0.64839506,
      "epoch": 1.582699137493658,
      "grad_norm": 5.1875,
      "learning_rate": 1.143752041708287e-06,
      "loss": 1.64312897,
      "memory(GiB)": 117.38,
      "step": 62390,
      "train_speed(iter/s)": 1.63676
    },
    {
      "acc": 0.65850468,
      "epoch": 1.582825976661593,
      "grad_norm": 10.125,
      "learning_rate": 1.1430846431029062e-06,
      "loss": 1.57215195,
      "memory(GiB)": 117.38,
      "step": 62395,
      "train_speed(iter/s)": 1.636774
    },
    {
      "acc": 0.64436893,
      "epoch": 1.582952815829528,
      "grad_norm": 6.8125,
      "learning_rate": 1.1424174141418815e-06,
      "loss": 1.68215828,
      "memory(GiB)": 117.38,
      "step": 62400,
      "train_speed(iter/s)": 1.636787
    },
    {
      "acc": 0.65718699,
      "epoch": 1.583079654997463,
      "grad_norm": 6.59375,
      "learning_rate": 1.1417503548545634e-06,
      "loss": 1.60325508,
      "memory(GiB)": 117.38,
      "step": 62405,
      "train_speed(iter/s)": 1.6368
    },
    {
      "acc": 0.64169102,
      "epoch": 1.5832064941653983,
      "grad_norm": 5.28125,
      "learning_rate": 1.1410834652702918e-06,
      "loss": 1.63935871,
      "memory(GiB)": 117.38,
      "step": 62410,
      "train_speed(iter/s)": 1.636815
    },
    {
      "acc": 0.6585021,
      "epoch": 1.5833333333333335,
      "grad_norm": 7.03125,
      "learning_rate": 1.1404167454183957e-06,
      "loss": 1.56873226,
      "memory(GiB)": 117.38,
      "step": 62415,
      "train_speed(iter/s)": 1.636829
    },
    {
      "acc": 0.63946252,
      "epoch": 1.5834601725012685,
      "grad_norm": 5.28125,
      "learning_rate": 1.1397501953282042e-06,
      "loss": 1.64788818,
      "memory(GiB)": 117.38,
      "step": 62420,
      "train_speed(iter/s)": 1.636841
    },
    {
      "acc": 0.64478893,
      "epoch": 1.5835870116692035,
      "grad_norm": 8.5,
      "learning_rate": 1.1390838150290334e-06,
      "loss": 1.61939316,
      "memory(GiB)": 117.38,
      "step": 62425,
      "train_speed(iter/s)": 1.636854
    },
    {
      "acc": 0.66862283,
      "epoch": 1.5837138508371384,
      "grad_norm": 7.125,
      "learning_rate": 1.1384176045501944e-06,
      "loss": 1.56827202,
      "memory(GiB)": 117.38,
      "step": 62430,
      "train_speed(iter/s)": 1.636869
    },
    {
      "acc": 0.64439735,
      "epoch": 1.5838406900050734,
      "grad_norm": 5.4375,
      "learning_rate": 1.1377515639209902e-06,
      "loss": 1.6099762,
      "memory(GiB)": 117.38,
      "step": 62435,
      "train_speed(iter/s)": 1.636884
    },
    {
      "acc": 0.66032171,
      "epoch": 1.5839675291730086,
      "grad_norm": 7.34375,
      "learning_rate": 1.1370856931707159e-06,
      "loss": 1.63359871,
      "memory(GiB)": 117.38,
      "step": 62440,
      "train_speed(iter/s)": 1.636898
    },
    {
      "acc": 0.65483494,
      "epoch": 1.5840943683409436,
      "grad_norm": 7.34375,
      "learning_rate": 1.1364199923286589e-06,
      "loss": 1.60862846,
      "memory(GiB)": 117.38,
      "step": 62445,
      "train_speed(iter/s)": 1.63691
    },
    {
      "acc": 0.67228365,
      "epoch": 1.5842212075088788,
      "grad_norm": 7.59375,
      "learning_rate": 1.1357544614241012e-06,
      "loss": 1.57928505,
      "memory(GiB)": 117.38,
      "step": 62450,
      "train_speed(iter/s)": 1.636925
    },
    {
      "acc": 0.66140218,
      "epoch": 1.5843480466768138,
      "grad_norm": 6.9375,
      "learning_rate": 1.135089100486314e-06,
      "loss": 1.59014063,
      "memory(GiB)": 117.38,
      "step": 62455,
      "train_speed(iter/s)": 1.63694
    },
    {
      "acc": 0.659905,
      "epoch": 1.5844748858447488,
      "grad_norm": 6.0,
      "learning_rate": 1.1344239095445642e-06,
      "loss": 1.57373943,
      "memory(GiB)": 117.38,
      "step": 62460,
      "train_speed(iter/s)": 1.636952
    },
    {
      "acc": 0.66732683,
      "epoch": 1.5846017250126838,
      "grad_norm": 4.8125,
      "learning_rate": 1.1337588886281092e-06,
      "loss": 1.50141506,
      "memory(GiB)": 117.38,
      "step": 62465,
      "train_speed(iter/s)": 1.636965
    },
    {
      "acc": 0.66228685,
      "epoch": 1.584728564180619,
      "grad_norm": 5.9375,
      "learning_rate": 1.1330940377662002e-06,
      "loss": 1.59462929,
      "memory(GiB)": 117.38,
      "step": 62470,
      "train_speed(iter/s)": 1.636978
    },
    {
      "acc": 0.65680861,
      "epoch": 1.584855403348554,
      "grad_norm": 5.4375,
      "learning_rate": 1.1324293569880795e-06,
      "loss": 1.62669182,
      "memory(GiB)": 117.38,
      "step": 62475,
      "train_speed(iter/s)": 1.636992
    },
    {
      "acc": 0.66022425,
      "epoch": 1.5849822425164892,
      "grad_norm": 6.8125,
      "learning_rate": 1.1317648463229835e-06,
      "loss": 1.56034737,
      "memory(GiB)": 117.38,
      "step": 62480,
      "train_speed(iter/s)": 1.637007
    },
    {
      "acc": 0.64006104,
      "epoch": 1.5851090816844242,
      "grad_norm": 6.75,
      "learning_rate": 1.1311005058001396e-06,
      "loss": 1.63513985,
      "memory(GiB)": 117.38,
      "step": 62485,
      "train_speed(iter/s)": 1.637021
    },
    {
      "acc": 0.64555693,
      "epoch": 1.5852359208523592,
      "grad_norm": 5.375,
      "learning_rate": 1.130436335448769e-06,
      "loss": 1.64385014,
      "memory(GiB)": 117.38,
      "step": 62490,
      "train_speed(iter/s)": 1.637033
    },
    {
      "acc": 0.65974517,
      "epoch": 1.5853627600202942,
      "grad_norm": 5.9375,
      "learning_rate": 1.129772335298085e-06,
      "loss": 1.60770988,
      "memory(GiB)": 117.38,
      "step": 62495,
      "train_speed(iter/s)": 1.637046
    },
    {
      "acc": 0.64732871,
      "epoch": 1.5854895991882292,
      "grad_norm": 6.03125,
      "learning_rate": 1.1291085053772926e-06,
      "loss": 1.62634773,
      "memory(GiB)": 117.38,
      "step": 62500,
      "train_speed(iter/s)": 1.637059
    },
    {
      "acc": 0.65343914,
      "epoch": 1.5856164383561644,
      "grad_norm": 5.625,
      "learning_rate": 1.1284448457155893e-06,
      "loss": 1.64265175,
      "memory(GiB)": 117.38,
      "step": 62505,
      "train_speed(iter/s)": 1.637072
    },
    {
      "acc": 0.65827475,
      "epoch": 1.5857432775240996,
      "grad_norm": 6.53125,
      "learning_rate": 1.1277813563421697e-06,
      "loss": 1.591856,
      "memory(GiB)": 117.38,
      "step": 62510,
      "train_speed(iter/s)": 1.637085
    },
    {
      "acc": 0.66083798,
      "epoch": 1.5858701166920346,
      "grad_norm": 5.15625,
      "learning_rate": 1.127118037286213e-06,
      "loss": 1.55919638,
      "memory(GiB)": 117.38,
      "step": 62515,
      "train_speed(iter/s)": 1.637098
    },
    {
      "acc": 0.65217314,
      "epoch": 1.5859969558599696,
      "grad_norm": 5.65625,
      "learning_rate": 1.1264548885768944e-06,
      "loss": 1.66456738,
      "memory(GiB)": 117.38,
      "step": 62520,
      "train_speed(iter/s)": 1.637112
    },
    {
      "acc": 0.6598146,
      "epoch": 1.5861237950279046,
      "grad_norm": 5.9375,
      "learning_rate": 1.1257919102433856e-06,
      "loss": 1.6061161,
      "memory(GiB)": 117.38,
      "step": 62525,
      "train_speed(iter/s)": 1.637125
    },
    {
      "acc": 0.65738888,
      "epoch": 1.5862506341958396,
      "grad_norm": 5.34375,
      "learning_rate": 1.125129102314847e-06,
      "loss": 1.58573866,
      "memory(GiB)": 117.38,
      "step": 62530,
      "train_speed(iter/s)": 1.637138
    },
    {
      "acc": 0.63902454,
      "epoch": 1.5863774733637748,
      "grad_norm": 5.28125,
      "learning_rate": 1.124466464820429e-06,
      "loss": 1.64181061,
      "memory(GiB)": 117.38,
      "step": 62535,
      "train_speed(iter/s)": 1.637152
    },
    {
      "acc": 0.65840492,
      "epoch": 1.5865043125317098,
      "grad_norm": 6.65625,
      "learning_rate": 1.123803997789278e-06,
      "loss": 1.54912586,
      "memory(GiB)": 117.38,
      "step": 62540,
      "train_speed(iter/s)": 1.637164
    },
    {
      "acc": 0.65408926,
      "epoch": 1.586631151699645,
      "grad_norm": 5.59375,
      "learning_rate": 1.1231417012505342e-06,
      "loss": 1.61890259,
      "memory(GiB)": 117.38,
      "step": 62545,
      "train_speed(iter/s)": 1.637177
    },
    {
      "acc": 0.66086903,
      "epoch": 1.58675799086758,
      "grad_norm": 6.0625,
      "learning_rate": 1.1224795752333283e-06,
      "loss": 1.58828669,
      "memory(GiB)": 117.38,
      "step": 62550,
      "train_speed(iter/s)": 1.63719
    },
    {
      "acc": 0.66073389,
      "epoch": 1.586884830035515,
      "grad_norm": 6.03125,
      "learning_rate": 1.1218176197667802e-06,
      "loss": 1.65278244,
      "memory(GiB)": 117.38,
      "step": 62555,
      "train_speed(iter/s)": 1.637203
    },
    {
      "acc": 0.64921055,
      "epoch": 1.58701166920345,
      "grad_norm": 4.875,
      "learning_rate": 1.1211558348800095e-06,
      "loss": 1.60193367,
      "memory(GiB)": 117.38,
      "step": 62560,
      "train_speed(iter/s)": 1.637216
    },
    {
      "acc": 0.65964622,
      "epoch": 1.587138508371385,
      "grad_norm": 5.90625,
      "learning_rate": 1.120494220602123e-06,
      "loss": 1.56493664,
      "memory(GiB)": 117.38,
      "step": 62565,
      "train_speed(iter/s)": 1.637229
    },
    {
      "acc": 0.6770113,
      "epoch": 1.5872653475393201,
      "grad_norm": 7.375,
      "learning_rate": 1.1198327769622224e-06,
      "loss": 1.57329111,
      "memory(GiB)": 117.38,
      "step": 62570,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.65843763,
      "epoch": 1.5873921867072553,
      "grad_norm": 5.03125,
      "learning_rate": 1.1191715039893975e-06,
      "loss": 1.55712767,
      "memory(GiB)": 117.38,
      "step": 62575,
      "train_speed(iter/s)": 1.637256
    },
    {
      "acc": 0.65831566,
      "epoch": 1.5875190258751903,
      "grad_norm": 6.03125,
      "learning_rate": 1.1185104017127379e-06,
      "loss": 1.71317406,
      "memory(GiB)": 117.38,
      "step": 62580,
      "train_speed(iter/s)": 1.63727
    },
    {
      "acc": 0.6672946,
      "epoch": 1.5876458650431253,
      "grad_norm": 6.34375,
      "learning_rate": 1.1178494701613202e-06,
      "loss": 1.63364353,
      "memory(GiB)": 117.38,
      "step": 62585,
      "train_speed(iter/s)": 1.637282
    },
    {
      "acc": 0.66725416,
      "epoch": 1.5877727042110603,
      "grad_norm": 5.84375,
      "learning_rate": 1.1171887093642158e-06,
      "loss": 1.55805531,
      "memory(GiB)": 117.38,
      "step": 62590,
      "train_speed(iter/s)": 1.637296
    },
    {
      "acc": 0.68033695,
      "epoch": 1.5878995433789953,
      "grad_norm": 5.09375,
      "learning_rate": 1.1165281193504873e-06,
      "loss": 1.53453856,
      "memory(GiB)": 117.38,
      "step": 62595,
      "train_speed(iter/s)": 1.63731
    },
    {
      "acc": 0.64932861,
      "epoch": 1.5880263825469305,
      "grad_norm": 6.25,
      "learning_rate": 1.1158677001491902e-06,
      "loss": 1.6040556,
      "memory(GiB)": 117.38,
      "step": 62600,
      "train_speed(iter/s)": 1.637323
    },
    {
      "acc": 0.64266424,
      "epoch": 1.5881532217148655,
      "grad_norm": 6.90625,
      "learning_rate": 1.1152074517893735e-06,
      "loss": 1.69743309,
      "memory(GiB)": 117.38,
      "step": 62605,
      "train_speed(iter/s)": 1.637336
    },
    {
      "acc": 0.64480076,
      "epoch": 1.5882800608828007,
      "grad_norm": 4.8125,
      "learning_rate": 1.1145473743000773e-06,
      "loss": 1.67770576,
      "memory(GiB)": 117.38,
      "step": 62610,
      "train_speed(iter/s)": 1.63735
    },
    {
      "acc": 0.66311855,
      "epoch": 1.5884069000507357,
      "grad_norm": 6.5625,
      "learning_rate": 1.1138874677103345e-06,
      "loss": 1.55446396,
      "memory(GiB)": 117.38,
      "step": 62615,
      "train_speed(iter/s)": 1.637363
    },
    {
      "acc": 0.66786499,
      "epoch": 1.5885337392186707,
      "grad_norm": 5.875,
      "learning_rate": 1.1132277320491713e-06,
      "loss": 1.55443649,
      "memory(GiB)": 117.38,
      "step": 62620,
      "train_speed(iter/s)": 1.637377
    },
    {
      "acc": 0.64248624,
      "epoch": 1.5886605783866057,
      "grad_norm": 5.3125,
      "learning_rate": 1.1125681673456062e-06,
      "loss": 1.60297012,
      "memory(GiB)": 117.38,
      "step": 62625,
      "train_speed(iter/s)": 1.63739
    },
    {
      "acc": 0.6707387,
      "epoch": 1.588787417554541,
      "grad_norm": 6.65625,
      "learning_rate": 1.1119087736286489e-06,
      "loss": 1.56099186,
      "memory(GiB)": 117.38,
      "step": 62630,
      "train_speed(iter/s)": 1.637403
    },
    {
      "acc": 0.64204235,
      "epoch": 1.5889142567224759,
      "grad_norm": 5.375,
      "learning_rate": 1.1112495509273025e-06,
      "loss": 1.65730133,
      "memory(GiB)": 117.38,
      "step": 62635,
      "train_speed(iter/s)": 1.637417
    },
    {
      "acc": 0.65478725,
      "epoch": 1.589041095890411,
      "grad_norm": 5.53125,
      "learning_rate": 1.110590499270563e-06,
      "loss": 1.57113113,
      "memory(GiB)": 117.38,
      "step": 62640,
      "train_speed(iter/s)": 1.637432
    },
    {
      "acc": 0.66281033,
      "epoch": 1.589167935058346,
      "grad_norm": 5.375,
      "learning_rate": 1.109931618687417e-06,
      "loss": 1.57591667,
      "memory(GiB)": 117.38,
      "step": 62645,
      "train_speed(iter/s)": 1.637446
    },
    {
      "acc": 0.64522371,
      "epoch": 1.589294774226281,
      "grad_norm": 6.40625,
      "learning_rate": 1.1092729092068495e-06,
      "loss": 1.659021,
      "memory(GiB)": 117.38,
      "step": 62650,
      "train_speed(iter/s)": 1.637458
    },
    {
      "acc": 0.65676346,
      "epoch": 1.589421613394216,
      "grad_norm": 6.15625,
      "learning_rate": 1.1086143708578285e-06,
      "loss": 1.67405357,
      "memory(GiB)": 117.38,
      "step": 62655,
      "train_speed(iter/s)": 1.637471
    },
    {
      "acc": 0.66228676,
      "epoch": 1.589548452562151,
      "grad_norm": 8.9375,
      "learning_rate": 1.107956003669321e-06,
      "loss": 1.5373064,
      "memory(GiB)": 117.38,
      "step": 62660,
      "train_speed(iter/s)": 1.637484
    },
    {
      "acc": 0.66487112,
      "epoch": 1.5896752917300863,
      "grad_norm": 6.03125,
      "learning_rate": 1.107297807670284e-06,
      "loss": 1.58625841,
      "memory(GiB)": 117.38,
      "step": 62665,
      "train_speed(iter/s)": 1.637496
    },
    {
      "acc": 0.65883241,
      "epoch": 1.5898021308980215,
      "grad_norm": 6.21875,
      "learning_rate": 1.106639782889672e-06,
      "loss": 1.60930405,
      "memory(GiB)": 117.38,
      "step": 62670,
      "train_speed(iter/s)": 1.637509
    },
    {
      "acc": 0.66167426,
      "epoch": 1.5899289700659565,
      "grad_norm": 5.59375,
      "learning_rate": 1.1059819293564233e-06,
      "loss": 1.54043045,
      "memory(GiB)": 117.38,
      "step": 62675,
      "train_speed(iter/s)": 1.637524
    },
    {
      "acc": 0.65661821,
      "epoch": 1.5900558092338914,
      "grad_norm": 5.4375,
      "learning_rate": 1.105324247099474e-06,
      "loss": 1.61117821,
      "memory(GiB)": 117.38,
      "step": 62680,
      "train_speed(iter/s)": 1.637536
    },
    {
      "acc": 0.65390182,
      "epoch": 1.5901826484018264,
      "grad_norm": 5.15625,
      "learning_rate": 1.1046667361477537e-06,
      "loss": 1.53831892,
      "memory(GiB)": 117.38,
      "step": 62685,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.67002668,
      "epoch": 1.5903094875697614,
      "grad_norm": 6.21875,
      "learning_rate": 1.1040093965301835e-06,
      "loss": 1.56735458,
      "memory(GiB)": 117.38,
      "step": 62690,
      "train_speed(iter/s)": 1.637563
    },
    {
      "acc": 0.6513566,
      "epoch": 1.5904363267376966,
      "grad_norm": 6.25,
      "learning_rate": 1.1033522282756716e-06,
      "loss": 1.60883369,
      "memory(GiB)": 117.38,
      "step": 62695,
      "train_speed(iter/s)": 1.637577
    },
    {
      "acc": 0.66003828,
      "epoch": 1.5905631659056316,
      "grad_norm": 5.625,
      "learning_rate": 1.1026952314131268e-06,
      "loss": 1.58817825,
      "memory(GiB)": 117.38,
      "step": 62700,
      "train_speed(iter/s)": 1.63759
    },
    {
      "acc": 0.67945404,
      "epoch": 1.5906900050735668,
      "grad_norm": 4.90625,
      "learning_rate": 1.1020384059714463e-06,
      "loss": 1.52609806,
      "memory(GiB)": 117.38,
      "step": 62705,
      "train_speed(iter/s)": 1.637602
    },
    {
      "acc": 0.67456856,
      "epoch": 1.5908168442415018,
      "grad_norm": 5.4375,
      "learning_rate": 1.1013817519795211e-06,
      "loss": 1.5038702,
      "memory(GiB)": 117.38,
      "step": 62710,
      "train_speed(iter/s)": 1.637615
    },
    {
      "acc": 0.65819688,
      "epoch": 1.5909436834094368,
      "grad_norm": 6.34375,
      "learning_rate": 1.1007252694662302e-06,
      "loss": 1.62090111,
      "memory(GiB)": 117.38,
      "step": 62715,
      "train_speed(iter/s)": 1.637629
    },
    {
      "acc": 0.67165565,
      "epoch": 1.5910705225773718,
      "grad_norm": 6.28125,
      "learning_rate": 1.1000689584604519e-06,
      "loss": 1.50161076,
      "memory(GiB)": 117.38,
      "step": 62720,
      "train_speed(iter/s)": 1.637643
    },
    {
      "acc": 0.64866056,
      "epoch": 1.5911973617453068,
      "grad_norm": 5.8125,
      "learning_rate": 1.099412818991053e-06,
      "loss": 1.63522167,
      "memory(GiB)": 117.38,
      "step": 62725,
      "train_speed(iter/s)": 1.637656
    },
    {
      "acc": 0.66707015,
      "epoch": 1.591324200913242,
      "grad_norm": 6.71875,
      "learning_rate": 1.098756851086893e-06,
      "loss": 1.56834793,
      "memory(GiB)": 117.38,
      "step": 62730,
      "train_speed(iter/s)": 1.637668
    },
    {
      "acc": 0.64237709,
      "epoch": 1.5914510400811772,
      "grad_norm": 5.5,
      "learning_rate": 1.0981010547768244e-06,
      "loss": 1.66979275,
      "memory(GiB)": 117.38,
      "step": 62735,
      "train_speed(iter/s)": 1.637681
    },
    {
      "acc": 0.65112667,
      "epoch": 1.5915778792491122,
      "grad_norm": 5.78125,
      "learning_rate": 1.0974454300896924e-06,
      "loss": 1.63156357,
      "memory(GiB)": 117.38,
      "step": 62740,
      "train_speed(iter/s)": 1.637695
    },
    {
      "acc": 0.65356369,
      "epoch": 1.5917047184170472,
      "grad_norm": 5.6875,
      "learning_rate": 1.0967899770543344e-06,
      "loss": 1.68217621,
      "memory(GiB)": 117.38,
      "step": 62745,
      "train_speed(iter/s)": 1.637709
    },
    {
      "acc": 0.65835867,
      "epoch": 1.5918315575849822,
      "grad_norm": 6.3125,
      "learning_rate": 1.0961346956995795e-06,
      "loss": 1.62671337,
      "memory(GiB)": 117.38,
      "step": 62750,
      "train_speed(iter/s)": 1.637721
    },
    {
      "acc": 0.663344,
      "epoch": 1.5919583967529172,
      "grad_norm": 6.40625,
      "learning_rate": 1.0954795860542495e-06,
      "loss": 1.54263611,
      "memory(GiB)": 117.38,
      "step": 62755,
      "train_speed(iter/s)": 1.637736
    },
    {
      "acc": 0.64407768,
      "epoch": 1.5920852359208524,
      "grad_norm": 5.3125,
      "learning_rate": 1.0948246481471603e-06,
      "loss": 1.66214676,
      "memory(GiB)": 117.38,
      "step": 62760,
      "train_speed(iter/s)": 1.637749
    },
    {
      "acc": 0.66005678,
      "epoch": 1.5922120750887874,
      "grad_norm": 6.6875,
      "learning_rate": 1.0941698820071183e-06,
      "loss": 1.54358101,
      "memory(GiB)": 117.38,
      "step": 62765,
      "train_speed(iter/s)": 1.637762
    },
    {
      "acc": 0.65918379,
      "epoch": 1.5923389142567226,
      "grad_norm": 5.46875,
      "learning_rate": 1.0935152876629234e-06,
      "loss": 1.49937801,
      "memory(GiB)": 117.38,
      "step": 62770,
      "train_speed(iter/s)": 1.637776
    },
    {
      "acc": 0.66754518,
      "epoch": 1.5924657534246576,
      "grad_norm": 6.46875,
      "learning_rate": 1.0928608651433675e-06,
      "loss": 1.55539265,
      "memory(GiB)": 117.38,
      "step": 62775,
      "train_speed(iter/s)": 1.637789
    },
    {
      "acc": 0.64254384,
      "epoch": 1.5925925925925926,
      "grad_norm": 6.84375,
      "learning_rate": 1.0922066144772342e-06,
      "loss": 1.6642046,
      "memory(GiB)": 117.38,
      "step": 62780,
      "train_speed(iter/s)": 1.637804
    },
    {
      "acc": 0.65602951,
      "epoch": 1.5927194317605275,
      "grad_norm": 6.375,
      "learning_rate": 1.0915525356933004e-06,
      "loss": 1.67031956,
      "memory(GiB)": 117.38,
      "step": 62785,
      "train_speed(iter/s)": 1.637817
    },
    {
      "acc": 0.66519742,
      "epoch": 1.5928462709284628,
      "grad_norm": 6.1875,
      "learning_rate": 1.0908986288203382e-06,
      "loss": 1.60669537,
      "memory(GiB)": 117.38,
      "step": 62790,
      "train_speed(iter/s)": 1.63783
    },
    {
      "acc": 0.66883984,
      "epoch": 1.5929731100963977,
      "grad_norm": 5.53125,
      "learning_rate": 1.0902448938871064e-06,
      "loss": 1.52913857,
      "memory(GiB)": 117.38,
      "step": 62795,
      "train_speed(iter/s)": 1.637843
    },
    {
      "acc": 0.66484003,
      "epoch": 1.593099949264333,
      "grad_norm": 5.03125,
      "learning_rate": 1.0895913309223594e-06,
      "loss": 1.58305054,
      "memory(GiB)": 117.38,
      "step": 62800,
      "train_speed(iter/s)": 1.637856
    },
    {
      "acc": 0.66887884,
      "epoch": 1.593226788432268,
      "grad_norm": 5.84375,
      "learning_rate": 1.0889379399548432e-06,
      "loss": 1.54780502,
      "memory(GiB)": 117.38,
      "step": 62805,
      "train_speed(iter/s)": 1.637869
    },
    {
      "acc": 0.64801464,
      "epoch": 1.593353627600203,
      "grad_norm": 6.84375,
      "learning_rate": 1.0882847210133007e-06,
      "loss": 1.62212696,
      "memory(GiB)": 117.38,
      "step": 62810,
      "train_speed(iter/s)": 1.637882
    },
    {
      "acc": 0.66171775,
      "epoch": 1.593480466768138,
      "grad_norm": 4.96875,
      "learning_rate": 1.0876316741264598e-06,
      "loss": 1.59099846,
      "memory(GiB)": 117.38,
      "step": 62815,
      "train_speed(iter/s)": 1.637895
    },
    {
      "acc": 0.65383253,
      "epoch": 1.593607305936073,
      "grad_norm": 5.0,
      "learning_rate": 1.0869787993230435e-06,
      "loss": 1.66024075,
      "memory(GiB)": 117.38,
      "step": 62820,
      "train_speed(iter/s)": 1.637908
    },
    {
      "acc": 0.66324282,
      "epoch": 1.5937341451040081,
      "grad_norm": 5.40625,
      "learning_rate": 1.0863260966317713e-06,
      "loss": 1.57417755,
      "memory(GiB)": 117.38,
      "step": 62825,
      "train_speed(iter/s)": 1.637922
    },
    {
      "acc": 0.65766764,
      "epoch": 1.5938609842719433,
      "grad_norm": 4.90625,
      "learning_rate": 1.0856735660813523e-06,
      "loss": 1.55127659,
      "memory(GiB)": 117.38,
      "step": 62830,
      "train_speed(iter/s)": 1.637935
    },
    {
      "acc": 0.65404606,
      "epoch": 1.5939878234398783,
      "grad_norm": 5.09375,
      "learning_rate": 1.0850212077004845e-06,
      "loss": 1.6048912,
      "memory(GiB)": 117.38,
      "step": 62835,
      "train_speed(iter/s)": 1.637948
    },
    {
      "acc": 0.66554499,
      "epoch": 1.5941146626078133,
      "grad_norm": 5.5625,
      "learning_rate": 1.084369021517862e-06,
      "loss": 1.57158184,
      "memory(GiB)": 117.38,
      "step": 62840,
      "train_speed(iter/s)": 1.637962
    },
    {
      "acc": 0.66656065,
      "epoch": 1.5942415017757483,
      "grad_norm": 5.625,
      "learning_rate": 1.0837170075621733e-06,
      "loss": 1.56499786,
      "memory(GiB)": 117.38,
      "step": 62845,
      "train_speed(iter/s)": 1.637976
    },
    {
      "acc": 0.66513052,
      "epoch": 1.5943683409436833,
      "grad_norm": 5.6875,
      "learning_rate": 1.0830651658620966e-06,
      "loss": 1.56380625,
      "memory(GiB)": 117.38,
      "step": 62850,
      "train_speed(iter/s)": 1.637989
    },
    {
      "acc": 0.68336992,
      "epoch": 1.5944951801116185,
      "grad_norm": 4.96875,
      "learning_rate": 1.0824134964462996e-06,
      "loss": 1.49707603,
      "memory(GiB)": 117.38,
      "step": 62855,
      "train_speed(iter/s)": 1.638003
    },
    {
      "acc": 0.66165628,
      "epoch": 1.5946220192795535,
      "grad_norm": 5.03125,
      "learning_rate": 1.0817619993434486e-06,
      "loss": 1.5601078,
      "memory(GiB)": 117.38,
      "step": 62860,
      "train_speed(iter/s)": 1.638017
    },
    {
      "acc": 0.64193153,
      "epoch": 1.5947488584474887,
      "grad_norm": 5.28125,
      "learning_rate": 1.0811106745821987e-06,
      "loss": 1.6751091,
      "memory(GiB)": 117.38,
      "step": 62865,
      "train_speed(iter/s)": 1.638031
    },
    {
      "acc": 0.66296811,
      "epoch": 1.5948756976154237,
      "grad_norm": 5.375,
      "learning_rate": 1.0804595221911978e-06,
      "loss": 1.60402641,
      "memory(GiB)": 117.38,
      "step": 62870,
      "train_speed(iter/s)": 1.638045
    },
    {
      "acc": 0.65604777,
      "epoch": 1.5950025367833587,
      "grad_norm": 5.78125,
      "learning_rate": 1.0798085421990867e-06,
      "loss": 1.59422054,
      "memory(GiB)": 117.38,
      "step": 62875,
      "train_speed(iter/s)": 1.638059
    },
    {
      "acc": 0.67371092,
      "epoch": 1.5951293759512937,
      "grad_norm": 6.3125,
      "learning_rate": 1.079157734634499e-06,
      "loss": 1.55721741,
      "memory(GiB)": 117.38,
      "step": 62880,
      "train_speed(iter/s)": 1.638074
    },
    {
      "acc": 0.65964994,
      "epoch": 1.5952562151192287,
      "grad_norm": 4.28125,
      "learning_rate": 1.0785070995260582e-06,
      "loss": 1.63469543,
      "memory(GiB)": 117.38,
      "step": 62885,
      "train_speed(iter/s)": 1.638087
    },
    {
      "acc": 0.67139902,
      "epoch": 1.5953830542871639,
      "grad_norm": 5.1875,
      "learning_rate": 1.0778566369023841e-06,
      "loss": 1.56823635,
      "memory(GiB)": 117.38,
      "step": 62890,
      "train_speed(iter/s)": 1.6381
    },
    {
      "acc": 0.6539577,
      "epoch": 1.595509893455099,
      "grad_norm": 5.125,
      "learning_rate": 1.0772063467920863e-06,
      "loss": 1.59358273,
      "memory(GiB)": 117.38,
      "step": 62895,
      "train_speed(iter/s)": 1.638113
    },
    {
      "acc": 0.65581532,
      "epoch": 1.595636732623034,
      "grad_norm": 5.75,
      "learning_rate": 1.076556229223767e-06,
      "loss": 1.61210937,
      "memory(GiB)": 117.38,
      "step": 62900,
      "train_speed(iter/s)": 1.638127
    },
    {
      "acc": 0.67634935,
      "epoch": 1.595763571790969,
      "grad_norm": 5.90625,
      "learning_rate": 1.0759062842260221e-06,
      "loss": 1.54376431,
      "memory(GiB)": 117.38,
      "step": 62905,
      "train_speed(iter/s)": 1.63814
    },
    {
      "acc": 0.66490984,
      "epoch": 1.595890410958904,
      "grad_norm": 6.15625,
      "learning_rate": 1.0752565118274383e-06,
      "loss": 1.62718925,
      "memory(GiB)": 117.38,
      "step": 62910,
      "train_speed(iter/s)": 1.638154
    },
    {
      "acc": 0.66052189,
      "epoch": 1.596017250126839,
      "grad_norm": 8.3125,
      "learning_rate": 1.0746069120565961e-06,
      "loss": 1.61135654,
      "memory(GiB)": 117.38,
      "step": 62915,
      "train_speed(iter/s)": 1.638166
    },
    {
      "acc": 0.67061701,
      "epoch": 1.5961440892947742,
      "grad_norm": 5.59375,
      "learning_rate": 1.073957484942067e-06,
      "loss": 1.55945606,
      "memory(GiB)": 117.38,
      "step": 62920,
      "train_speed(iter/s)": 1.638179
    },
    {
      "acc": 0.65347948,
      "epoch": 1.5962709284627092,
      "grad_norm": 5.1875,
      "learning_rate": 1.0733082305124166e-06,
      "loss": 1.58091908,
      "memory(GiB)": 117.38,
      "step": 62925,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.65152078,
      "epoch": 1.5963977676306444,
      "grad_norm": 6.40625,
      "learning_rate": 1.0726591487962018e-06,
      "loss": 1.61406746,
      "memory(GiB)": 117.38,
      "step": 62930,
      "train_speed(iter/s)": 1.638207
    },
    {
      "acc": 0.65219269,
      "epoch": 1.5965246067985794,
      "grad_norm": 6.0625,
      "learning_rate": 1.0720102398219716e-06,
      "loss": 1.61286201,
      "memory(GiB)": 117.38,
      "step": 62935,
      "train_speed(iter/s)": 1.638222
    },
    {
      "acc": 0.6677989,
      "epoch": 1.5966514459665144,
      "grad_norm": 6.0625,
      "learning_rate": 1.0713615036182684e-06,
      "loss": 1.56803465,
      "memory(GiB)": 117.38,
      "step": 62940,
      "train_speed(iter/s)": 1.638236
    },
    {
      "acc": 0.66225033,
      "epoch": 1.5967782851344494,
      "grad_norm": 7.0625,
      "learning_rate": 1.0707129402136252e-06,
      "loss": 1.59018974,
      "memory(GiB)": 117.38,
      "step": 62945,
      "train_speed(iter/s)": 1.638248
    },
    {
      "acc": 0.65961542,
      "epoch": 1.5969051243023846,
      "grad_norm": 6.40625,
      "learning_rate": 1.0700645496365725e-06,
      "loss": 1.57412443,
      "memory(GiB)": 117.38,
      "step": 62950,
      "train_speed(iter/s)": 1.638262
    },
    {
      "acc": 0.65433278,
      "epoch": 1.5970319634703196,
      "grad_norm": 5.28125,
      "learning_rate": 1.0694163319156254e-06,
      "loss": 1.58099747,
      "memory(GiB)": 117.38,
      "step": 62955,
      "train_speed(iter/s)": 1.638275
    },
    {
      "acc": 0.66535811,
      "epoch": 1.5971588026382548,
      "grad_norm": 5.625,
      "learning_rate": 1.0687682870792953e-06,
      "loss": 1.55464916,
      "memory(GiB)": 117.38,
      "step": 62960,
      "train_speed(iter/s)": 1.638288
    },
    {
      "acc": 0.64707308,
      "epoch": 1.5972856418061898,
      "grad_norm": 7.65625,
      "learning_rate": 1.0681204151560891e-06,
      "loss": 1.60858459,
      "memory(GiB)": 117.38,
      "step": 62965,
      "train_speed(iter/s)": 1.638302
    },
    {
      "acc": 0.64534931,
      "epoch": 1.5974124809741248,
      "grad_norm": 5.625,
      "learning_rate": 1.067472716174503e-06,
      "loss": 1.64561844,
      "memory(GiB)": 117.38,
      "step": 62970,
      "train_speed(iter/s)": 1.638315
    },
    {
      "acc": 0.65165234,
      "epoch": 1.5975393201420598,
      "grad_norm": 5.9375,
      "learning_rate": 1.0668251901630227e-06,
      "loss": 1.6185854,
      "memory(GiB)": 117.38,
      "step": 62975,
      "train_speed(iter/s)": 1.638328
    },
    {
      "acc": 0.67528396,
      "epoch": 1.5976661593099948,
      "grad_norm": 6.65625,
      "learning_rate": 1.0661778371501303e-06,
      "loss": 1.55878162,
      "memory(GiB)": 117.38,
      "step": 62980,
      "train_speed(iter/s)": 1.638342
    },
    {
      "acc": 0.67354817,
      "epoch": 1.59779299847793,
      "grad_norm": 6.84375,
      "learning_rate": 1.0655306571643004e-06,
      "loss": 1.61317043,
      "memory(GiB)": 117.38,
      "step": 62985,
      "train_speed(iter/s)": 1.638355
    },
    {
      "acc": 0.63953276,
      "epoch": 1.5979198376458652,
      "grad_norm": 5.9375,
      "learning_rate": 1.0648836502339998e-06,
      "loss": 1.66579113,
      "memory(GiB)": 117.38,
      "step": 62990,
      "train_speed(iter/s)": 1.638369
    },
    {
      "acc": 0.65304585,
      "epoch": 1.5980466768138002,
      "grad_norm": 5.90625,
      "learning_rate": 1.0642368163876832e-06,
      "loss": 1.58958511,
      "memory(GiB)": 117.38,
      "step": 62995,
      "train_speed(iter/s)": 1.638382
    },
    {
      "acc": 0.64993043,
      "epoch": 1.5981735159817352,
      "grad_norm": 5.21875,
      "learning_rate": 1.0635901556538042e-06,
      "loss": 1.58726244,
      "memory(GiB)": 117.38,
      "step": 63000,
      "train_speed(iter/s)": 1.638394
    },
    {
      "epoch": 1.5981735159817352,
      "eval_acc": 0.6463198825536642,
      "eval_loss": 1.5732152462005615,
      "eval_runtime": 58.4796,
      "eval_samples_per_second": 108.927,
      "eval_steps_per_second": 27.24,
      "step": 63000
    },
    {
      "acc": 0.67321014,
      "epoch": 1.5983003551496702,
      "grad_norm": 5.1875,
      "learning_rate": 1.0629436680608051e-06,
      "loss": 1.52636433,
      "memory(GiB)": 117.38,
      "step": 63005,
      "train_speed(iter/s)": 1.635737
    },
    {
      "acc": 0.66551385,
      "epoch": 1.5984271943176052,
      "grad_norm": 5.375,
      "learning_rate": 1.0622973536371223e-06,
      "loss": 1.53802767,
      "memory(GiB)": 117.38,
      "step": 63010,
      "train_speed(iter/s)": 1.63575
    },
    {
      "acc": 0.66202545,
      "epoch": 1.5985540334855404,
      "grad_norm": 5.84375,
      "learning_rate": 1.06165121241118e-06,
      "loss": 1.53575535,
      "memory(GiB)": 117.38,
      "step": 63015,
      "train_speed(iter/s)": 1.635764
    },
    {
      "acc": 0.65583754,
      "epoch": 1.5986808726534754,
      "grad_norm": 6.25,
      "learning_rate": 1.0610052444114023e-06,
      "loss": 1.59028873,
      "memory(GiB)": 117.38,
      "step": 63020,
      "train_speed(iter/s)": 1.635778
    },
    {
      "acc": 0.64928775,
      "epoch": 1.5988077118214106,
      "grad_norm": 6.375,
      "learning_rate": 1.0603594496662001e-06,
      "loss": 1.64033756,
      "memory(GiB)": 117.38,
      "step": 63025,
      "train_speed(iter/s)": 1.635791
    },
    {
      "acc": 0.6517415,
      "epoch": 1.5989345509893456,
      "grad_norm": 6.1875,
      "learning_rate": 1.0597138282039786e-06,
      "loss": 1.55655165,
      "memory(GiB)": 117.38,
      "step": 63030,
      "train_speed(iter/s)": 1.635804
    },
    {
      "acc": 0.66357374,
      "epoch": 1.5990613901572805,
      "grad_norm": 5.5625,
      "learning_rate": 1.0590683800531348e-06,
      "loss": 1.53634586,
      "memory(GiB)": 117.38,
      "step": 63035,
      "train_speed(iter/s)": 1.635818
    },
    {
      "acc": 0.66937804,
      "epoch": 1.5991882293252155,
      "grad_norm": 5.53125,
      "learning_rate": 1.058423105242059e-06,
      "loss": 1.56521683,
      "memory(GiB)": 117.38,
      "step": 63040,
      "train_speed(iter/s)": 1.635831
    },
    {
      "acc": 0.66186819,
      "epoch": 1.5993150684931505,
      "grad_norm": 5.90625,
      "learning_rate": 1.057778003799133e-06,
      "loss": 1.59319096,
      "memory(GiB)": 117.38,
      "step": 63045,
      "train_speed(iter/s)": 1.635843
    },
    {
      "acc": 0.65413027,
      "epoch": 1.5994419076610857,
      "grad_norm": 5.875,
      "learning_rate": 1.0571330757527309e-06,
      "loss": 1.63452091,
      "memory(GiB)": 117.38,
      "step": 63050,
      "train_speed(iter/s)": 1.635856
    },
    {
      "acc": 0.64303432,
      "epoch": 1.599568746829021,
      "grad_norm": 5.8125,
      "learning_rate": 1.0564883211312199e-06,
      "loss": 1.68220978,
      "memory(GiB)": 117.38,
      "step": 63055,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.65824366,
      "epoch": 1.599695585996956,
      "grad_norm": 6.28125,
      "learning_rate": 1.0558437399629584e-06,
      "loss": 1.58481245,
      "memory(GiB)": 117.38,
      "step": 63060,
      "train_speed(iter/s)": 1.635884
    },
    {
      "acc": 0.67312307,
      "epoch": 1.599822425164891,
      "grad_norm": 8.9375,
      "learning_rate": 1.0551993322762994e-06,
      "loss": 1.59917183,
      "memory(GiB)": 117.38,
      "step": 63065,
      "train_speed(iter/s)": 1.635897
    },
    {
      "acc": 0.65208092,
      "epoch": 1.599949264332826,
      "grad_norm": 6.28125,
      "learning_rate": 1.0545550980995857e-06,
      "loss": 1.56832161,
      "memory(GiB)": 117.38,
      "step": 63070,
      "train_speed(iter/s)": 1.63591
    },
    {
      "acc": 0.67117381,
      "epoch": 1.600076103500761,
      "grad_norm": 5.96875,
      "learning_rate": 1.0539110374611538e-06,
      "loss": 1.56425953,
      "memory(GiB)": 117.38,
      "step": 63075,
      "train_speed(iter/s)": 1.635923
    },
    {
      "acc": 0.66582322,
      "epoch": 1.6002029426686961,
      "grad_norm": 5.84375,
      "learning_rate": 1.0532671503893328e-06,
      "loss": 1.51314793,
      "memory(GiB)": 117.38,
      "step": 63080,
      "train_speed(iter/s)": 1.635936
    },
    {
      "acc": 0.65818939,
      "epoch": 1.600329781836631,
      "grad_norm": 7.0625,
      "learning_rate": 1.052623436912442e-06,
      "loss": 1.55100384,
      "memory(GiB)": 117.38,
      "step": 63085,
      "train_speed(iter/s)": 1.63595
    },
    {
      "acc": 0.65439148,
      "epoch": 1.6004566210045663,
      "grad_norm": 5.78125,
      "learning_rate": 1.0519798970587992e-06,
      "loss": 1.58639488,
      "memory(GiB)": 117.38,
      "step": 63090,
      "train_speed(iter/s)": 1.635963
    },
    {
      "acc": 0.64888353,
      "epoch": 1.6005834601725013,
      "grad_norm": 5.1875,
      "learning_rate": 1.0513365308567054e-06,
      "loss": 1.61622467,
      "memory(GiB)": 117.38,
      "step": 63095,
      "train_speed(iter/s)": 1.635978
    },
    {
      "acc": 0.64919519,
      "epoch": 1.6007102993404363,
      "grad_norm": 4.9375,
      "learning_rate": 1.0506933383344602e-06,
      "loss": 1.64282627,
      "memory(GiB)": 117.38,
      "step": 63100,
      "train_speed(iter/s)": 1.635991
    },
    {
      "acc": 0.65175028,
      "epoch": 1.6008371385083713,
      "grad_norm": 5.15625,
      "learning_rate": 1.0500503195203537e-06,
      "loss": 1.62532806,
      "memory(GiB)": 117.38,
      "step": 63105,
      "train_speed(iter/s)": 1.636002
    },
    {
      "acc": 0.65444393,
      "epoch": 1.6009639776763065,
      "grad_norm": 5.1875,
      "learning_rate": 1.049407474442672e-06,
      "loss": 1.62911377,
      "memory(GiB)": 117.38,
      "step": 63110,
      "train_speed(iter/s)": 1.636016
    },
    {
      "acc": 0.64304824,
      "epoch": 1.6010908168442415,
      "grad_norm": 5.59375,
      "learning_rate": 1.048764803129686e-06,
      "loss": 1.64158211,
      "memory(GiB)": 117.38,
      "step": 63115,
      "train_speed(iter/s)": 1.636028
    },
    {
      "acc": 0.67096558,
      "epoch": 1.6012176560121767,
      "grad_norm": 6.28125,
      "learning_rate": 1.0481223056096635e-06,
      "loss": 1.54007549,
      "memory(GiB)": 117.38,
      "step": 63120,
      "train_speed(iter/s)": 1.636042
    },
    {
      "acc": 0.65090814,
      "epoch": 1.6013444951801117,
      "grad_norm": 5.875,
      "learning_rate": 1.0474799819108677e-06,
      "loss": 1.61761513,
      "memory(GiB)": 117.38,
      "step": 63125,
      "train_speed(iter/s)": 1.636054
    },
    {
      "acc": 0.65764327,
      "epoch": 1.6014713343480467,
      "grad_norm": 6.4375,
      "learning_rate": 1.0468378320615502e-06,
      "loss": 1.57159176,
      "memory(GiB)": 117.38,
      "step": 63130,
      "train_speed(iter/s)": 1.636068
    },
    {
      "acc": 0.65537605,
      "epoch": 1.6015981735159817,
      "grad_norm": 5.5,
      "learning_rate": 1.0461958560899516e-06,
      "loss": 1.60358562,
      "memory(GiB)": 117.38,
      "step": 63135,
      "train_speed(iter/s)": 1.636081
    },
    {
      "acc": 0.66216707,
      "epoch": 1.6017250126839166,
      "grad_norm": 5.875,
      "learning_rate": 1.045554054024313e-06,
      "loss": 1.53025026,
      "memory(GiB)": 117.38,
      "step": 63140,
      "train_speed(iter/s)": 1.636095
    },
    {
      "acc": 0.65361862,
      "epoch": 1.6018518518518519,
      "grad_norm": 5.6875,
      "learning_rate": 1.0449124258928627e-06,
      "loss": 1.59925518,
      "memory(GiB)": 117.38,
      "step": 63145,
      "train_speed(iter/s)": 1.636109
    },
    {
      "acc": 0.65425138,
      "epoch": 1.601978691019787,
      "grad_norm": 6.125,
      "learning_rate": 1.044270971723823e-06,
      "loss": 1.58673649,
      "memory(GiB)": 117.38,
      "step": 63150,
      "train_speed(iter/s)": 1.636122
    },
    {
      "acc": 0.65936747,
      "epoch": 1.602105530187722,
      "grad_norm": 6.875,
      "learning_rate": 1.0436296915454048e-06,
      "loss": 1.60257912,
      "memory(GiB)": 117.38,
      "step": 63155,
      "train_speed(iter/s)": 1.636135
    },
    {
      "acc": 0.64741063,
      "epoch": 1.602232369355657,
      "grad_norm": 6.25,
      "learning_rate": 1.042988585385818e-06,
      "loss": 1.65389118,
      "memory(GiB)": 117.38,
      "step": 63160,
      "train_speed(iter/s)": 1.636149
    },
    {
      "acc": 0.66876011,
      "epoch": 1.602359208523592,
      "grad_norm": 5.9375,
      "learning_rate": 1.0423476532732602e-06,
      "loss": 1.5550148,
      "memory(GiB)": 117.38,
      "step": 63165,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.65169382,
      "epoch": 1.602486047691527,
      "grad_norm": 7.375,
      "learning_rate": 1.0417068952359216e-06,
      "loss": 1.58853741,
      "memory(GiB)": 117.38,
      "step": 63170,
      "train_speed(iter/s)": 1.636176
    },
    {
      "acc": 0.68113117,
      "epoch": 1.6026128868594622,
      "grad_norm": 4.65625,
      "learning_rate": 1.0410663113019869e-06,
      "loss": 1.53986454,
      "memory(GiB)": 117.38,
      "step": 63175,
      "train_speed(iter/s)": 1.636191
    },
    {
      "acc": 0.66872358,
      "epoch": 1.6027397260273972,
      "grad_norm": 5.53125,
      "learning_rate": 1.040425901499631e-06,
      "loss": 1.57916679,
      "memory(GiB)": 117.38,
      "step": 63180,
      "train_speed(iter/s)": 1.636204
    },
    {
      "acc": 0.67065878,
      "epoch": 1.6028665651953324,
      "grad_norm": 6.375,
      "learning_rate": 1.0397856658570216e-06,
      "loss": 1.52845459,
      "memory(GiB)": 117.38,
      "step": 63185,
      "train_speed(iter/s)": 1.636218
    },
    {
      "acc": 0.65603676,
      "epoch": 1.6029934043632674,
      "grad_norm": 6.875,
      "learning_rate": 1.03914560440232e-06,
      "loss": 1.60039139,
      "memory(GiB)": 117.38,
      "step": 63190,
      "train_speed(iter/s)": 1.636232
    },
    {
      "acc": 0.64850883,
      "epoch": 1.6031202435312024,
      "grad_norm": 5.84375,
      "learning_rate": 1.0385057171636791e-06,
      "loss": 1.64557247,
      "memory(GiB)": 117.38,
      "step": 63195,
      "train_speed(iter/s)": 1.636245
    },
    {
      "acc": 0.67557688,
      "epoch": 1.6032470826991374,
      "grad_norm": 5.71875,
      "learning_rate": 1.0378660041692435e-06,
      "loss": 1.58251038,
      "memory(GiB)": 117.38,
      "step": 63200,
      "train_speed(iter/s)": 1.63626
    },
    {
      "acc": 0.66818404,
      "epoch": 1.6033739218670724,
      "grad_norm": 5.03125,
      "learning_rate": 1.0372264654471504e-06,
      "loss": 1.54091148,
      "memory(GiB)": 117.38,
      "step": 63205,
      "train_speed(iter/s)": 1.636273
    },
    {
      "acc": 0.66613693,
      "epoch": 1.6035007610350076,
      "grad_norm": 5.71875,
      "learning_rate": 1.0365871010255303e-06,
      "loss": 1.61775627,
      "memory(GiB)": 117.38,
      "step": 63210,
      "train_speed(iter/s)": 1.636286
    },
    {
      "acc": 0.65165963,
      "epoch": 1.6036276002029428,
      "grad_norm": 4.90625,
      "learning_rate": 1.0359479109325043e-06,
      "loss": 1.61288433,
      "memory(GiB)": 117.38,
      "step": 63215,
      "train_speed(iter/s)": 1.6363
    },
    {
      "acc": 0.67193794,
      "epoch": 1.6037544393708778,
      "grad_norm": 6.375,
      "learning_rate": 1.0353088951961877e-06,
      "loss": 1.5572052,
      "memory(GiB)": 117.38,
      "step": 63220,
      "train_speed(iter/s)": 1.636313
    },
    {
      "acc": 0.66676292,
      "epoch": 1.6038812785388128,
      "grad_norm": 5.8125,
      "learning_rate": 1.0346700538446853e-06,
      "loss": 1.55466948,
      "memory(GiB)": 117.38,
      "step": 63225,
      "train_speed(iter/s)": 1.636327
    },
    {
      "acc": 0.66083107,
      "epoch": 1.6040081177067478,
      "grad_norm": 5.34375,
      "learning_rate": 1.034031386906101e-06,
      "loss": 1.55668364,
      "memory(GiB)": 117.38,
      "step": 63230,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65881033,
      "epoch": 1.6041349568746828,
      "grad_norm": 4.40625,
      "learning_rate": 1.0333928944085214e-06,
      "loss": 1.57668314,
      "memory(GiB)": 117.38,
      "step": 63235,
      "train_speed(iter/s)": 1.636354
    },
    {
      "acc": 0.65985451,
      "epoch": 1.604261796042618,
      "grad_norm": 6.21875,
      "learning_rate": 1.0327545763800322e-06,
      "loss": 1.60206566,
      "memory(GiB)": 117.38,
      "step": 63240,
      "train_speed(iter/s)": 1.636368
    },
    {
      "acc": 0.66663475,
      "epoch": 1.604388635210553,
      "grad_norm": 12.8125,
      "learning_rate": 1.0321164328487077e-06,
      "loss": 1.5337532,
      "memory(GiB)": 117.38,
      "step": 63245,
      "train_speed(iter/s)": 1.636381
    },
    {
      "acc": 0.64081488,
      "epoch": 1.6045154743784882,
      "grad_norm": 5.46875,
      "learning_rate": 1.0314784638426207e-06,
      "loss": 1.62483768,
      "memory(GiB)": 117.38,
      "step": 63250,
      "train_speed(iter/s)": 1.636394
    },
    {
      "acc": 0.65010276,
      "epoch": 1.6046423135464232,
      "grad_norm": 5.5,
      "learning_rate": 1.030840669389827e-06,
      "loss": 1.60685387,
      "memory(GiB)": 117.38,
      "step": 63255,
      "train_speed(iter/s)": 1.636408
    },
    {
      "acc": 0.65588932,
      "epoch": 1.6047691527143582,
      "grad_norm": 6.34375,
      "learning_rate": 1.0302030495183812e-06,
      "loss": 1.61661301,
      "memory(GiB)": 117.38,
      "step": 63260,
      "train_speed(iter/s)": 1.636421
    },
    {
      "acc": 0.65224919,
      "epoch": 1.6048959918822931,
      "grad_norm": 6.15625,
      "learning_rate": 1.0295656042563302e-06,
      "loss": 1.58130646,
      "memory(GiB)": 117.38,
      "step": 63265,
      "train_speed(iter/s)": 1.636433
    },
    {
      "acc": 0.65845861,
      "epoch": 1.6050228310502284,
      "grad_norm": 5.34375,
      "learning_rate": 1.0289283336317119e-06,
      "loss": 1.59415398,
      "memory(GiB)": 117.38,
      "step": 63270,
      "train_speed(iter/s)": 1.636448
    },
    {
      "acc": 0.66230006,
      "epoch": 1.6051496702181633,
      "grad_norm": 5.84375,
      "learning_rate": 1.0282912376725535e-06,
      "loss": 1.62725029,
      "memory(GiB)": 117.38,
      "step": 63275,
      "train_speed(iter/s)": 1.636463
    },
    {
      "acc": 0.65525117,
      "epoch": 1.6052765093860986,
      "grad_norm": 5.71875,
      "learning_rate": 1.0276543164068776e-06,
      "loss": 1.63565178,
      "memory(GiB)": 117.38,
      "step": 63280,
      "train_speed(iter/s)": 1.636477
    },
    {
      "acc": 0.66614847,
      "epoch": 1.6054033485540335,
      "grad_norm": 6.40625,
      "learning_rate": 1.0270175698627015e-06,
      "loss": 1.54426994,
      "memory(GiB)": 117.38,
      "step": 63285,
      "train_speed(iter/s)": 1.636491
    },
    {
      "acc": 0.65716567,
      "epoch": 1.6055301877219685,
      "grad_norm": 5.6875,
      "learning_rate": 1.0263809980680323e-06,
      "loss": 1.57237215,
      "memory(GiB)": 117.38,
      "step": 63290,
      "train_speed(iter/s)": 1.636504
    },
    {
      "acc": 0.65186343,
      "epoch": 1.6056570268899035,
      "grad_norm": 6.4375,
      "learning_rate": 1.0257446010508648e-06,
      "loss": 1.56913452,
      "memory(GiB)": 117.38,
      "step": 63295,
      "train_speed(iter/s)": 1.636519
    },
    {
      "acc": 0.64145284,
      "epoch": 1.6057838660578385,
      "grad_norm": 5.46875,
      "learning_rate": 1.0251083788391952e-06,
      "loss": 1.75254288,
      "memory(GiB)": 117.38,
      "step": 63300,
      "train_speed(iter/s)": 1.636533
    },
    {
      "acc": 0.64904003,
      "epoch": 1.6059107052257737,
      "grad_norm": 6.84375,
      "learning_rate": 1.0244723314610055e-06,
      "loss": 1.60689201,
      "memory(GiB)": 117.38,
      "step": 63305,
      "train_speed(iter/s)": 1.636547
    },
    {
      "acc": 0.65746627,
      "epoch": 1.606037544393709,
      "grad_norm": 5.0,
      "learning_rate": 1.0238364589442723e-06,
      "loss": 1.67799797,
      "memory(GiB)": 117.38,
      "step": 63310,
      "train_speed(iter/s)": 1.63656
    },
    {
      "acc": 0.66065416,
      "epoch": 1.606164383561644,
      "grad_norm": 5.65625,
      "learning_rate": 1.0232007613169637e-06,
      "loss": 1.62901859,
      "memory(GiB)": 117.38,
      "step": 63315,
      "train_speed(iter/s)": 1.636575
    },
    {
      "acc": 0.66691303,
      "epoch": 1.606291222729579,
      "grad_norm": 7.625,
      "learning_rate": 1.0225652386070406e-06,
      "loss": 1.51083708,
      "memory(GiB)": 117.38,
      "step": 63320,
      "train_speed(iter/s)": 1.636587
    },
    {
      "acc": 0.65998182,
      "epoch": 1.606418061897514,
      "grad_norm": 6.59375,
      "learning_rate": 1.0219298908424568e-06,
      "loss": 1.60301971,
      "memory(GiB)": 117.38,
      "step": 63325,
      "train_speed(iter/s)": 1.636601
    },
    {
      "acc": 0.65262203,
      "epoch": 1.606544901065449,
      "grad_norm": 7.28125,
      "learning_rate": 1.0212947180511567e-06,
      "loss": 1.6119339,
      "memory(GiB)": 117.38,
      "step": 63330,
      "train_speed(iter/s)": 1.636615
    },
    {
      "acc": 0.6560483,
      "epoch": 1.606671740233384,
      "grad_norm": 6.09375,
      "learning_rate": 1.020659720261079e-06,
      "loss": 1.63825474,
      "memory(GiB)": 117.38,
      "step": 63335,
      "train_speed(iter/s)": 1.636629
    },
    {
      "acc": 0.66190338,
      "epoch": 1.606798579401319,
      "grad_norm": 5.46875,
      "learning_rate": 1.020024897500153e-06,
      "loss": 1.58186569,
      "memory(GiB)": 117.38,
      "step": 63340,
      "train_speed(iter/s)": 1.636643
    },
    {
      "acc": 0.66282234,
      "epoch": 1.6069254185692543,
      "grad_norm": 5.625,
      "learning_rate": 1.0193902497963014e-06,
      "loss": 1.55005302,
      "memory(GiB)": 117.38,
      "step": 63345,
      "train_speed(iter/s)": 1.636657
    },
    {
      "acc": 0.66492939,
      "epoch": 1.6070522577371893,
      "grad_norm": 7.4375,
      "learning_rate": 1.0187557771774387e-06,
      "loss": 1.61549568,
      "memory(GiB)": 117.38,
      "step": 63350,
      "train_speed(iter/s)": 1.636671
    },
    {
      "acc": 0.65569873,
      "epoch": 1.6071790969051243,
      "grad_norm": 5.84375,
      "learning_rate": 1.0181214796714717e-06,
      "loss": 1.66181927,
      "memory(GiB)": 117.38,
      "step": 63355,
      "train_speed(iter/s)": 1.636684
    },
    {
      "acc": 0.65417132,
      "epoch": 1.6073059360730593,
      "grad_norm": 5.0,
      "learning_rate": 1.0174873573062998e-06,
      "loss": 1.67353935,
      "memory(GiB)": 117.38,
      "step": 63360,
      "train_speed(iter/s)": 1.636697
    },
    {
      "acc": 0.66234956,
      "epoch": 1.6074327752409943,
      "grad_norm": 6.3125,
      "learning_rate": 1.0168534101098148e-06,
      "loss": 1.59113045,
      "memory(GiB)": 117.38,
      "step": 63365,
      "train_speed(iter/s)": 1.636711
    },
    {
      "acc": 0.6557631,
      "epoch": 1.6075596144089295,
      "grad_norm": 5.40625,
      "learning_rate": 1.0162196381099004e-06,
      "loss": 1.61100235,
      "memory(GiB)": 117.38,
      "step": 63370,
      "train_speed(iter/s)": 1.636725
    },
    {
      "acc": 0.66415567,
      "epoch": 1.6076864535768647,
      "grad_norm": 6.09375,
      "learning_rate": 1.0155860413344327e-06,
      "loss": 1.57860041,
      "memory(GiB)": 117.38,
      "step": 63375,
      "train_speed(iter/s)": 1.636738
    },
    {
      "acc": 0.66326399,
      "epoch": 1.6078132927447997,
      "grad_norm": 6.53125,
      "learning_rate": 1.0149526198112797e-06,
      "loss": 1.55059662,
      "memory(GiB)": 117.38,
      "step": 63380,
      "train_speed(iter/s)": 1.636751
    },
    {
      "acc": 0.66914272,
      "epoch": 1.6079401319127347,
      "grad_norm": 5.3125,
      "learning_rate": 1.0143193735683016e-06,
      "loss": 1.58386917,
      "memory(GiB)": 117.38,
      "step": 63385,
      "train_speed(iter/s)": 1.636764
    },
    {
      "acc": 0.6591526,
      "epoch": 1.6080669710806696,
      "grad_norm": 6.40625,
      "learning_rate": 1.0136863026333543e-06,
      "loss": 1.60580235,
      "memory(GiB)": 117.38,
      "step": 63390,
      "train_speed(iter/s)": 1.636778
    },
    {
      "acc": 0.65191479,
      "epoch": 1.6081938102486046,
      "grad_norm": 5.65625,
      "learning_rate": 1.0130534070342802e-06,
      "loss": 1.62703438,
      "memory(GiB)": 117.38,
      "step": 63395,
      "train_speed(iter/s)": 1.636791
    },
    {
      "acc": 0.66894474,
      "epoch": 1.6083206494165398,
      "grad_norm": 6.09375,
      "learning_rate": 1.0124206867989157e-06,
      "loss": 1.61279545,
      "memory(GiB)": 117.38,
      "step": 63400,
      "train_speed(iter/s)": 1.636804
    },
    {
      "acc": 0.64818106,
      "epoch": 1.6084474885844748,
      "grad_norm": 5.1875,
      "learning_rate": 1.0117881419550945e-06,
      "loss": 1.62728996,
      "memory(GiB)": 117.38,
      "step": 63405,
      "train_speed(iter/s)": 1.636818
    },
    {
      "acc": 0.66701317,
      "epoch": 1.60857432775241,
      "grad_norm": 6.125,
      "learning_rate": 1.0111557725306382e-06,
      "loss": 1.57233353,
      "memory(GiB)": 117.38,
      "step": 63410,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.63460493,
      "epoch": 1.608701166920345,
      "grad_norm": 6.4375,
      "learning_rate": 1.0105235785533585e-06,
      "loss": 1.6997776,
      "memory(GiB)": 117.38,
      "step": 63415,
      "train_speed(iter/s)": 1.636844
    },
    {
      "acc": 0.64710674,
      "epoch": 1.60882800608828,
      "grad_norm": 5.8125,
      "learning_rate": 1.0098915600510623e-06,
      "loss": 1.64802704,
      "memory(GiB)": 117.38,
      "step": 63420,
      "train_speed(iter/s)": 1.636858
    },
    {
      "acc": 0.66063366,
      "epoch": 1.608954845256215,
      "grad_norm": 9.5,
      "learning_rate": 1.0092597170515512e-06,
      "loss": 1.55857086,
      "memory(GiB)": 117.38,
      "step": 63425,
      "train_speed(iter/s)": 1.636872
    },
    {
      "acc": 0.65296106,
      "epoch": 1.6090816844241502,
      "grad_norm": 5.25,
      "learning_rate": 1.0086280495826161e-06,
      "loss": 1.61271439,
      "memory(GiB)": 117.38,
      "step": 63430,
      "train_speed(iter/s)": 1.636885
    },
    {
      "acc": 0.66210623,
      "epoch": 1.6092085235920852,
      "grad_norm": 5.125,
      "learning_rate": 1.0079965576720375e-06,
      "loss": 1.54774313,
      "memory(GiB)": 117.38,
      "step": 63435,
      "train_speed(iter/s)": 1.636898
    },
    {
      "acc": 0.66705546,
      "epoch": 1.6093353627600204,
      "grad_norm": 6.125,
      "learning_rate": 1.0073652413475936e-06,
      "loss": 1.57116737,
      "memory(GiB)": 117.38,
      "step": 63440,
      "train_speed(iter/s)": 1.636911
    },
    {
      "acc": 0.67672005,
      "epoch": 1.6094622019279554,
      "grad_norm": 7.15625,
      "learning_rate": 1.0067341006370535e-06,
      "loss": 1.5560276,
      "memory(GiB)": 117.38,
      "step": 63445,
      "train_speed(iter/s)": 1.636923
    },
    {
      "acc": 0.64681377,
      "epoch": 1.6095890410958904,
      "grad_norm": 5.53125,
      "learning_rate": 1.0061031355681766e-06,
      "loss": 1.55814152,
      "memory(GiB)": 117.38,
      "step": 63450,
      "train_speed(iter/s)": 1.636937
    },
    {
      "acc": 0.65562639,
      "epoch": 1.6097158802638254,
      "grad_norm": 5.15625,
      "learning_rate": 1.0054723461687133e-06,
      "loss": 1.63386269,
      "memory(GiB)": 117.38,
      "step": 63455,
      "train_speed(iter/s)": 1.63695
    },
    {
      "acc": 0.67083125,
      "epoch": 1.6098427194317604,
      "grad_norm": 4.875,
      "learning_rate": 1.0048417324664118e-06,
      "loss": 1.5490263,
      "memory(GiB)": 117.38,
      "step": 63460,
      "train_speed(iter/s)": 1.636964
    },
    {
      "acc": 0.66167526,
      "epoch": 1.6099695585996956,
      "grad_norm": 5.28125,
      "learning_rate": 1.0042112944890075e-06,
      "loss": 1.60895329,
      "memory(GiB)": 117.38,
      "step": 63465,
      "train_speed(iter/s)": 1.636977
    },
    {
      "acc": 0.67047067,
      "epoch": 1.6100963977676308,
      "grad_norm": 6.875,
      "learning_rate": 1.003581032264231e-06,
      "loss": 1.5283864,
      "memory(GiB)": 117.38,
      "step": 63470,
      "train_speed(iter/s)": 1.63699
    },
    {
      "acc": 0.66012039,
      "epoch": 1.6102232369355658,
      "grad_norm": 6.0,
      "learning_rate": 1.0029509458198027e-06,
      "loss": 1.56061935,
      "memory(GiB)": 117.38,
      "step": 63475,
      "train_speed(iter/s)": 1.637005
    },
    {
      "acc": 0.66320295,
      "epoch": 1.6103500761035008,
      "grad_norm": 5.84375,
      "learning_rate": 1.0023210351834378e-06,
      "loss": 1.61925793,
      "memory(GiB)": 117.38,
      "step": 63480,
      "train_speed(iter/s)": 1.637018
    },
    {
      "acc": 0.65719404,
      "epoch": 1.6104769152714358,
      "grad_norm": 5.8125,
      "learning_rate": 1.001691300382842e-06,
      "loss": 1.61541443,
      "memory(GiB)": 117.38,
      "step": 63485,
      "train_speed(iter/s)": 1.637031
    },
    {
      "acc": 0.66457138,
      "epoch": 1.6106037544393708,
      "grad_norm": 5.3125,
      "learning_rate": 1.001061741445714e-06,
      "loss": 1.53346319,
      "memory(GiB)": 117.38,
      "step": 63490,
      "train_speed(iter/s)": 1.637044
    },
    {
      "acc": 0.64835467,
      "epoch": 1.610730593607306,
      "grad_norm": 6.0625,
      "learning_rate": 1.000432358399745e-06,
      "loss": 1.59138069,
      "memory(GiB)": 117.38,
      "step": 63495,
      "train_speed(iter/s)": 1.637057
    },
    {
      "acc": 0.67487812,
      "epoch": 1.610857432775241,
      "grad_norm": 6.125,
      "learning_rate": 9.99803151272617e-07,
      "loss": 1.54327688,
      "memory(GiB)": 117.38,
      "step": 63500,
      "train_speed(iter/s)": 1.637069
    },
    {
      "acc": 0.66052384,
      "epoch": 1.6109842719431762,
      "grad_norm": 6.96875,
      "learning_rate": 9.991741200920062e-07,
      "loss": 1.64372787,
      "memory(GiB)": 117.38,
      "step": 63505,
      "train_speed(iter/s)": 1.637083
    },
    {
      "acc": 0.6670166,
      "epoch": 1.6111111111111112,
      "grad_norm": 5.28125,
      "learning_rate": 9.985452648855803e-07,
      "loss": 1.56333094,
      "memory(GiB)": 117.38,
      "step": 63510,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.65294228,
      "epoch": 1.6112379502790461,
      "grad_norm": 5.4375,
      "learning_rate": 9.979165856809985e-07,
      "loss": 1.5945447,
      "memory(GiB)": 117.38,
      "step": 63515,
      "train_speed(iter/s)": 1.63711
    },
    {
      "acc": 0.65545826,
      "epoch": 1.6113647894469811,
      "grad_norm": 6.21875,
      "learning_rate": 9.972880825059134e-07,
      "loss": 1.64409695,
      "memory(GiB)": 117.38,
      "step": 63520,
      "train_speed(iter/s)": 1.637124
    },
    {
      "acc": 0.65820494,
      "epoch": 1.6114916286149161,
      "grad_norm": 5.65625,
      "learning_rate": 9.966597553879681e-07,
      "loss": 1.5589798,
      "memory(GiB)": 117.38,
      "step": 63525,
      "train_speed(iter/s)": 1.637137
    },
    {
      "acc": 0.66281343,
      "epoch": 1.6116184677828513,
      "grad_norm": 6.03125,
      "learning_rate": 9.96031604354803e-07,
      "loss": 1.65268631,
      "memory(GiB)": 117.38,
      "step": 63530,
      "train_speed(iter/s)": 1.637151
    },
    {
      "acc": 0.65771389,
      "epoch": 1.6117453069507865,
      "grad_norm": 6.84375,
      "learning_rate": 9.954036294340425e-07,
      "loss": 1.64918041,
      "memory(GiB)": 117.38,
      "step": 63535,
      "train_speed(iter/s)": 1.637165
    },
    {
      "acc": 0.66634417,
      "epoch": 1.6118721461187215,
      "grad_norm": 7.0625,
      "learning_rate": 9.947758306533101e-07,
      "loss": 1.5305583,
      "memory(GiB)": 117.38,
      "step": 63540,
      "train_speed(iter/s)": 1.637179
    },
    {
      "acc": 0.66381764,
      "epoch": 1.6119989852866565,
      "grad_norm": 5.15625,
      "learning_rate": 9.941482080402177e-07,
      "loss": 1.57927227,
      "memory(GiB)": 117.38,
      "step": 63545,
      "train_speed(iter/s)": 1.637193
    },
    {
      "acc": 0.6571826,
      "epoch": 1.6121258244545915,
      "grad_norm": 6.375,
      "learning_rate": 9.935207616223741e-07,
      "loss": 1.58145609,
      "memory(GiB)": 117.38,
      "step": 63550,
      "train_speed(iter/s)": 1.637207
    },
    {
      "acc": 0.66311655,
      "epoch": 1.6122526636225265,
      "grad_norm": 6.09375,
      "learning_rate": 9.928934914273735e-07,
      "loss": 1.57143135,
      "memory(GiB)": 117.38,
      "step": 63555,
      "train_speed(iter/s)": 1.63722
    },
    {
      "acc": 0.66612053,
      "epoch": 1.6123795027904617,
      "grad_norm": 5.03125,
      "learning_rate": 9.922663974828066e-07,
      "loss": 1.60198345,
      "memory(GiB)": 117.38,
      "step": 63560,
      "train_speed(iter/s)": 1.637234
    },
    {
      "acc": 0.64748149,
      "epoch": 1.6125063419583967,
      "grad_norm": 5.25,
      "learning_rate": 9.916394798162582e-07,
      "loss": 1.60756569,
      "memory(GiB)": 117.38,
      "step": 63565,
      "train_speed(iter/s)": 1.637248
    },
    {
      "acc": 0.65283551,
      "epoch": 1.612633181126332,
      "grad_norm": 5.6875,
      "learning_rate": 9.91012738455303e-07,
      "loss": 1.59576664,
      "memory(GiB)": 117.38,
      "step": 63570,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.6569778,
      "epoch": 1.612760020294267,
      "grad_norm": 5.84375,
      "learning_rate": 9.903861734275032e-07,
      "loss": 1.53625526,
      "memory(GiB)": 117.38,
      "step": 63575,
      "train_speed(iter/s)": 1.637276
    },
    {
      "acc": 0.67263918,
      "epoch": 1.612886859462202,
      "grad_norm": 5.8125,
      "learning_rate": 9.897597847604228e-07,
      "loss": 1.58514309,
      "memory(GiB)": 117.38,
      "step": 63580,
      "train_speed(iter/s)": 1.63729
    },
    {
      "acc": 0.63764133,
      "epoch": 1.6130136986301369,
      "grad_norm": 8.0625,
      "learning_rate": 9.89133572481612e-07,
      "loss": 1.66299706,
      "memory(GiB)": 117.38,
      "step": 63585,
      "train_speed(iter/s)": 1.637304
    },
    {
      "acc": 0.65980377,
      "epoch": 1.613140537798072,
      "grad_norm": 5.6875,
      "learning_rate": 9.885075366186148e-07,
      "loss": 1.59518061,
      "memory(GiB)": 117.38,
      "step": 63590,
      "train_speed(iter/s)": 1.637318
    },
    {
      "acc": 0.6663126,
      "epoch": 1.613267376966007,
      "grad_norm": 4.40625,
      "learning_rate": 9.87881677198963e-07,
      "loss": 1.5243556,
      "memory(GiB)": 117.38,
      "step": 63595,
      "train_speed(iter/s)": 1.637332
    },
    {
      "acc": 0.64048948,
      "epoch": 1.6133942161339423,
      "grad_norm": 5.71875,
      "learning_rate": 9.872559942501897e-07,
      "loss": 1.71440315,
      "memory(GiB)": 117.38,
      "step": 63600,
      "train_speed(iter/s)": 1.637347
    },
    {
      "acc": 0.63816881,
      "epoch": 1.6135210553018773,
      "grad_norm": 5.125,
      "learning_rate": 9.866304877998134e-07,
      "loss": 1.66364098,
      "memory(GiB)": 117.38,
      "step": 63605,
      "train_speed(iter/s)": 1.637362
    },
    {
      "acc": 0.65931482,
      "epoch": 1.6136478944698123,
      "grad_norm": 6.09375,
      "learning_rate": 9.860051578753466e-07,
      "loss": 1.62982101,
      "memory(GiB)": 117.38,
      "step": 63610,
      "train_speed(iter/s)": 1.637375
    },
    {
      "acc": 0.65058303,
      "epoch": 1.6137747336377473,
      "grad_norm": 6.03125,
      "learning_rate": 9.853800045042938e-07,
      "loss": 1.60727596,
      "memory(GiB)": 117.38,
      "step": 63615,
      "train_speed(iter/s)": 1.637388
    },
    {
      "acc": 0.65808096,
      "epoch": 1.6139015728056822,
      "grad_norm": 4.9375,
      "learning_rate": 9.847550277141526e-07,
      "loss": 1.59751825,
      "memory(GiB)": 117.38,
      "step": 63620,
      "train_speed(iter/s)": 1.637402
    },
    {
      "acc": 0.65432973,
      "epoch": 1.6140284119736175,
      "grad_norm": 6.25,
      "learning_rate": 9.841302275324128e-07,
      "loss": 1.57043667,
      "memory(GiB)": 117.38,
      "step": 63625,
      "train_speed(iter/s)": 1.637415
    },
    {
      "acc": 0.65543289,
      "epoch": 1.6141552511415527,
      "grad_norm": 6.59375,
      "learning_rate": 9.835056039865542e-07,
      "loss": 1.61725349,
      "memory(GiB)": 117.38,
      "step": 63630,
      "train_speed(iter/s)": 1.637429
    },
    {
      "acc": 0.64806819,
      "epoch": 1.6142820903094877,
      "grad_norm": 9.375,
      "learning_rate": 9.82881157104052e-07,
      "loss": 1.66217003,
      "memory(GiB)": 117.38,
      "step": 63635,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.64420433,
      "epoch": 1.6144089294774226,
      "grad_norm": 4.9375,
      "learning_rate": 9.822568869123712e-07,
      "loss": 1.69623299,
      "memory(GiB)": 117.38,
      "step": 63640,
      "train_speed(iter/s)": 1.637456
    },
    {
      "acc": 0.65110359,
      "epoch": 1.6145357686453576,
      "grad_norm": 4.78125,
      "learning_rate": 9.816327934389707e-07,
      "loss": 1.58794594,
      "memory(GiB)": 117.38,
      "step": 63645,
      "train_speed(iter/s)": 1.63747
    },
    {
      "acc": 0.66972895,
      "epoch": 1.6146626078132926,
      "grad_norm": 5.09375,
      "learning_rate": 9.810088767113008e-07,
      "loss": 1.52417698,
      "memory(GiB)": 117.38,
      "step": 63650,
      "train_speed(iter/s)": 1.637483
    },
    {
      "acc": 0.64999018,
      "epoch": 1.6147894469812278,
      "grad_norm": 5.8125,
      "learning_rate": 9.80385136756804e-07,
      "loss": 1.64205666,
      "memory(GiB)": 117.38,
      "step": 63655,
      "train_speed(iter/s)": 1.637496
    },
    {
      "acc": 0.65315814,
      "epoch": 1.6149162861491628,
      "grad_norm": 6.40625,
      "learning_rate": 9.797615736029148e-07,
      "loss": 1.62650452,
      "memory(GiB)": 117.38,
      "step": 63660,
      "train_speed(iter/s)": 1.637511
    },
    {
      "acc": 0.66172247,
      "epoch": 1.615043125317098,
      "grad_norm": 5.0625,
      "learning_rate": 9.791381872770594e-07,
      "loss": 1.60049706,
      "memory(GiB)": 117.38,
      "step": 63665,
      "train_speed(iter/s)": 1.637524
    },
    {
      "acc": 0.65203691,
      "epoch": 1.615169964485033,
      "grad_norm": 5.8125,
      "learning_rate": 9.785149778066615e-07,
      "loss": 1.64893551,
      "memory(GiB)": 117.38,
      "step": 63670,
      "train_speed(iter/s)": 1.637537
    },
    {
      "acc": 0.66618047,
      "epoch": 1.615296803652968,
      "grad_norm": 6.25,
      "learning_rate": 9.778919452191277e-07,
      "loss": 1.58618259,
      "memory(GiB)": 117.38,
      "step": 63675,
      "train_speed(iter/s)": 1.637551
    },
    {
      "acc": 0.68306036,
      "epoch": 1.615423642820903,
      "grad_norm": 7.0,
      "learning_rate": 9.77269089541864e-07,
      "loss": 1.52129574,
      "memory(GiB)": 117.38,
      "step": 63680,
      "train_speed(iter/s)": 1.637565
    },
    {
      "acc": 0.65918469,
      "epoch": 1.615550481988838,
      "grad_norm": 5.4375,
      "learning_rate": 9.766464108022644e-07,
      "loss": 1.66668892,
      "memory(GiB)": 117.38,
      "step": 63685,
      "train_speed(iter/s)": 1.637579
    },
    {
      "acc": 0.66751914,
      "epoch": 1.6156773211567732,
      "grad_norm": 5.90625,
      "learning_rate": 9.760239090277213e-07,
      "loss": 1.57237225,
      "memory(GiB)": 117.38,
      "step": 63690,
      "train_speed(iter/s)": 1.637592
    },
    {
      "acc": 0.67409949,
      "epoch": 1.6158041603247084,
      "grad_norm": 6.5,
      "learning_rate": 9.75401584245611e-07,
      "loss": 1.55902081,
      "memory(GiB)": 117.38,
      "step": 63695,
      "train_speed(iter/s)": 1.637606
    },
    {
      "acc": 0.6373436,
      "epoch": 1.6159309994926434,
      "grad_norm": 6.0625,
      "learning_rate": 9.747794364833063e-07,
      "loss": 1.71150436,
      "memory(GiB)": 117.38,
      "step": 63700,
      "train_speed(iter/s)": 1.63762
    },
    {
      "acc": 0.64090705,
      "epoch": 1.6160578386605784,
      "grad_norm": 5.3125,
      "learning_rate": 9.741574657681747e-07,
      "loss": 1.65247822,
      "memory(GiB)": 117.38,
      "step": 63705,
      "train_speed(iter/s)": 1.637633
    },
    {
      "acc": 0.64660087,
      "epoch": 1.6161846778285134,
      "grad_norm": 5.75,
      "learning_rate": 9.735356721275734e-07,
      "loss": 1.60407181,
      "memory(GiB)": 117.38,
      "step": 63710,
      "train_speed(iter/s)": 1.637646
    },
    {
      "acc": 0.65684347,
      "epoch": 1.6163115169964484,
      "grad_norm": 6.46875,
      "learning_rate": 9.729140555888483e-07,
      "loss": 1.4954215,
      "memory(GiB)": 117.38,
      "step": 63715,
      "train_speed(iter/s)": 1.63766
    },
    {
      "acc": 0.66543336,
      "epoch": 1.6164383561643836,
      "grad_norm": 6.5,
      "learning_rate": 9.722926161793417e-07,
      "loss": 1.5744091,
      "memory(GiB)": 117.38,
      "step": 63720,
      "train_speed(iter/s)": 1.637673
    },
    {
      "acc": 0.65732365,
      "epoch": 1.6165651953323186,
      "grad_norm": 5.78125,
      "learning_rate": 9.716713539263895e-07,
      "loss": 1.63254242,
      "memory(GiB)": 117.38,
      "step": 63725,
      "train_speed(iter/s)": 1.637686
    },
    {
      "acc": 0.65619011,
      "epoch": 1.6166920345002538,
      "grad_norm": 5.34375,
      "learning_rate": 9.710502688573175e-07,
      "loss": 1.60967789,
      "memory(GiB)": 117.38,
      "step": 63730,
      "train_speed(iter/s)": 1.6377
    },
    {
      "acc": 0.65877409,
      "epoch": 1.6168188736681888,
      "grad_norm": 5.15625,
      "learning_rate": 9.704293609994403e-07,
      "loss": 1.61505661,
      "memory(GiB)": 117.38,
      "step": 63735,
      "train_speed(iter/s)": 1.637713
    },
    {
      "acc": 0.6580019,
      "epoch": 1.6169457128361238,
      "grad_norm": 5.40625,
      "learning_rate": 9.69808630380072e-07,
      "loss": 1.63126183,
      "memory(GiB)": 117.38,
      "step": 63740,
      "train_speed(iter/s)": 1.637726
    },
    {
      "acc": 0.66299114,
      "epoch": 1.6170725520040587,
      "grad_norm": 5.875,
      "learning_rate": 9.691880770265132e-07,
      "loss": 1.5350893,
      "memory(GiB)": 117.38,
      "step": 63745,
      "train_speed(iter/s)": 1.637738
    },
    {
      "acc": 0.6549655,
      "epoch": 1.617199391171994,
      "grad_norm": 6.375,
      "learning_rate": 9.685677009660587e-07,
      "loss": 1.64639626,
      "memory(GiB)": 117.38,
      "step": 63750,
      "train_speed(iter/s)": 1.637751
    },
    {
      "acc": 0.64077358,
      "epoch": 1.617326230339929,
      "grad_norm": 7.59375,
      "learning_rate": 9.679475022259965e-07,
      "loss": 1.69925995,
      "memory(GiB)": 117.38,
      "step": 63755,
      "train_speed(iter/s)": 1.637765
    },
    {
      "acc": 0.65708489,
      "epoch": 1.6174530695078642,
      "grad_norm": 6.25,
      "learning_rate": 9.673274808336047e-07,
      "loss": 1.63301392,
      "memory(GiB)": 117.38,
      "step": 63760,
      "train_speed(iter/s)": 1.637778
    },
    {
      "acc": 0.65188012,
      "epoch": 1.6175799086757991,
      "grad_norm": 5.90625,
      "learning_rate": 9.66707636816155e-07,
      "loss": 1.69917145,
      "memory(GiB)": 117.38,
      "step": 63765,
      "train_speed(iter/s)": 1.637791
    },
    {
      "acc": 0.65312948,
      "epoch": 1.6177067478437341,
      "grad_norm": 6.28125,
      "learning_rate": 9.660879702009106e-07,
      "loss": 1.68750992,
      "memory(GiB)": 117.38,
      "step": 63770,
      "train_speed(iter/s)": 1.637805
    },
    {
      "acc": 0.63581977,
      "epoch": 1.6178335870116691,
      "grad_norm": 7.15625,
      "learning_rate": 9.654684810151276e-07,
      "loss": 1.69862938,
      "memory(GiB)": 117.38,
      "step": 63775,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.65914712,
      "epoch": 1.6179604261796041,
      "grad_norm": 5.875,
      "learning_rate": 9.648491692860534e-07,
      "loss": 1.5766984,
      "memory(GiB)": 117.38,
      "step": 63780,
      "train_speed(iter/s)": 1.637833
    },
    {
      "acc": 0.65791836,
      "epoch": 1.6180872653475393,
      "grad_norm": 5.78125,
      "learning_rate": 9.642300350409289e-07,
      "loss": 1.59953127,
      "memory(GiB)": 117.38,
      "step": 63785,
      "train_speed(iter/s)": 1.637846
    },
    {
      "acc": 0.64600964,
      "epoch": 1.6182141045154745,
      "grad_norm": 6.96875,
      "learning_rate": 9.636110783069852e-07,
      "loss": 1.70841484,
      "memory(GiB)": 117.38,
      "step": 63790,
      "train_speed(iter/s)": 1.637858
    },
    {
      "acc": 0.64777722,
      "epoch": 1.6183409436834095,
      "grad_norm": 5.59375,
      "learning_rate": 9.629922991114482e-07,
      "loss": 1.67116699,
      "memory(GiB)": 117.38,
      "step": 63795,
      "train_speed(iter/s)": 1.637872
    },
    {
      "acc": 0.64868355,
      "epoch": 1.6184677828513445,
      "grad_norm": 5.375,
      "learning_rate": 9.623736974815334e-07,
      "loss": 1.60910454,
      "memory(GiB)": 117.38,
      "step": 63800,
      "train_speed(iter/s)": 1.637885
    },
    {
      "acc": 0.67498946,
      "epoch": 1.6185946220192795,
      "grad_norm": 6.3125,
      "learning_rate": 9.617552734444502e-07,
      "loss": 1.49038763,
      "memory(GiB)": 117.38,
      "step": 63805,
      "train_speed(iter/s)": 1.637899
    },
    {
      "acc": 0.66613836,
      "epoch": 1.6187214611872145,
      "grad_norm": 4.625,
      "learning_rate": 9.611370270273996e-07,
      "loss": 1.50617428,
      "memory(GiB)": 117.38,
      "step": 63810,
      "train_speed(iter/s)": 1.637912
    },
    {
      "acc": 0.65473881,
      "epoch": 1.6188483003551497,
      "grad_norm": 5.09375,
      "learning_rate": 9.605189582575741e-07,
      "loss": 1.57273035,
      "memory(GiB)": 117.38,
      "step": 63815,
      "train_speed(iter/s)": 1.637926
    },
    {
      "acc": 0.65938063,
      "epoch": 1.6189751395230847,
      "grad_norm": 5.96875,
      "learning_rate": 9.599010671621605e-07,
      "loss": 1.63436584,
      "memory(GiB)": 117.38,
      "step": 63820,
      "train_speed(iter/s)": 1.637938
    },
    {
      "acc": 0.64908562,
      "epoch": 1.61910197869102,
      "grad_norm": 5.65625,
      "learning_rate": 9.592833537683344e-07,
      "loss": 1.63700027,
      "memory(GiB)": 117.38,
      "step": 63825,
      "train_speed(iter/s)": 1.637952
    },
    {
      "acc": 0.64169598,
      "epoch": 1.619228817858955,
      "grad_norm": 6.28125,
      "learning_rate": 9.586658181032693e-07,
      "loss": 1.65708389,
      "memory(GiB)": 117.38,
      "step": 63830,
      "train_speed(iter/s)": 1.637966
    },
    {
      "acc": 0.65639119,
      "epoch": 1.6193556570268899,
      "grad_norm": 4.96875,
      "learning_rate": 9.580484601941237e-07,
      "loss": 1.55072823,
      "memory(GiB)": 117.38,
      "step": 63835,
      "train_speed(iter/s)": 1.637978
    },
    {
      "acc": 0.65030546,
      "epoch": 1.6194824961948249,
      "grad_norm": 6.53125,
      "learning_rate": 9.574312800680514e-07,
      "loss": 1.65216236,
      "memory(GiB)": 117.38,
      "step": 63840,
      "train_speed(iter/s)": 1.637991
    },
    {
      "acc": 0.65782022,
      "epoch": 1.6196093353627599,
      "grad_norm": 6.125,
      "learning_rate": 9.56814277752201e-07,
      "loss": 1.57568588,
      "memory(GiB)": 117.38,
      "step": 63845,
      "train_speed(iter/s)": 1.638004
    },
    {
      "acc": 0.64032516,
      "epoch": 1.619736174530695,
      "grad_norm": 8.0625,
      "learning_rate": 9.561974532737124e-07,
      "loss": 1.68418541,
      "memory(GiB)": 117.38,
      "step": 63850,
      "train_speed(iter/s)": 1.638018
    },
    {
      "acc": 0.64176078,
      "epoch": 1.6198630136986303,
      "grad_norm": 5.0625,
      "learning_rate": 9.555808066597123e-07,
      "loss": 1.56404171,
      "memory(GiB)": 117.38,
      "step": 63855,
      "train_speed(iter/s)": 1.638032
    },
    {
      "acc": 0.66833396,
      "epoch": 1.6199898528665653,
      "grad_norm": 6.03125,
      "learning_rate": 9.549643379373236e-07,
      "loss": 1.56283779,
      "memory(GiB)": 117.38,
      "step": 63860,
      "train_speed(iter/s)": 1.638046
    },
    {
      "acc": 0.65467529,
      "epoch": 1.6201166920345003,
      "grad_norm": 6.46875,
      "learning_rate": 9.543480471336652e-07,
      "loss": 1.58498173,
      "memory(GiB)": 117.38,
      "step": 63865,
      "train_speed(iter/s)": 1.638059
    },
    {
      "acc": 0.68168626,
      "epoch": 1.6202435312024352,
      "grad_norm": 5.40625,
      "learning_rate": 9.537319342758434e-07,
      "loss": 1.51265812,
      "memory(GiB)": 117.38,
      "step": 63870,
      "train_speed(iter/s)": 1.638074
    },
    {
      "acc": 0.64589157,
      "epoch": 1.6203703703703702,
      "grad_norm": 7.65625,
      "learning_rate": 9.531159993909533e-07,
      "loss": 1.59957457,
      "memory(GiB)": 117.38,
      "step": 63875,
      "train_speed(iter/s)": 1.638088
    },
    {
      "acc": 0.67579846,
      "epoch": 1.6204972095383054,
      "grad_norm": 5.8125,
      "learning_rate": 9.525002425060914e-07,
      "loss": 1.48189831,
      "memory(GiB)": 117.38,
      "step": 63880,
      "train_speed(iter/s)": 1.638101
    },
    {
      "acc": 0.6530838,
      "epoch": 1.6206240487062404,
      "grad_norm": 5.40625,
      "learning_rate": 9.518846636483392e-07,
      "loss": 1.62513905,
      "memory(GiB)": 117.38,
      "step": 63885,
      "train_speed(iter/s)": 1.638114
    },
    {
      "acc": 0.63869891,
      "epoch": 1.6207508878741756,
      "grad_norm": 4.84375,
      "learning_rate": 9.512692628447745e-07,
      "loss": 1.68012791,
      "memory(GiB)": 117.38,
      "step": 63890,
      "train_speed(iter/s)": 1.638126
    },
    {
      "acc": 0.65494466,
      "epoch": 1.6208777270421106,
      "grad_norm": 6.03125,
      "learning_rate": 9.506540401224612e-07,
      "loss": 1.58157272,
      "memory(GiB)": 117.38,
      "step": 63895,
      "train_speed(iter/s)": 1.638139
    },
    {
      "acc": 0.65738754,
      "epoch": 1.6210045662100456,
      "grad_norm": 5.03125,
      "learning_rate": 9.500389955084638e-07,
      "loss": 1.58201752,
      "memory(GiB)": 117.38,
      "step": 63900,
      "train_speed(iter/s)": 1.638152
    },
    {
      "acc": 0.66312809,
      "epoch": 1.6211314053779806,
      "grad_norm": 7.4375,
      "learning_rate": 9.494241290298334e-07,
      "loss": 1.60181408,
      "memory(GiB)": 117.38,
      "step": 63905,
      "train_speed(iter/s)": 1.638166
    },
    {
      "acc": 0.66506815,
      "epoch": 1.6212582445459158,
      "grad_norm": 6.375,
      "learning_rate": 9.488094407136139e-07,
      "loss": 1.6024065,
      "memory(GiB)": 117.38,
      "step": 63910,
      "train_speed(iter/s)": 1.63818
    },
    {
      "acc": 0.65851274,
      "epoch": 1.6213850837138508,
      "grad_norm": 6.71875,
      "learning_rate": 9.481949305868421e-07,
      "loss": 1.60496998,
      "memory(GiB)": 117.38,
      "step": 63915,
      "train_speed(iter/s)": 1.638193
    },
    {
      "acc": 0.64477277,
      "epoch": 1.621511922881786,
      "grad_norm": 7.1875,
      "learning_rate": 9.475805986765479e-07,
      "loss": 1.67787819,
      "memory(GiB)": 117.38,
      "step": 63920,
      "train_speed(iter/s)": 1.638207
    },
    {
      "acc": 0.64880447,
      "epoch": 1.621638762049721,
      "grad_norm": 4.9375,
      "learning_rate": 9.469664450097515e-07,
      "loss": 1.56779327,
      "memory(GiB)": 117.38,
      "step": 63925,
      "train_speed(iter/s)": 1.63822
    },
    {
      "acc": 0.6680727,
      "epoch": 1.621765601217656,
      "grad_norm": 5.65625,
      "learning_rate": 9.463524696134663e-07,
      "loss": 1.62146034,
      "memory(GiB)": 117.38,
      "step": 63930,
      "train_speed(iter/s)": 1.638234
    },
    {
      "acc": 0.65873756,
      "epoch": 1.621892440385591,
      "grad_norm": 5.875,
      "learning_rate": 9.457386725146978e-07,
      "loss": 1.57686338,
      "memory(GiB)": 117.38,
      "step": 63935,
      "train_speed(iter/s)": 1.638246
    },
    {
      "acc": 0.64717083,
      "epoch": 1.622019279553526,
      "grad_norm": 5.03125,
      "learning_rate": 9.451250537404433e-07,
      "loss": 1.63794975,
      "memory(GiB)": 117.38,
      "step": 63940,
      "train_speed(iter/s)": 1.63826
    },
    {
      "acc": 0.65129347,
      "epoch": 1.6221461187214612,
      "grad_norm": 5.03125,
      "learning_rate": 9.44511613317693e-07,
      "loss": 1.61691475,
      "memory(GiB)": 117.38,
      "step": 63945,
      "train_speed(iter/s)": 1.638273
    },
    {
      "acc": 0.65615168,
      "epoch": 1.6222729578893964,
      "grad_norm": 7.34375,
      "learning_rate": 9.438983512734279e-07,
      "loss": 1.60385094,
      "memory(GiB)": 117.38,
      "step": 63950,
      "train_speed(iter/s)": 1.638286
    },
    {
      "acc": 0.66852179,
      "epoch": 1.6223997970573314,
      "grad_norm": 5.90625,
      "learning_rate": 9.432852676346233e-07,
      "loss": 1.56938505,
      "memory(GiB)": 117.38,
      "step": 63955,
      "train_speed(iter/s)": 1.638301
    },
    {
      "acc": 0.65511212,
      "epoch": 1.6225266362252664,
      "grad_norm": 6.875,
      "learning_rate": 9.426723624282436e-07,
      "loss": 1.62275887,
      "memory(GiB)": 117.38,
      "step": 63960,
      "train_speed(iter/s)": 1.638315
    },
    {
      "acc": 0.6600563,
      "epoch": 1.6226534753932014,
      "grad_norm": 5.625,
      "learning_rate": 9.420596356812473e-07,
      "loss": 1.59218225,
      "memory(GiB)": 117.38,
      "step": 63965,
      "train_speed(iter/s)": 1.638328
    },
    {
      "acc": 0.6611042,
      "epoch": 1.6227803145611364,
      "grad_norm": 5.78125,
      "learning_rate": 9.414470874205883e-07,
      "loss": 1.52277365,
      "memory(GiB)": 117.38,
      "step": 63970,
      "train_speed(iter/s)": 1.638342
    },
    {
      "acc": 0.65541153,
      "epoch": 1.6229071537290716,
      "grad_norm": 7.28125,
      "learning_rate": 9.408347176732053e-07,
      "loss": 1.62493229,
      "memory(GiB)": 117.38,
      "step": 63975,
      "train_speed(iter/s)": 1.638356
    },
    {
      "acc": 0.67996044,
      "epoch": 1.6230339928970066,
      "grad_norm": 5.3125,
      "learning_rate": 9.402225264660336e-07,
      "loss": 1.4980999,
      "memory(GiB)": 117.38,
      "step": 63980,
      "train_speed(iter/s)": 1.638369
    },
    {
      "acc": 0.6695806,
      "epoch": 1.6231608320649418,
      "grad_norm": 6.78125,
      "learning_rate": 9.396105138259997e-07,
      "loss": 1.54065456,
      "memory(GiB)": 117.38,
      "step": 63985,
      "train_speed(iter/s)": 1.638383
    },
    {
      "acc": 0.66249571,
      "epoch": 1.6232876712328768,
      "grad_norm": 6.46875,
      "learning_rate": 9.389986797800271e-07,
      "loss": 1.56858435,
      "memory(GiB)": 117.38,
      "step": 63990,
      "train_speed(iter/s)": 1.638396
    },
    {
      "acc": 0.65884171,
      "epoch": 1.6234145104008117,
      "grad_norm": 5.59375,
      "learning_rate": 9.383870243550214e-07,
      "loss": 1.57668962,
      "memory(GiB)": 117.38,
      "step": 63995,
      "train_speed(iter/s)": 1.63841
    },
    {
      "acc": 0.65744395,
      "epoch": 1.6235413495687467,
      "grad_norm": 4.8125,
      "learning_rate": 9.377755475778877e-07,
      "loss": 1.5917057,
      "memory(GiB)": 117.38,
      "step": 64000,
      "train_speed(iter/s)": 1.638425
    },
    {
      "epoch": 1.6235413495687467,
      "eval_acc": 0.646299417153764,
      "eval_loss": 1.5733387470245361,
      "eval_runtime": 58.7285,
      "eval_samples_per_second": 108.465,
      "eval_steps_per_second": 27.125,
      "step": 64000
    },
    {
      "acc": 0.65668488,
      "epoch": 1.6236681887366817,
      "grad_norm": 4.53125,
      "learning_rate": 9.371642494755223e-07,
      "loss": 1.5850708,
      "memory(GiB)": 117.38,
      "step": 64005,
      "train_speed(iter/s)": 1.635799
    },
    {
      "acc": 0.66200113,
      "epoch": 1.623795027904617,
      "grad_norm": 5.21875,
      "learning_rate": 9.365531300748143e-07,
      "loss": 1.56423283,
      "memory(GiB)": 117.38,
      "step": 64010,
      "train_speed(iter/s)": 1.635812
    },
    {
      "acc": 0.66863675,
      "epoch": 1.6239218670725521,
      "grad_norm": 5.78125,
      "learning_rate": 9.359421894026394e-07,
      "loss": 1.50921106,
      "memory(GiB)": 117.38,
      "step": 64015,
      "train_speed(iter/s)": 1.635825
    },
    {
      "acc": 0.62581825,
      "epoch": 1.6240487062404871,
      "grad_norm": 5.8125,
      "learning_rate": 9.353314274858726e-07,
      "loss": 1.69167366,
      "memory(GiB)": 117.38,
      "step": 64020,
      "train_speed(iter/s)": 1.635839
    },
    {
      "acc": 0.64224763,
      "epoch": 1.6241755454084221,
      "grad_norm": 6.0,
      "learning_rate": 9.347208443513773e-07,
      "loss": 1.61048775,
      "memory(GiB)": 117.38,
      "step": 64025,
      "train_speed(iter/s)": 1.635852
    },
    {
      "acc": 0.65394907,
      "epoch": 1.6243023845763571,
      "grad_norm": 5.28125,
      "learning_rate": 9.341104400260103e-07,
      "loss": 1.63138447,
      "memory(GiB)": 117.38,
      "step": 64030,
      "train_speed(iter/s)": 1.635865
    },
    {
      "acc": 0.65322199,
      "epoch": 1.624429223744292,
      "grad_norm": 5.4375,
      "learning_rate": 9.335002145366167e-07,
      "loss": 1.64252357,
      "memory(GiB)": 117.38,
      "step": 64035,
      "train_speed(iter/s)": 1.635879
    },
    {
      "acc": 0.65276918,
      "epoch": 1.6245560629122273,
      "grad_norm": 6.21875,
      "learning_rate": 9.328901679100405e-07,
      "loss": 1.59390326,
      "memory(GiB)": 117.38,
      "step": 64040,
      "train_speed(iter/s)": 1.635894
    },
    {
      "acc": 0.63086853,
      "epoch": 1.6246829020801623,
      "grad_norm": 4.875,
      "learning_rate": 9.322803001731128e-07,
      "loss": 1.66026917,
      "memory(GiB)": 117.38,
      "step": 64045,
      "train_speed(iter/s)": 1.635907
    },
    {
      "acc": 0.64947605,
      "epoch": 1.6248097412480975,
      "grad_norm": 7.03125,
      "learning_rate": 9.316706113526591e-07,
      "loss": 1.62306824,
      "memory(GiB)": 117.38,
      "step": 64050,
      "train_speed(iter/s)": 1.635921
    },
    {
      "acc": 0.6578341,
      "epoch": 1.6249365804160325,
      "grad_norm": 5.28125,
      "learning_rate": 9.310611014754956e-07,
      "loss": 1.64150639,
      "memory(GiB)": 117.38,
      "step": 64055,
      "train_speed(iter/s)": 1.635935
    },
    {
      "acc": 0.63564014,
      "epoch": 1.6250634195839675,
      "grad_norm": 5.8125,
      "learning_rate": 9.304517705684308e-07,
      "loss": 1.65794983,
      "memory(GiB)": 117.38,
      "step": 64060,
      "train_speed(iter/s)": 1.635948
    },
    {
      "acc": 0.65869732,
      "epoch": 1.6251902587519025,
      "grad_norm": 6.65625,
      "learning_rate": 9.298426186582671e-07,
      "loss": 1.6023632,
      "memory(GiB)": 117.38,
      "step": 64065,
      "train_speed(iter/s)": 1.635962
    },
    {
      "acc": 0.64586973,
      "epoch": 1.6253170979198377,
      "grad_norm": 5.5625,
      "learning_rate": 9.292336457717965e-07,
      "loss": 1.63211555,
      "memory(GiB)": 117.38,
      "step": 64070,
      "train_speed(iter/s)": 1.635976
    },
    {
      "acc": 0.64693317,
      "epoch": 1.6254439370877727,
      "grad_norm": 6.34375,
      "learning_rate": 9.286248519358049e-07,
      "loss": 1.65288639,
      "memory(GiB)": 117.38,
      "step": 64075,
      "train_speed(iter/s)": 1.63599
    },
    {
      "acc": 0.6616128,
      "epoch": 1.625570776255708,
      "grad_norm": 6.0,
      "learning_rate": 9.280162371770696e-07,
      "loss": 1.50730543,
      "memory(GiB)": 117.38,
      "step": 64080,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.66329699,
      "epoch": 1.6256976154236429,
      "grad_norm": 5.03125,
      "learning_rate": 9.274078015223603e-07,
      "loss": 1.58291759,
      "memory(GiB)": 117.38,
      "step": 64085,
      "train_speed(iter/s)": 1.636016
    },
    {
      "acc": 0.65693541,
      "epoch": 1.6258244545915779,
      "grad_norm": 5.84375,
      "learning_rate": 9.26799544998439e-07,
      "loss": 1.59871483,
      "memory(GiB)": 117.38,
      "step": 64090,
      "train_speed(iter/s)": 1.63603
    },
    {
      "acc": 0.67011118,
      "epoch": 1.6259512937595129,
      "grad_norm": 6.125,
      "learning_rate": 9.261914676320594e-07,
      "loss": 1.49998903,
      "memory(GiB)": 117.38,
      "step": 64095,
      "train_speed(iter/s)": 1.636045
    },
    {
      "acc": 0.65041304,
      "epoch": 1.6260781329274479,
      "grad_norm": 6.03125,
      "learning_rate": 9.255835694499665e-07,
      "loss": 1.5893405,
      "memory(GiB)": 117.38,
      "step": 64100,
      "train_speed(iter/s)": 1.636058
    },
    {
      "acc": 0.68258257,
      "epoch": 1.626204972095383,
      "grad_norm": 5.25,
      "learning_rate": 9.249758504788986e-07,
      "loss": 1.56823673,
      "memory(GiB)": 117.38,
      "step": 64105,
      "train_speed(iter/s)": 1.636072
    },
    {
      "acc": 0.65966063,
      "epoch": 1.6263318112633183,
      "grad_norm": 5.03125,
      "learning_rate": 9.243683107455886e-07,
      "loss": 1.56983013,
      "memory(GiB)": 117.38,
      "step": 64110,
      "train_speed(iter/s)": 1.636086
    },
    {
      "acc": 0.66664734,
      "epoch": 1.6264586504312533,
      "grad_norm": 5.71875,
      "learning_rate": 9.237609502767558e-07,
      "loss": 1.5638628,
      "memory(GiB)": 117.38,
      "step": 64115,
      "train_speed(iter/s)": 1.636098
    },
    {
      "acc": 0.64629068,
      "epoch": 1.6265854895991883,
      "grad_norm": 5.4375,
      "learning_rate": 9.231537690991155e-07,
      "loss": 1.62178955,
      "memory(GiB)": 117.38,
      "step": 64120,
      "train_speed(iter/s)": 1.636112
    },
    {
      "acc": 0.66311121,
      "epoch": 1.6267123287671232,
      "grad_norm": 6.9375,
      "learning_rate": 9.225467672393729e-07,
      "loss": 1.58112679,
      "memory(GiB)": 117.38,
      "step": 64125,
      "train_speed(iter/s)": 1.636126
    },
    {
      "acc": 0.66931963,
      "epoch": 1.6268391679350582,
      "grad_norm": 6.5,
      "learning_rate": 9.2193994472423e-07,
      "loss": 1.58022137,
      "memory(GiB)": 117.38,
      "step": 64130,
      "train_speed(iter/s)": 1.636139
    },
    {
      "acc": 0.64963408,
      "epoch": 1.6269660071029934,
      "grad_norm": 5.4375,
      "learning_rate": 9.21333301580375e-07,
      "loss": 1.61292439,
      "memory(GiB)": 117.38,
      "step": 64135,
      "train_speed(iter/s)": 1.636153
    },
    {
      "acc": 0.64832935,
      "epoch": 1.6270928462709284,
      "grad_norm": 5.0,
      "learning_rate": 9.207268378344896e-07,
      "loss": 1.66601448,
      "memory(GiB)": 117.38,
      "step": 64140,
      "train_speed(iter/s)": 1.636166
    },
    {
      "acc": 0.66399355,
      "epoch": 1.6272196854388636,
      "grad_norm": 6.0625,
      "learning_rate": 9.201205535132523e-07,
      "loss": 1.62433567,
      "memory(GiB)": 117.38,
      "step": 64145,
      "train_speed(iter/s)": 1.636179
    },
    {
      "acc": 0.64774213,
      "epoch": 1.6273465246067986,
      "grad_norm": 5.71875,
      "learning_rate": 9.195144486433294e-07,
      "loss": 1.63149204,
      "memory(GiB)": 117.38,
      "step": 64150,
      "train_speed(iter/s)": 1.636193
    },
    {
      "acc": 0.68049345,
      "epoch": 1.6274733637747336,
      "grad_norm": 5.25,
      "learning_rate": 9.189085232513773e-07,
      "loss": 1.52996807,
      "memory(GiB)": 117.38,
      "step": 64155,
      "train_speed(iter/s)": 1.636207
    },
    {
      "acc": 0.65328145,
      "epoch": 1.6276002029426686,
      "grad_norm": 5.53125,
      "learning_rate": 9.183027773640485e-07,
      "loss": 1.66289978,
      "memory(GiB)": 117.38,
      "step": 64160,
      "train_speed(iter/s)": 1.63622
    },
    {
      "acc": 0.65506711,
      "epoch": 1.6277270421106036,
      "grad_norm": 4.46875,
      "learning_rate": 9.176972110079879e-07,
      "loss": 1.48536968,
      "memory(GiB)": 117.38,
      "step": 64165,
      "train_speed(iter/s)": 1.636234
    },
    {
      "acc": 0.66621208,
      "epoch": 1.6278538812785388,
      "grad_norm": 6.0,
      "learning_rate": 9.17091824209832e-07,
      "loss": 1.58151608,
      "memory(GiB)": 117.38,
      "step": 64170,
      "train_speed(iter/s)": 1.636247
    },
    {
      "acc": 0.66377249,
      "epoch": 1.627980720446474,
      "grad_norm": 5.34375,
      "learning_rate": 9.164866169962034e-07,
      "loss": 1.62506218,
      "memory(GiB)": 117.38,
      "step": 64175,
      "train_speed(iter/s)": 1.636261
    },
    {
      "acc": 0.67016363,
      "epoch": 1.628107559614409,
      "grad_norm": 7.21875,
      "learning_rate": 9.158815893937268e-07,
      "loss": 1.61708374,
      "memory(GiB)": 117.38,
      "step": 64180,
      "train_speed(iter/s)": 1.636275
    },
    {
      "acc": 0.66039228,
      "epoch": 1.628234398782344,
      "grad_norm": 6.875,
      "learning_rate": 9.152767414290115e-07,
      "loss": 1.6024477,
      "memory(GiB)": 117.38,
      "step": 64185,
      "train_speed(iter/s)": 1.63629
    },
    {
      "acc": 0.66371355,
      "epoch": 1.628361237950279,
      "grad_norm": 5.875,
      "learning_rate": 9.146720731286623e-07,
      "loss": 1.56843948,
      "memory(GiB)": 117.38,
      "step": 64190,
      "train_speed(iter/s)": 1.636303
    },
    {
      "acc": 0.65737667,
      "epoch": 1.628488077118214,
      "grad_norm": 6.65625,
      "learning_rate": 9.140675845192754e-07,
      "loss": 1.64151211,
      "memory(GiB)": 117.38,
      "step": 64195,
      "train_speed(iter/s)": 1.636107
    },
    {
      "acc": 0.65039558,
      "epoch": 1.6286149162861492,
      "grad_norm": 5.5,
      "learning_rate": 9.134632756274381e-07,
      "loss": 1.55718536,
      "memory(GiB)": 117.38,
      "step": 64200,
      "train_speed(iter/s)": 1.636121
    },
    {
      "acc": 0.66105261,
      "epoch": 1.6287417554540842,
      "grad_norm": 5.96875,
      "learning_rate": 9.12859146479731e-07,
      "loss": 1.63983612,
      "memory(GiB)": 117.38,
      "step": 64205,
      "train_speed(iter/s)": 1.636134
    },
    {
      "acc": 0.66798487,
      "epoch": 1.6288685946220194,
      "grad_norm": 5.9375,
      "learning_rate": 9.122551971027266e-07,
      "loss": 1.56808758,
      "memory(GiB)": 117.38,
      "step": 64210,
      "train_speed(iter/s)": 1.636148
    },
    {
      "acc": 0.64251666,
      "epoch": 1.6289954337899544,
      "grad_norm": 5.8125,
      "learning_rate": 9.116514275229892e-07,
      "loss": 1.65047455,
      "memory(GiB)": 117.38,
      "step": 64215,
      "train_speed(iter/s)": 1.636162
    },
    {
      "acc": 0.67602401,
      "epoch": 1.6291222729578894,
      "grad_norm": 5.25,
      "learning_rate": 9.110478377670751e-07,
      "loss": 1.54542303,
      "memory(GiB)": 117.38,
      "step": 64220,
      "train_speed(iter/s)": 1.636176
    },
    {
      "acc": 0.68549647,
      "epoch": 1.6292491121258244,
      "grad_norm": 6.21875,
      "learning_rate": 9.104444278615327e-07,
      "loss": 1.52364578,
      "memory(GiB)": 117.38,
      "step": 64225,
      "train_speed(iter/s)": 1.63619
    },
    {
      "acc": 0.66653528,
      "epoch": 1.6293759512937596,
      "grad_norm": 5.875,
      "learning_rate": 9.098411978329031e-07,
      "loss": 1.58440819,
      "memory(GiB)": 117.38,
      "step": 64230,
      "train_speed(iter/s)": 1.636205
    },
    {
      "acc": 0.68131003,
      "epoch": 1.6295027904616946,
      "grad_norm": 4.90625,
      "learning_rate": 9.092381477077189e-07,
      "loss": 1.53496552,
      "memory(GiB)": 117.38,
      "step": 64235,
      "train_speed(iter/s)": 1.636218
    },
    {
      "acc": 0.65149441,
      "epoch": 1.6296296296296298,
      "grad_norm": 5.125,
      "learning_rate": 9.086352775125046e-07,
      "loss": 1.63113022,
      "memory(GiB)": 117.38,
      "step": 64240,
      "train_speed(iter/s)": 1.636232
    },
    {
      "acc": 0.65774584,
      "epoch": 1.6297564687975648,
      "grad_norm": 5.34375,
      "learning_rate": 9.080325872737766e-07,
      "loss": 1.63655815,
      "memory(GiB)": 117.38,
      "step": 64245,
      "train_speed(iter/s)": 1.636243
    },
    {
      "acc": 0.67013774,
      "epoch": 1.6298833079654997,
      "grad_norm": 5.875,
      "learning_rate": 9.074300770180472e-07,
      "loss": 1.59241638,
      "memory(GiB)": 117.38,
      "step": 64250,
      "train_speed(iter/s)": 1.636257
    },
    {
      "acc": 0.66312466,
      "epoch": 1.6300101471334347,
      "grad_norm": 5.65625,
      "learning_rate": 9.068277467718134e-07,
      "loss": 1.5373848,
      "memory(GiB)": 117.38,
      "step": 64255,
      "train_speed(iter/s)": 1.636271
    },
    {
      "acc": 0.66634731,
      "epoch": 1.6301369863013697,
      "grad_norm": 6.0,
      "learning_rate": 9.062255965615701e-07,
      "loss": 1.57371206,
      "memory(GiB)": 117.38,
      "step": 64260,
      "train_speed(iter/s)": 1.636284
    },
    {
      "acc": 0.65798616,
      "epoch": 1.630263825469305,
      "grad_norm": 6.46875,
      "learning_rate": 9.056236264138013e-07,
      "loss": 1.5157733,
      "memory(GiB)": 117.38,
      "step": 64265,
      "train_speed(iter/s)": 1.636298
    },
    {
      "acc": 0.67357807,
      "epoch": 1.6303906646372401,
      "grad_norm": 6.03125,
      "learning_rate": 9.050218363549885e-07,
      "loss": 1.57546177,
      "memory(GiB)": 117.38,
      "step": 64270,
      "train_speed(iter/s)": 1.636311
    },
    {
      "acc": 0.65279284,
      "epoch": 1.6305175038051751,
      "grad_norm": 5.9375,
      "learning_rate": 9.044202264115958e-07,
      "loss": 1.58305664,
      "memory(GiB)": 117.38,
      "step": 64275,
      "train_speed(iter/s)": 1.636325
    },
    {
      "acc": 0.66002922,
      "epoch": 1.6306443429731101,
      "grad_norm": 5.71875,
      "learning_rate": 9.038187966100864e-07,
      "loss": 1.54904785,
      "memory(GiB)": 117.38,
      "step": 64280,
      "train_speed(iter/s)": 1.636338
    },
    {
      "acc": 0.65746746,
      "epoch": 1.630771182141045,
      "grad_norm": 5.8125,
      "learning_rate": 9.032175469769155e-07,
      "loss": 1.61362495,
      "memory(GiB)": 117.38,
      "step": 64285,
      "train_speed(iter/s)": 1.636351
    },
    {
      "acc": 0.65153384,
      "epoch": 1.63089802130898,
      "grad_norm": 6.375,
      "learning_rate": 9.026164775385294e-07,
      "loss": 1.53693304,
      "memory(GiB)": 117.38,
      "step": 64290,
      "train_speed(iter/s)": 1.636365
    },
    {
      "acc": 0.65331411,
      "epoch": 1.6310248604769153,
      "grad_norm": 5.625,
      "learning_rate": 9.020155883213627e-07,
      "loss": 1.55656967,
      "memory(GiB)": 117.38,
      "step": 64295,
      "train_speed(iter/s)": 1.636378
    },
    {
      "acc": 0.65635805,
      "epoch": 1.6311516996448503,
      "grad_norm": 5.5,
      "learning_rate": 9.014148793518451e-07,
      "loss": 1.55267391,
      "memory(GiB)": 117.38,
      "step": 64300,
      "train_speed(iter/s)": 1.636392
    },
    {
      "acc": 0.65309649,
      "epoch": 1.6312785388127855,
      "grad_norm": 5.40625,
      "learning_rate": 9.00814350656401e-07,
      "loss": 1.61679459,
      "memory(GiB)": 117.38,
      "step": 64305,
      "train_speed(iter/s)": 1.636405
    },
    {
      "acc": 0.63942447,
      "epoch": 1.6314053779807205,
      "grad_norm": 6.0,
      "learning_rate": 9.002140022614452e-07,
      "loss": 1.65091991,
      "memory(GiB)": 117.38,
      "step": 64310,
      "train_speed(iter/s)": 1.636418
    },
    {
      "acc": 0.6526722,
      "epoch": 1.6315322171486555,
      "grad_norm": 5.84375,
      "learning_rate": 8.996138341933786e-07,
      "loss": 1.6415432,
      "memory(GiB)": 117.38,
      "step": 64315,
      "train_speed(iter/s)": 1.636431
    },
    {
      "acc": 0.65593309,
      "epoch": 1.6316590563165905,
      "grad_norm": 5.90625,
      "learning_rate": 8.990138464786041e-07,
      "loss": 1.65720711,
      "memory(GiB)": 117.38,
      "step": 64320,
      "train_speed(iter/s)": 1.636444
    },
    {
      "acc": 0.66309729,
      "epoch": 1.6317858954845255,
      "grad_norm": 5.0,
      "learning_rate": 8.984140391435103e-07,
      "loss": 1.5448595,
      "memory(GiB)": 117.38,
      "step": 64325,
      "train_speed(iter/s)": 1.636457
    },
    {
      "acc": 0.64228854,
      "epoch": 1.6319127346524607,
      "grad_norm": 6.0625,
      "learning_rate": 8.978144122144805e-07,
      "loss": 1.61040192,
      "memory(GiB)": 117.38,
      "step": 64330,
      "train_speed(iter/s)": 1.63647
    },
    {
      "acc": 0.64958572,
      "epoch": 1.6320395738203959,
      "grad_norm": 5.125,
      "learning_rate": 8.972149657178852e-07,
      "loss": 1.57734375,
      "memory(GiB)": 117.38,
      "step": 64335,
      "train_speed(iter/s)": 1.636483
    },
    {
      "acc": 0.63760729,
      "epoch": 1.6321664129883309,
      "grad_norm": 7.71875,
      "learning_rate": 8.966156996800951e-07,
      "loss": 1.68630714,
      "memory(GiB)": 117.38,
      "step": 64340,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.65831456,
      "epoch": 1.6322932521562659,
      "grad_norm": 6.0,
      "learning_rate": 8.960166141274662e-07,
      "loss": 1.61654854,
      "memory(GiB)": 117.38,
      "step": 64345,
      "train_speed(iter/s)": 1.636511
    },
    {
      "acc": 0.65944462,
      "epoch": 1.6324200913242009,
      "grad_norm": 6.3125,
      "learning_rate": 8.954177090863497e-07,
      "loss": 1.5868371,
      "memory(GiB)": 117.38,
      "step": 64350,
      "train_speed(iter/s)": 1.636524
    },
    {
      "acc": 0.64768248,
      "epoch": 1.6325469304921358,
      "grad_norm": 6.96875,
      "learning_rate": 8.948189845830879e-07,
      "loss": 1.61492577,
      "memory(GiB)": 117.38,
      "step": 64355,
      "train_speed(iter/s)": 1.636537
    },
    {
      "acc": 0.65644655,
      "epoch": 1.632673769660071,
      "grad_norm": 6.125,
      "learning_rate": 8.942204406440159e-07,
      "loss": 1.58328247,
      "memory(GiB)": 117.38,
      "step": 64360,
      "train_speed(iter/s)": 1.636551
    },
    {
      "acc": 0.65405083,
      "epoch": 1.632800608828006,
      "grad_norm": 6.1875,
      "learning_rate": 8.936220772954595e-07,
      "loss": 1.65491486,
      "memory(GiB)": 117.38,
      "step": 64365,
      "train_speed(iter/s)": 1.636565
    },
    {
      "acc": 0.66227741,
      "epoch": 1.6329274479959413,
      "grad_norm": 5.0,
      "learning_rate": 8.930238945637381e-07,
      "loss": 1.63686008,
      "memory(GiB)": 117.38,
      "step": 64370,
      "train_speed(iter/s)": 1.636579
    },
    {
      "acc": 0.66622677,
      "epoch": 1.6330542871638762,
      "grad_norm": 7.65625,
      "learning_rate": 8.924258924751617e-07,
      "loss": 1.53165569,
      "memory(GiB)": 117.38,
      "step": 64375,
      "train_speed(iter/s)": 1.636593
    },
    {
      "acc": 0.66397347,
      "epoch": 1.6331811263318112,
      "grad_norm": 5.40625,
      "learning_rate": 8.918280710560339e-07,
      "loss": 1.51180067,
      "memory(GiB)": 117.38,
      "step": 64380,
      "train_speed(iter/s)": 1.636606
    },
    {
      "acc": 0.65786042,
      "epoch": 1.6333079654997462,
      "grad_norm": 5.6875,
      "learning_rate": 8.912304303326491e-07,
      "loss": 1.56389637,
      "memory(GiB)": 117.38,
      "step": 64385,
      "train_speed(iter/s)": 1.63662
    },
    {
      "acc": 0.65657158,
      "epoch": 1.6334348046676814,
      "grad_norm": 5.46875,
      "learning_rate": 8.906329703312943e-07,
      "loss": 1.61333218,
      "memory(GiB)": 117.38,
      "step": 64390,
      "train_speed(iter/s)": 1.636634
    },
    {
      "acc": 0.65686479,
      "epoch": 1.6335616438356164,
      "grad_norm": 5.53125,
      "learning_rate": 8.900356910782487e-07,
      "loss": 1.58123417,
      "memory(GiB)": 117.38,
      "step": 64395,
      "train_speed(iter/s)": 1.636647
    },
    {
      "acc": 0.6510046,
      "epoch": 1.6336884830035516,
      "grad_norm": 6.15625,
      "learning_rate": 8.894385925997828e-07,
      "loss": 1.63838387,
      "memory(GiB)": 117.38,
      "step": 64400,
      "train_speed(iter/s)": 1.636659
    },
    {
      "acc": 0.66055756,
      "epoch": 1.6338153221714866,
      "grad_norm": 5.0625,
      "learning_rate": 8.88841674922159e-07,
      "loss": 1.55805283,
      "memory(GiB)": 117.38,
      "step": 64405,
      "train_speed(iter/s)": 1.636673
    },
    {
      "acc": 0.65006514,
      "epoch": 1.6339421613394216,
      "grad_norm": 5.21875,
      "learning_rate": 8.882449380716351e-07,
      "loss": 1.63462143,
      "memory(GiB)": 117.38,
      "step": 64410,
      "train_speed(iter/s)": 1.636685
    },
    {
      "acc": 0.67003059,
      "epoch": 1.6340690005073566,
      "grad_norm": 5.84375,
      "learning_rate": 8.876483820744558e-07,
      "loss": 1.53705578,
      "memory(GiB)": 117.38,
      "step": 64415,
      "train_speed(iter/s)": 1.636697
    },
    {
      "acc": 0.6776444,
      "epoch": 1.6341958396752916,
      "grad_norm": 6.53125,
      "learning_rate": 8.8705200695686e-07,
      "loss": 1.50939484,
      "memory(GiB)": 117.38,
      "step": 64420,
      "train_speed(iter/s)": 1.63671
    },
    {
      "acc": 0.65658298,
      "epoch": 1.6343226788432268,
      "grad_norm": 5.46875,
      "learning_rate": 8.864558127450807e-07,
      "loss": 1.64592018,
      "memory(GiB)": 117.38,
      "step": 64425,
      "train_speed(iter/s)": 1.636724
    },
    {
      "acc": 0.67021289,
      "epoch": 1.634449518011162,
      "grad_norm": 4.875,
      "learning_rate": 8.858597994653417e-07,
      "loss": 1.56698513,
      "memory(GiB)": 117.38,
      "step": 64430,
      "train_speed(iter/s)": 1.636737
    },
    {
      "acc": 0.6748867,
      "epoch": 1.634576357179097,
      "grad_norm": 5.90625,
      "learning_rate": 8.852639671438562e-07,
      "loss": 1.50503607,
      "memory(GiB)": 117.38,
      "step": 64435,
      "train_speed(iter/s)": 1.636749
    },
    {
      "acc": 0.65975094,
      "epoch": 1.634703196347032,
      "grad_norm": 4.6875,
      "learning_rate": 8.846683158068309e-07,
      "loss": 1.53684912,
      "memory(GiB)": 117.38,
      "step": 64440,
      "train_speed(iter/s)": 1.636762
    },
    {
      "acc": 0.65763922,
      "epoch": 1.634830035514967,
      "grad_norm": 7.65625,
      "learning_rate": 8.840728454804676e-07,
      "loss": 1.57818031,
      "memory(GiB)": 117.38,
      "step": 64445,
      "train_speed(iter/s)": 1.636774
    },
    {
      "acc": 0.66745706,
      "epoch": 1.634956874682902,
      "grad_norm": 5.625,
      "learning_rate": 8.834775561909587e-07,
      "loss": 1.56720648,
      "memory(GiB)": 117.38,
      "step": 64450,
      "train_speed(iter/s)": 1.636787
    },
    {
      "acc": 0.64632545,
      "epoch": 1.6350837138508372,
      "grad_norm": 5.1875,
      "learning_rate": 8.828824479644827e-07,
      "loss": 1.7071949,
      "memory(GiB)": 117.38,
      "step": 64455,
      "train_speed(iter/s)": 1.636799
    },
    {
      "acc": 0.64682975,
      "epoch": 1.6352105530187722,
      "grad_norm": 6.9375,
      "learning_rate": 8.822875208272203e-07,
      "loss": 1.57792473,
      "memory(GiB)": 117.38,
      "step": 64460,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.64925599,
      "epoch": 1.6353373921867074,
      "grad_norm": 7.0625,
      "learning_rate": 8.816927748053361e-07,
      "loss": 1.63652573,
      "memory(GiB)": 117.38,
      "step": 64465,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.64419012,
      "epoch": 1.6354642313546424,
      "grad_norm": 6.03125,
      "learning_rate": 8.810982099249926e-07,
      "loss": 1.66783848,
      "memory(GiB)": 117.38,
      "step": 64470,
      "train_speed(iter/s)": 1.63684
    },
    {
      "acc": 0.66090884,
      "epoch": 1.6355910705225774,
      "grad_norm": 5.96875,
      "learning_rate": 8.805038262123361e-07,
      "loss": 1.61365662,
      "memory(GiB)": 117.38,
      "step": 64475,
      "train_speed(iter/s)": 1.636853
    },
    {
      "acc": 0.6645216,
      "epoch": 1.6357179096905123,
      "grad_norm": 5.9375,
      "learning_rate": 8.799096236935156e-07,
      "loss": 1.60325813,
      "memory(GiB)": 117.38,
      "step": 64480,
      "train_speed(iter/s)": 1.636865
    },
    {
      "acc": 0.65705695,
      "epoch": 1.6358447488584473,
      "grad_norm": 5.625,
      "learning_rate": 8.793156023946641e-07,
      "loss": 1.60500793,
      "memory(GiB)": 117.38,
      "step": 64485,
      "train_speed(iter/s)": 1.636879
    },
    {
      "acc": 0.65052996,
      "epoch": 1.6359715880263825,
      "grad_norm": 5.9375,
      "learning_rate": 8.787217623419104e-07,
      "loss": 1.64182587,
      "memory(GiB)": 117.38,
      "step": 64490,
      "train_speed(iter/s)": 1.636892
    },
    {
      "acc": 0.66665125,
      "epoch": 1.6360984271943178,
      "grad_norm": 6.03125,
      "learning_rate": 8.781281035613737e-07,
      "loss": 1.56315975,
      "memory(GiB)": 117.38,
      "step": 64495,
      "train_speed(iter/s)": 1.636906
    },
    {
      "acc": 0.65590496,
      "epoch": 1.6362252663622527,
      "grad_norm": 5.53125,
      "learning_rate": 8.775346260791656e-07,
      "loss": 1.6054451,
      "memory(GiB)": 117.38,
      "step": 64500,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.6460043,
      "epoch": 1.6363521055301877,
      "grad_norm": 6.59375,
      "learning_rate": 8.769413299213903e-07,
      "loss": 1.64523582,
      "memory(GiB)": 117.38,
      "step": 64505,
      "train_speed(iter/s)": 1.636931
    },
    {
      "acc": 0.65839362,
      "epoch": 1.6364789446981227,
      "grad_norm": 7.15625,
      "learning_rate": 8.763482151141434e-07,
      "loss": 1.60169888,
      "memory(GiB)": 117.38,
      "step": 64510,
      "train_speed(iter/s)": 1.636944
    },
    {
      "acc": 0.65998778,
      "epoch": 1.6366057838660577,
      "grad_norm": 5.625,
      "learning_rate": 8.75755281683513e-07,
      "loss": 1.59584618,
      "memory(GiB)": 117.38,
      "step": 64515,
      "train_speed(iter/s)": 1.636957
    },
    {
      "acc": 0.66168942,
      "epoch": 1.636732623033993,
      "grad_norm": 6.21875,
      "learning_rate": 8.751625296555782e-07,
      "loss": 1.57391157,
      "memory(GiB)": 117.38,
      "step": 64520,
      "train_speed(iter/s)": 1.63697
    },
    {
      "acc": 0.65584211,
      "epoch": 1.636859462201928,
      "grad_norm": 5.6875,
      "learning_rate": 8.745699590564122e-07,
      "loss": 1.59686718,
      "memory(GiB)": 117.38,
      "step": 64525,
      "train_speed(iter/s)": 1.636982
    },
    {
      "acc": 0.66042361,
      "epoch": 1.6369863013698631,
      "grad_norm": 7.0,
      "learning_rate": 8.739775699120773e-07,
      "loss": 1.54451761,
      "memory(GiB)": 117.38,
      "step": 64530,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.65598269,
      "epoch": 1.637113140537798,
      "grad_norm": 6.53125,
      "learning_rate": 8.733853622486305e-07,
      "loss": 1.57635136,
      "memory(GiB)": 117.38,
      "step": 64535,
      "train_speed(iter/s)": 1.637009
    },
    {
      "acc": 0.66244283,
      "epoch": 1.637239979705733,
      "grad_norm": 5.8125,
      "learning_rate": 8.727933360921198e-07,
      "loss": 1.54611378,
      "memory(GiB)": 117.38,
      "step": 64540,
      "train_speed(iter/s)": 1.637022
    },
    {
      "acc": 0.6711031,
      "epoch": 1.637366818873668,
      "grad_norm": 5.09375,
      "learning_rate": 8.722014914685834e-07,
      "loss": 1.55091829,
      "memory(GiB)": 117.38,
      "step": 64545,
      "train_speed(iter/s)": 1.637035
    },
    {
      "acc": 0.64969845,
      "epoch": 1.6374936580416033,
      "grad_norm": 8.25,
      "learning_rate": 8.71609828404057e-07,
      "loss": 1.70320358,
      "memory(GiB)": 117.38,
      "step": 64550,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.65629444,
      "epoch": 1.6376204972095383,
      "grad_norm": 5.1875,
      "learning_rate": 8.710183469245614e-07,
      "loss": 1.61533871,
      "memory(GiB)": 117.38,
      "step": 64555,
      "train_speed(iter/s)": 1.63706
    },
    {
      "acc": 0.66461658,
      "epoch": 1.6377473363774735,
      "grad_norm": 5.28125,
      "learning_rate": 8.704270470561132e-07,
      "loss": 1.52329006,
      "memory(GiB)": 117.38,
      "step": 64560,
      "train_speed(iter/s)": 1.637073
    },
    {
      "acc": 0.65518188,
      "epoch": 1.6378741755454085,
      "grad_norm": 5.71875,
      "learning_rate": 8.698359288247194e-07,
      "loss": 1.57330151,
      "memory(GiB)": 117.38,
      "step": 64565,
      "train_speed(iter/s)": 1.637087
    },
    {
      "acc": 0.65063562,
      "epoch": 1.6380010147133435,
      "grad_norm": 7.0625,
      "learning_rate": 8.692449922563839e-07,
      "loss": 1.63121166,
      "memory(GiB)": 117.38,
      "step": 64570,
      "train_speed(iter/s)": 1.6371
    },
    {
      "acc": 0.65189147,
      "epoch": 1.6381278538812785,
      "grad_norm": 6.5,
      "learning_rate": 8.686542373770951e-07,
      "loss": 1.63193474,
      "memory(GiB)": 117.38,
      "step": 64575,
      "train_speed(iter/s)": 1.637113
    },
    {
      "acc": 0.67223182,
      "epoch": 1.6382546930492135,
      "grad_norm": 5.125,
      "learning_rate": 8.680636642128365e-07,
      "loss": 1.58498354,
      "memory(GiB)": 117.38,
      "step": 64580,
      "train_speed(iter/s)": 1.637127
    },
    {
      "acc": 0.65705099,
      "epoch": 1.6383815322171487,
      "grad_norm": 5.21875,
      "learning_rate": 8.674732727895874e-07,
      "loss": 1.54217386,
      "memory(GiB)": 117.38,
      "step": 64585,
      "train_speed(iter/s)": 1.63714
    },
    {
      "acc": 0.64737816,
      "epoch": 1.6385083713850839,
      "grad_norm": 5.59375,
      "learning_rate": 8.668830631333147e-07,
      "loss": 1.75328445,
      "memory(GiB)": 117.38,
      "step": 64590,
      "train_speed(iter/s)": 1.637155
    },
    {
      "acc": 0.65661049,
      "epoch": 1.6386352105530189,
      "grad_norm": 5.84375,
      "learning_rate": 8.662930352699761e-07,
      "loss": 1.60997772,
      "memory(GiB)": 117.38,
      "step": 64595,
      "train_speed(iter/s)": 1.637167
    },
    {
      "acc": 0.67823992,
      "epoch": 1.6387620497209539,
      "grad_norm": 5.90625,
      "learning_rate": 8.657031892255263e-07,
      "loss": 1.54094582,
      "memory(GiB)": 117.38,
      "step": 64600,
      "train_speed(iter/s)": 1.637181
    },
    {
      "acc": 0.67424746,
      "epoch": 1.6388888888888888,
      "grad_norm": 6.0,
      "learning_rate": 8.651135250259091e-07,
      "loss": 1.50796165,
      "memory(GiB)": 117.38,
      "step": 64605,
      "train_speed(iter/s)": 1.637194
    },
    {
      "acc": 0.64996595,
      "epoch": 1.6390157280568238,
      "grad_norm": 7.375,
      "learning_rate": 8.645240426970608e-07,
      "loss": 1.60432434,
      "memory(GiB)": 117.38,
      "step": 64610,
      "train_speed(iter/s)": 1.637208
    },
    {
      "acc": 0.64492421,
      "epoch": 1.639142567224759,
      "grad_norm": 5.59375,
      "learning_rate": 8.639347422649058e-07,
      "loss": 1.59048653,
      "memory(GiB)": 117.38,
      "step": 64615,
      "train_speed(iter/s)": 1.637221
    },
    {
      "acc": 0.66685057,
      "epoch": 1.639269406392694,
      "grad_norm": 5.40625,
      "learning_rate": 8.633456237553689e-07,
      "loss": 1.52912235,
      "memory(GiB)": 117.38,
      "step": 64620,
      "train_speed(iter/s)": 1.637235
    },
    {
      "acc": 0.65340443,
      "epoch": 1.6393962455606292,
      "grad_norm": 5.71875,
      "learning_rate": 8.627566871943604e-07,
      "loss": 1.60479507,
      "memory(GiB)": 117.38,
      "step": 64625,
      "train_speed(iter/s)": 1.637248
    },
    {
      "acc": 0.65466242,
      "epoch": 1.6395230847285642,
      "grad_norm": 5.3125,
      "learning_rate": 8.621679326077836e-07,
      "loss": 1.56817703,
      "memory(GiB)": 117.38,
      "step": 64630,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.67594848,
      "epoch": 1.6396499238964992,
      "grad_norm": 6.5,
      "learning_rate": 8.61579360021536e-07,
      "loss": 1.51457767,
      "memory(GiB)": 117.38,
      "step": 64635,
      "train_speed(iter/s)": 1.637275
    },
    {
      "acc": 0.66544476,
      "epoch": 1.6397767630644342,
      "grad_norm": 5.71875,
      "learning_rate": 8.609909694615043e-07,
      "loss": 1.60106907,
      "memory(GiB)": 117.38,
      "step": 64640,
      "train_speed(iter/s)": 1.637288
    },
    {
      "acc": 0.65676861,
      "epoch": 1.6399036022323692,
      "grad_norm": 5.875,
      "learning_rate": 8.604027609535693e-07,
      "loss": 1.56086674,
      "memory(GiB)": 117.38,
      "step": 64645,
      "train_speed(iter/s)": 1.637301
    },
    {
      "acc": 0.6629344,
      "epoch": 1.6400304414003044,
      "grad_norm": 5.875,
      "learning_rate": 8.598147345236029e-07,
      "loss": 1.55487366,
      "memory(GiB)": 117.38,
      "step": 64650,
      "train_speed(iter/s)": 1.637315
    },
    {
      "acc": 0.64780807,
      "epoch": 1.6401572805682396,
      "grad_norm": 5.78125,
      "learning_rate": 8.592268901974688e-07,
      "loss": 1.58347492,
      "memory(GiB)": 117.38,
      "step": 64655,
      "train_speed(iter/s)": 1.637329
    },
    {
      "acc": 0.65043926,
      "epoch": 1.6402841197361746,
      "grad_norm": 7.75,
      "learning_rate": 8.586392280010237e-07,
      "loss": 1.66943111,
      "memory(GiB)": 117.38,
      "step": 64660,
      "train_speed(iter/s)": 1.637342
    },
    {
      "acc": 0.63919687,
      "epoch": 1.6404109589041096,
      "grad_norm": 6.75,
      "learning_rate": 8.580517479601147e-07,
      "loss": 1.67320137,
      "memory(GiB)": 117.38,
      "step": 64665,
      "train_speed(iter/s)": 1.637356
    },
    {
      "acc": 0.66788187,
      "epoch": 1.6405377980720446,
      "grad_norm": 7.3125,
      "learning_rate": 8.57464450100583e-07,
      "loss": 1.5314394,
      "memory(GiB)": 117.38,
      "step": 64670,
      "train_speed(iter/s)": 1.637369
    },
    {
      "acc": 0.66296434,
      "epoch": 1.6406646372399796,
      "grad_norm": 8.5,
      "learning_rate": 8.568773344482595e-07,
      "loss": 1.58386822,
      "memory(GiB)": 117.38,
      "step": 64675,
      "train_speed(iter/s)": 1.637382
    },
    {
      "acc": 0.66180706,
      "epoch": 1.6407914764079148,
      "grad_norm": 6.0625,
      "learning_rate": 8.562904010289685e-07,
      "loss": 1.59790993,
      "memory(GiB)": 117.38,
      "step": 64680,
      "train_speed(iter/s)": 1.637395
    },
    {
      "acc": 0.65477934,
      "epoch": 1.6409183155758498,
      "grad_norm": 6.0,
      "learning_rate": 8.557036498685245e-07,
      "loss": 1.60289841,
      "memory(GiB)": 117.38,
      "step": 64685,
      "train_speed(iter/s)": 1.637409
    },
    {
      "acc": 0.65912018,
      "epoch": 1.641045154743785,
      "grad_norm": 5.625,
      "learning_rate": 8.551170809927394e-07,
      "loss": 1.62870331,
      "memory(GiB)": 117.38,
      "step": 64690,
      "train_speed(iter/s)": 1.637424
    },
    {
      "acc": 0.6732944,
      "epoch": 1.64117199391172,
      "grad_norm": 6.40625,
      "learning_rate": 8.545306944274101e-07,
      "loss": 1.53669538,
      "memory(GiB)": 117.38,
      "step": 64695,
      "train_speed(iter/s)": 1.637437
    },
    {
      "acc": 0.66373663,
      "epoch": 1.641298833079655,
      "grad_norm": 5.28125,
      "learning_rate": 8.539444901983284e-07,
      "loss": 1.55918283,
      "memory(GiB)": 117.38,
      "step": 64700,
      "train_speed(iter/s)": 1.63745
    },
    {
      "acc": 0.66447511,
      "epoch": 1.64142567224759,
      "grad_norm": 6.90625,
      "learning_rate": 8.533584683312779e-07,
      "loss": 1.60827999,
      "memory(GiB)": 117.38,
      "step": 64705,
      "train_speed(iter/s)": 1.637464
    },
    {
      "acc": 0.64863806,
      "epoch": 1.6415525114155252,
      "grad_norm": 7.03125,
      "learning_rate": 8.527726288520377e-07,
      "loss": 1.64336967,
      "memory(GiB)": 117.38,
      "step": 64710,
      "train_speed(iter/s)": 1.637479
    },
    {
      "acc": 0.65318751,
      "epoch": 1.6416793505834602,
      "grad_norm": 5.65625,
      "learning_rate": 8.521869717863723e-07,
      "loss": 1.61412544,
      "memory(GiB)": 117.38,
      "step": 64715,
      "train_speed(iter/s)": 1.637492
    },
    {
      "acc": 0.65827713,
      "epoch": 1.6418061897513954,
      "grad_norm": 5.78125,
      "learning_rate": 8.516014971600411e-07,
      "loss": 1.61229076,
      "memory(GiB)": 117.38,
      "step": 64720,
      "train_speed(iter/s)": 1.637506
    },
    {
      "acc": 0.64206033,
      "epoch": 1.6419330289193304,
      "grad_norm": 6.8125,
      "learning_rate": 8.510162049987986e-07,
      "loss": 1.64799881,
      "memory(GiB)": 117.38,
      "step": 64725,
      "train_speed(iter/s)": 1.63752
    },
    {
      "acc": 0.6657445,
      "epoch": 1.6420598680872653,
      "grad_norm": 5.375,
      "learning_rate": 8.504310953283884e-07,
      "loss": 1.59640121,
      "memory(GiB)": 117.38,
      "step": 64730,
      "train_speed(iter/s)": 1.637534
    },
    {
      "acc": 0.65431252,
      "epoch": 1.6421867072552003,
      "grad_norm": 6.4375,
      "learning_rate": 8.49846168174544e-07,
      "loss": 1.60152779,
      "memory(GiB)": 117.38,
      "step": 64735,
      "train_speed(iter/s)": 1.637547
    },
    {
      "acc": 0.66547089,
      "epoch": 1.6423135464231353,
      "grad_norm": 5.8125,
      "learning_rate": 8.492614235629932e-07,
      "loss": 1.61319695,
      "memory(GiB)": 117.38,
      "step": 64740,
      "train_speed(iter/s)": 1.637561
    },
    {
      "acc": 0.66679211,
      "epoch": 1.6424403855910705,
      "grad_norm": 6.40625,
      "learning_rate": 8.486768615194579e-07,
      "loss": 1.65451031,
      "memory(GiB)": 117.38,
      "step": 64745,
      "train_speed(iter/s)": 1.637574
    },
    {
      "acc": 0.67732229,
      "epoch": 1.6425672247590057,
      "grad_norm": 5.53125,
      "learning_rate": 8.480924820696495e-07,
      "loss": 1.49379568,
      "memory(GiB)": 117.38,
      "step": 64750,
      "train_speed(iter/s)": 1.637588
    },
    {
      "acc": 0.65555902,
      "epoch": 1.6426940639269407,
      "grad_norm": 5.28125,
      "learning_rate": 8.475082852392685e-07,
      "loss": 1.63650131,
      "memory(GiB)": 117.38,
      "step": 64755,
      "train_speed(iter/s)": 1.637601
    },
    {
      "acc": 0.66189566,
      "epoch": 1.6428209030948757,
      "grad_norm": 5.21875,
      "learning_rate": 8.469242710540138e-07,
      "loss": 1.57348881,
      "memory(GiB)": 117.38,
      "step": 64760,
      "train_speed(iter/s)": 1.637615
    },
    {
      "acc": 0.64581356,
      "epoch": 1.6429477422628107,
      "grad_norm": 6.0,
      "learning_rate": 8.463404395395713e-07,
      "loss": 1.68014965,
      "memory(GiB)": 117.38,
      "step": 64765,
      "train_speed(iter/s)": 1.637629
    },
    {
      "acc": 0.65202537,
      "epoch": 1.6430745814307457,
      "grad_norm": 5.875,
      "learning_rate": 8.45756790721623e-07,
      "loss": 1.56607771,
      "memory(GiB)": 117.38,
      "step": 64770,
      "train_speed(iter/s)": 1.637642
    },
    {
      "acc": 0.64969139,
      "epoch": 1.643201420598681,
      "grad_norm": 6.3125,
      "learning_rate": 8.451733246258354e-07,
      "loss": 1.63686638,
      "memory(GiB)": 117.38,
      "step": 64775,
      "train_speed(iter/s)": 1.637656
    },
    {
      "acc": 0.6547502,
      "epoch": 1.643328259766616,
      "grad_norm": 5.84375,
      "learning_rate": 8.445900412778768e-07,
      "loss": 1.65765114,
      "memory(GiB)": 117.38,
      "step": 64780,
      "train_speed(iter/s)": 1.63767
    },
    {
      "acc": 0.6582253,
      "epoch": 1.643455098934551,
      "grad_norm": 4.75,
      "learning_rate": 8.440069407034002e-07,
      "loss": 1.54100533,
      "memory(GiB)": 117.38,
      "step": 64785,
      "train_speed(iter/s)": 1.637684
    },
    {
      "acc": 0.67139196,
      "epoch": 1.643581938102486,
      "grad_norm": 6.9375,
      "learning_rate": 8.434240229280538e-07,
      "loss": 1.55577374,
      "memory(GiB)": 117.38,
      "step": 64790,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.66159654,
      "epoch": 1.643708777270421,
      "grad_norm": 5.59375,
      "learning_rate": 8.428412879774767e-07,
      "loss": 1.65142803,
      "memory(GiB)": 117.38,
      "step": 64795,
      "train_speed(iter/s)": 1.637711
    },
    {
      "acc": 0.66612983,
      "epoch": 1.643835616438356,
      "grad_norm": 5.5,
      "learning_rate": 8.422587358772999e-07,
      "loss": 1.57532177,
      "memory(GiB)": 128.51,
      "step": 64800,
      "train_speed(iter/s)": 1.637723
    },
    {
      "acc": 0.65604439,
      "epoch": 1.643962455606291,
      "grad_norm": 6.375,
      "learning_rate": 8.416763666531468e-07,
      "loss": 1.61523132,
      "memory(GiB)": 128.51,
      "step": 64805,
      "train_speed(iter/s)": 1.637736
    },
    {
      "acc": 0.65885863,
      "epoch": 1.6440892947742263,
      "grad_norm": 6.15625,
      "learning_rate": 8.410941803306327e-07,
      "loss": 1.52201481,
      "memory(GiB)": 128.51,
      "step": 64810,
      "train_speed(iter/s)": 1.63775
    },
    {
      "acc": 0.65644484,
      "epoch": 1.6442161339421615,
      "grad_norm": 6.09375,
      "learning_rate": 8.405121769353647e-07,
      "loss": 1.64983139,
      "memory(GiB)": 128.51,
      "step": 64815,
      "train_speed(iter/s)": 1.637764
    },
    {
      "acc": 0.6522748,
      "epoch": 1.6443429731100965,
      "grad_norm": 5.375,
      "learning_rate": 8.399303564929423e-07,
      "loss": 1.57061501,
      "memory(GiB)": 128.51,
      "step": 64820,
      "train_speed(iter/s)": 1.637777
    },
    {
      "acc": 0.65681,
      "epoch": 1.6444698122780315,
      "grad_norm": 6.5,
      "learning_rate": 8.393487190289562e-07,
      "loss": 1.55727634,
      "memory(GiB)": 128.51,
      "step": 64825,
      "train_speed(iter/s)": 1.637791
    },
    {
      "acc": 0.64951587,
      "epoch": 1.6445966514459665,
      "grad_norm": 5.25,
      "learning_rate": 8.387672645689887e-07,
      "loss": 1.62064781,
      "memory(GiB)": 128.51,
      "step": 64830,
      "train_speed(iter/s)": 1.637805
    },
    {
      "acc": 0.66965523,
      "epoch": 1.6447234906139014,
      "grad_norm": 6.125,
      "learning_rate": 8.381859931386161e-07,
      "loss": 1.56142445,
      "memory(GiB)": 128.51,
      "step": 64835,
      "train_speed(iter/s)": 1.637818
    },
    {
      "acc": 0.63712821,
      "epoch": 1.6448503297818367,
      "grad_norm": 5.9375,
      "learning_rate": 8.37604904763405e-07,
      "loss": 1.5931572,
      "memory(GiB)": 128.51,
      "step": 64840,
      "train_speed(iter/s)": 1.637832
    },
    {
      "acc": 0.64802895,
      "epoch": 1.6449771689497716,
      "grad_norm": 5.65625,
      "learning_rate": 8.370239994689123e-07,
      "loss": 1.58122149,
      "memory(GiB)": 128.51,
      "step": 64845,
      "train_speed(iter/s)": 1.637846
    },
    {
      "acc": 0.6716279,
      "epoch": 1.6451040081177069,
      "grad_norm": 5.78125,
      "learning_rate": 8.364432772806924e-07,
      "loss": 1.55457172,
      "memory(GiB)": 128.51,
      "step": 64850,
      "train_speed(iter/s)": 1.637636
    },
    {
      "acc": 0.66971855,
      "epoch": 1.6452308472856418,
      "grad_norm": 5.03125,
      "learning_rate": 8.358627382242857e-07,
      "loss": 1.5141942,
      "memory(GiB)": 128.51,
      "step": 64855,
      "train_speed(iter/s)": 1.63765
    },
    {
      "acc": 0.6456254,
      "epoch": 1.6453576864535768,
      "grad_norm": 8.25,
      "learning_rate": 8.352823823252254e-07,
      "loss": 1.65165844,
      "memory(GiB)": 128.51,
      "step": 64860,
      "train_speed(iter/s)": 1.637663
    },
    {
      "acc": 0.63329515,
      "epoch": 1.6454845256215118,
      "grad_norm": 5.34375,
      "learning_rate": 8.347022096090418e-07,
      "loss": 1.69572487,
      "memory(GiB)": 128.51,
      "step": 64865,
      "train_speed(iter/s)": 1.637676
    },
    {
      "acc": 0.66516323,
      "epoch": 1.645611364789447,
      "grad_norm": 6.96875,
      "learning_rate": 8.341222201012527e-07,
      "loss": 1.55688972,
      "memory(GiB)": 128.51,
      "step": 64870,
      "train_speed(iter/s)": 1.637689
    },
    {
      "acc": 0.65153027,
      "epoch": 1.645738203957382,
      "grad_norm": 6.21875,
      "learning_rate": 8.335424138273668e-07,
      "loss": 1.64163017,
      "memory(GiB)": 128.51,
      "step": 64875,
      "train_speed(iter/s)": 1.637702
    },
    {
      "acc": 0.65290432,
      "epoch": 1.6458650431253172,
      "grad_norm": 6.8125,
      "learning_rate": 8.329627908128857e-07,
      "loss": 1.60109043,
      "memory(GiB)": 128.51,
      "step": 64880,
      "train_speed(iter/s)": 1.637715
    },
    {
      "acc": 0.66773653,
      "epoch": 1.6459918822932522,
      "grad_norm": 6.4375,
      "learning_rate": 8.323833510833068e-07,
      "loss": 1.55382442,
      "memory(GiB)": 128.51,
      "step": 64885,
      "train_speed(iter/s)": 1.637729
    },
    {
      "acc": 0.65851893,
      "epoch": 1.6461187214611872,
      "grad_norm": 5.5,
      "learning_rate": 8.318040946641171e-07,
      "loss": 1.64109154,
      "memory(GiB)": 128.51,
      "step": 64890,
      "train_speed(iter/s)": 1.637743
    },
    {
      "acc": 0.66323276,
      "epoch": 1.6462455606291222,
      "grad_norm": 5.90625,
      "learning_rate": 8.312250215807898e-07,
      "loss": 1.6147213,
      "memory(GiB)": 128.51,
      "step": 64895,
      "train_speed(iter/s)": 1.637756
    },
    {
      "acc": 0.66610441,
      "epoch": 1.6463723997970572,
      "grad_norm": 5.90625,
      "learning_rate": 8.306461318587999e-07,
      "loss": 1.56968899,
      "memory(GiB)": 128.51,
      "step": 64900,
      "train_speed(iter/s)": 1.63777
    },
    {
      "acc": 0.63668189,
      "epoch": 1.6464992389649924,
      "grad_norm": 5.75,
      "learning_rate": 8.300674255236074e-07,
      "loss": 1.66660309,
      "memory(GiB)": 128.51,
      "step": 64905,
      "train_speed(iter/s)": 1.637784
    },
    {
      "acc": 0.6446743,
      "epoch": 1.6466260781329276,
      "grad_norm": 5.40625,
      "learning_rate": 8.294889026006686e-07,
      "loss": 1.64032745,
      "memory(GiB)": 128.51,
      "step": 64910,
      "train_speed(iter/s)": 1.637798
    },
    {
      "acc": 0.62644181,
      "epoch": 1.6467529173008626,
      "grad_norm": 6.1875,
      "learning_rate": 8.289105631154254e-07,
      "loss": 1.71175785,
      "memory(GiB)": 128.51,
      "step": 64915,
      "train_speed(iter/s)": 1.637812
    },
    {
      "acc": 0.65595312,
      "epoch": 1.6468797564687976,
      "grad_norm": 4.84375,
      "learning_rate": 8.283324070933196e-07,
      "loss": 1.59741316,
      "memory(GiB)": 128.51,
      "step": 64920,
      "train_speed(iter/s)": 1.637824
    },
    {
      "acc": 0.64044676,
      "epoch": 1.6470065956367326,
      "grad_norm": 5.4375,
      "learning_rate": 8.277544345597793e-07,
      "loss": 1.68474922,
      "memory(GiB)": 128.51,
      "step": 64925,
      "train_speed(iter/s)": 1.637838
    },
    {
      "acc": 0.63849506,
      "epoch": 1.6471334348046676,
      "grad_norm": 5.71875,
      "learning_rate": 8.271766455402269e-07,
      "loss": 1.68146992,
      "memory(GiB)": 128.51,
      "step": 64930,
      "train_speed(iter/s)": 1.63785
    },
    {
      "acc": 0.65692158,
      "epoch": 1.6472602739726028,
      "grad_norm": 5.4375,
      "learning_rate": 8.26599040060076e-07,
      "loss": 1.58570471,
      "memory(GiB)": 128.51,
      "step": 64935,
      "train_speed(iter/s)": 1.637863
    },
    {
      "acc": 0.65198889,
      "epoch": 1.6473871131405378,
      "grad_norm": 6.3125,
      "learning_rate": 8.260216181447323e-07,
      "loss": 1.57542915,
      "memory(GiB)": 128.51,
      "step": 64940,
      "train_speed(iter/s)": 1.637877
    },
    {
      "acc": 0.66240969,
      "epoch": 1.647513952308473,
      "grad_norm": 6.34375,
      "learning_rate": 8.254443798195932e-07,
      "loss": 1.52223663,
      "memory(GiB)": 128.51,
      "step": 64945,
      "train_speed(iter/s)": 1.637889
    },
    {
      "acc": 0.6646276,
      "epoch": 1.647640791476408,
      "grad_norm": 6.6875,
      "learning_rate": 8.248673251100481e-07,
      "loss": 1.61537323,
      "memory(GiB)": 128.51,
      "step": 64950,
      "train_speed(iter/s)": 1.637903
    },
    {
      "acc": 0.65987582,
      "epoch": 1.647767630644343,
      "grad_norm": 6.125,
      "learning_rate": 8.242904540414787e-07,
      "loss": 1.61644516,
      "memory(GiB)": 128.51,
      "step": 64955,
      "train_speed(iter/s)": 1.637916
    },
    {
      "acc": 0.65645065,
      "epoch": 1.647894469812278,
      "grad_norm": 6.75,
      "learning_rate": 8.237137666392586e-07,
      "loss": 1.5946147,
      "memory(GiB)": 128.51,
      "step": 64960,
      "train_speed(iter/s)": 1.637931
    },
    {
      "acc": 0.64823518,
      "epoch": 1.648021308980213,
      "grad_norm": 5.3125,
      "learning_rate": 8.231372629287526e-07,
      "loss": 1.65992336,
      "memory(GiB)": 128.51,
      "step": 64965,
      "train_speed(iter/s)": 1.637944
    },
    {
      "acc": 0.6629631,
      "epoch": 1.6481481481481481,
      "grad_norm": 5.9375,
      "learning_rate": 8.225609429353187e-07,
      "loss": 1.60226593,
      "memory(GiB)": 128.51,
      "step": 64970,
      "train_speed(iter/s)": 1.637958
    },
    {
      "acc": 0.67679634,
      "epoch": 1.6482749873160834,
      "grad_norm": 6.46875,
      "learning_rate": 8.219848066843051e-07,
      "loss": 1.52095451,
      "memory(GiB)": 128.51,
      "step": 64975,
      "train_speed(iter/s)": 1.637971
    },
    {
      "acc": 0.64920592,
      "epoch": 1.6484018264840183,
      "grad_norm": 5.40625,
      "learning_rate": 8.214088542010529e-07,
      "loss": 1.61804352,
      "memory(GiB)": 128.51,
      "step": 64980,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.64906683,
      "epoch": 1.6485286656519533,
      "grad_norm": 5.65625,
      "learning_rate": 8.208330855108943e-07,
      "loss": 1.61961708,
      "memory(GiB)": 128.51,
      "step": 64985,
      "train_speed(iter/s)": 1.637999
    },
    {
      "acc": 0.65655632,
      "epoch": 1.6486555048198883,
      "grad_norm": 6.40625,
      "learning_rate": 8.202575006391577e-07,
      "loss": 1.60711651,
      "memory(GiB)": 128.51,
      "step": 64990,
      "train_speed(iter/s)": 1.638012
    },
    {
      "acc": 0.65767536,
      "epoch": 1.6487823439878233,
      "grad_norm": 6.1875,
      "learning_rate": 8.19682099611156e-07,
      "loss": 1.51675682,
      "memory(GiB)": 128.51,
      "step": 64995,
      "train_speed(iter/s)": 1.638026
    },
    {
      "acc": 0.64426832,
      "epoch": 1.6489091831557585,
      "grad_norm": 5.375,
      "learning_rate": 8.191068824521998e-07,
      "loss": 1.64707069,
      "memory(GiB)": 128.51,
      "step": 65000,
      "train_speed(iter/s)": 1.638039
    },
    {
      "epoch": 1.6489091831557585,
      "eval_acc": 0.6462789517538639,
      "eval_loss": 1.5731784105300903,
      "eval_runtime": 58.7741,
      "eval_samples_per_second": 108.381,
      "eval_steps_per_second": 27.104,
      "step": 65000
    },
    {
      "acc": 0.64181662,
      "epoch": 1.6490360223236935,
      "grad_norm": 6.75,
      "learning_rate": 8.185318491875876e-07,
      "loss": 1.64678955,
      "memory(GiB)": 128.51,
      "step": 65005,
      "train_speed(iter/s)": 1.635451
    },
    {
      "acc": 0.67020946,
      "epoch": 1.6491628614916287,
      "grad_norm": 6.28125,
      "learning_rate": 8.179569998426162e-07,
      "loss": 1.56151409,
      "memory(GiB)": 128.51,
      "step": 65010,
      "train_speed(iter/s)": 1.635463
    },
    {
      "acc": 0.65577993,
      "epoch": 1.6492897006595637,
      "grad_norm": 6.40625,
      "learning_rate": 8.173823344425663e-07,
      "loss": 1.56394186,
      "memory(GiB)": 128.51,
      "step": 65015,
      "train_speed(iter/s)": 1.635476
    },
    {
      "acc": 0.66167994,
      "epoch": 1.6494165398274987,
      "grad_norm": 6.28125,
      "learning_rate": 8.168078530127138e-07,
      "loss": 1.5656702,
      "memory(GiB)": 128.51,
      "step": 65020,
      "train_speed(iter/s)": 1.635488
    },
    {
      "acc": 0.6601758,
      "epoch": 1.6495433789954337,
      "grad_norm": 4.875,
      "learning_rate": 8.162335555783301e-07,
      "loss": 1.59683399,
      "memory(GiB)": 128.51,
      "step": 65025,
      "train_speed(iter/s)": 1.6355
    },
    {
      "acc": 0.64978333,
      "epoch": 1.649670218163369,
      "grad_norm": 5.34375,
      "learning_rate": 8.156594421646752e-07,
      "loss": 1.59758263,
      "memory(GiB)": 128.51,
      "step": 65030,
      "train_speed(iter/s)": 1.635513
    },
    {
      "acc": 0.66675954,
      "epoch": 1.6497970573313039,
      "grad_norm": 5.59375,
      "learning_rate": 8.15085512796997e-07,
      "loss": 1.54028206,
      "memory(GiB)": 128.51,
      "step": 65035,
      "train_speed(iter/s)": 1.635525
    },
    {
      "acc": 0.6516674,
      "epoch": 1.649923896499239,
      "grad_norm": 5.25,
      "learning_rate": 8.145117675005431e-07,
      "loss": 1.57562218,
      "memory(GiB)": 128.51,
      "step": 65040,
      "train_speed(iter/s)": 1.635539
    },
    {
      "acc": 0.64809451,
      "epoch": 1.650050735667174,
      "grad_norm": 7.71875,
      "learning_rate": 8.13938206300549e-07,
      "loss": 1.66687222,
      "memory(GiB)": 128.51,
      "step": 65045,
      "train_speed(iter/s)": 1.635552
    },
    {
      "acc": 0.64147387,
      "epoch": 1.650177574835109,
      "grad_norm": 5.53125,
      "learning_rate": 8.133648292222435e-07,
      "loss": 1.62831974,
      "memory(GiB)": 128.51,
      "step": 65050,
      "train_speed(iter/s)": 1.635564
    },
    {
      "acc": 0.65581045,
      "epoch": 1.650304414003044,
      "grad_norm": 6.0,
      "learning_rate": 8.127916362908422e-07,
      "loss": 1.62946377,
      "memory(GiB)": 128.51,
      "step": 65055,
      "train_speed(iter/s)": 1.635577
    },
    {
      "acc": 0.64711485,
      "epoch": 1.650431253170979,
      "grad_norm": 5.84375,
      "learning_rate": 8.122186275315602e-07,
      "loss": 1.65491161,
      "memory(GiB)": 128.51,
      "step": 65060,
      "train_speed(iter/s)": 1.63559
    },
    {
      "acc": 0.64933577,
      "epoch": 1.6505580923389143,
      "grad_norm": 6.0625,
      "learning_rate": 8.116458029696e-07,
      "loss": 1.65905991,
      "memory(GiB)": 128.51,
      "step": 65065,
      "train_speed(iter/s)": 1.635602
    },
    {
      "acc": 0.65868735,
      "epoch": 1.6506849315068495,
      "grad_norm": 5.90625,
      "learning_rate": 8.110731626301577e-07,
      "loss": 1.59663143,
      "memory(GiB)": 128.51,
      "step": 65070,
      "train_speed(iter/s)": 1.635615
    },
    {
      "acc": 0.65231023,
      "epoch": 1.6508117706747845,
      "grad_norm": 4.875,
      "learning_rate": 8.105007065384191e-07,
      "loss": 1.57585659,
      "memory(GiB)": 128.51,
      "step": 65075,
      "train_speed(iter/s)": 1.635628
    },
    {
      "acc": 0.66797714,
      "epoch": 1.6509386098427195,
      "grad_norm": 5.28125,
      "learning_rate": 8.099284347195646e-07,
      "loss": 1.55437717,
      "memory(GiB)": 128.51,
      "step": 65080,
      "train_speed(iter/s)": 1.635641
    },
    {
      "acc": 0.67555838,
      "epoch": 1.6510654490106544,
      "grad_norm": 8.125,
      "learning_rate": 8.093563471987647e-07,
      "loss": 1.55635719,
      "memory(GiB)": 128.51,
      "step": 65085,
      "train_speed(iter/s)": 1.635653
    },
    {
      "acc": 0.67573853,
      "epoch": 1.6511922881785894,
      "grad_norm": 5.9375,
      "learning_rate": 8.087844440011828e-07,
      "loss": 1.50783558,
      "memory(GiB)": 128.51,
      "step": 65090,
      "train_speed(iter/s)": 1.635665
    },
    {
      "acc": 0.66688285,
      "epoch": 1.6513191273465246,
      "grad_norm": 4.875,
      "learning_rate": 8.082127251519733e-07,
      "loss": 1.57410526,
      "memory(GiB)": 128.51,
      "step": 65095,
      "train_speed(iter/s)": 1.635676
    },
    {
      "acc": 0.66919785,
      "epoch": 1.6514459665144596,
      "grad_norm": 5.59375,
      "learning_rate": 8.076411906762832e-07,
      "loss": 1.56066799,
      "memory(GiB)": 128.51,
      "step": 65100,
      "train_speed(iter/s)": 1.635689
    },
    {
      "acc": 0.66150475,
      "epoch": 1.6515728056823948,
      "grad_norm": 6.09375,
      "learning_rate": 8.070698405992511e-07,
      "loss": 1.65026131,
      "memory(GiB)": 128.51,
      "step": 65105,
      "train_speed(iter/s)": 1.635702
    },
    {
      "acc": 0.65168858,
      "epoch": 1.6516996448503298,
      "grad_norm": 6.1875,
      "learning_rate": 8.064986749460075e-07,
      "loss": 1.6122757,
      "memory(GiB)": 128.51,
      "step": 65110,
      "train_speed(iter/s)": 1.635714
    },
    {
      "acc": 0.65405087,
      "epoch": 1.6518264840182648,
      "grad_norm": 7.34375,
      "learning_rate": 8.059276937416744e-07,
      "loss": 1.63427658,
      "memory(GiB)": 128.51,
      "step": 65115,
      "train_speed(iter/s)": 1.635727
    },
    {
      "acc": 0.65276527,
      "epoch": 1.6519533231861998,
      "grad_norm": 6.625,
      "learning_rate": 8.053568970113667e-07,
      "loss": 1.66405487,
      "memory(GiB)": 128.51,
      "step": 65120,
      "train_speed(iter/s)": 1.63574
    },
    {
      "acc": 0.63322868,
      "epoch": 1.6520801623541348,
      "grad_norm": 4.875,
      "learning_rate": 8.047862847801896e-07,
      "loss": 1.70209656,
      "memory(GiB)": 128.51,
      "step": 65125,
      "train_speed(iter/s)": 1.635749
    },
    {
      "acc": 0.66581726,
      "epoch": 1.65220700152207,
      "grad_norm": 5.71875,
      "learning_rate": 8.042158570732444e-07,
      "loss": 1.52808619,
      "memory(GiB)": 128.51,
      "step": 65130,
      "train_speed(iter/s)": 1.635761
    },
    {
      "acc": 0.64536476,
      "epoch": 1.6523338406900052,
      "grad_norm": 6.59375,
      "learning_rate": 8.036456139156168e-07,
      "loss": 1.66580391,
      "memory(GiB)": 128.51,
      "step": 65135,
      "train_speed(iter/s)": 1.635774
    },
    {
      "acc": 0.66046152,
      "epoch": 1.6524606798579402,
      "grad_norm": 5.875,
      "learning_rate": 8.03075555332391e-07,
      "loss": 1.70430145,
      "memory(GiB)": 128.51,
      "step": 65140,
      "train_speed(iter/s)": 1.635787
    },
    {
      "acc": 0.67178168,
      "epoch": 1.6525875190258752,
      "grad_norm": 9.625,
      "learning_rate": 8.025056813486387e-07,
      "loss": 1.59192867,
      "memory(GiB)": 128.51,
      "step": 65145,
      "train_speed(iter/s)": 1.635799
    },
    {
      "acc": 0.64705992,
      "epoch": 1.6527143581938102,
      "grad_norm": 6.0,
      "learning_rate": 8.019359919894299e-07,
      "loss": 1.64672737,
      "memory(GiB)": 128.51,
      "step": 65150,
      "train_speed(iter/s)": 1.635812
    },
    {
      "acc": 0.65914907,
      "epoch": 1.6528411973617452,
      "grad_norm": 7.375,
      "learning_rate": 8.013664872798172e-07,
      "loss": 1.56427727,
      "memory(GiB)": 128.51,
      "step": 65155,
      "train_speed(iter/s)": 1.635824
    },
    {
      "acc": 0.64570169,
      "epoch": 1.6529680365296804,
      "grad_norm": 6.3125,
      "learning_rate": 8.007971672448511e-07,
      "loss": 1.61004753,
      "memory(GiB)": 128.51,
      "step": 65160,
      "train_speed(iter/s)": 1.635837
    },
    {
      "acc": 0.65401773,
      "epoch": 1.6530948756976154,
      "grad_norm": 6.34375,
      "learning_rate": 8.002280319095751e-07,
      "loss": 1.63278847,
      "memory(GiB)": 128.51,
      "step": 65165,
      "train_speed(iter/s)": 1.63585
    },
    {
      "acc": 0.65552754,
      "epoch": 1.6532217148655506,
      "grad_norm": 5.3125,
      "learning_rate": 7.996590812990219e-07,
      "loss": 1.58666821,
      "memory(GiB)": 128.51,
      "step": 65170,
      "train_speed(iter/s)": 1.635863
    },
    {
      "acc": 0.67017293,
      "epoch": 1.6533485540334856,
      "grad_norm": 4.6875,
      "learning_rate": 7.990903154382145e-07,
      "loss": 1.52340727,
      "memory(GiB)": 128.51,
      "step": 65175,
      "train_speed(iter/s)": 1.635875
    },
    {
      "acc": 0.65655279,
      "epoch": 1.6534753932014206,
      "grad_norm": 5.90625,
      "learning_rate": 7.985217343521695e-07,
      "loss": 1.56786976,
      "memory(GiB)": 128.51,
      "step": 65180,
      "train_speed(iter/s)": 1.635888
    },
    {
      "acc": 0.65701056,
      "epoch": 1.6536022323693556,
      "grad_norm": 4.96875,
      "learning_rate": 7.979533380658977e-07,
      "loss": 1.61558609,
      "memory(GiB)": 128.51,
      "step": 65185,
      "train_speed(iter/s)": 1.635901
    },
    {
      "acc": 0.67105217,
      "epoch": 1.6537290715372908,
      "grad_norm": 7.1875,
      "learning_rate": 7.973851266044003e-07,
      "loss": 1.5201436,
      "memory(GiB)": 128.51,
      "step": 65190,
      "train_speed(iter/s)": 1.635916
    },
    {
      "acc": 0.6506609,
      "epoch": 1.6538559107052258,
      "grad_norm": 9.75,
      "learning_rate": 7.968170999926661e-07,
      "loss": 1.69856205,
      "memory(GiB)": 128.51,
      "step": 65195,
      "train_speed(iter/s)": 1.63593
    },
    {
      "acc": 0.66018848,
      "epoch": 1.653982749873161,
      "grad_norm": 6.03125,
      "learning_rate": 7.962492582556825e-07,
      "loss": 1.58916445,
      "memory(GiB)": 128.51,
      "step": 65200,
      "train_speed(iter/s)": 1.635943
    },
    {
      "acc": 0.64859557,
      "epoch": 1.654109589041096,
      "grad_norm": 5.78125,
      "learning_rate": 7.956816014184254e-07,
      "loss": 1.62006416,
      "memory(GiB)": 128.51,
      "step": 65205,
      "train_speed(iter/s)": 1.635956
    },
    {
      "acc": 0.66832409,
      "epoch": 1.654236428209031,
      "grad_norm": 6.875,
      "learning_rate": 7.951141295058618e-07,
      "loss": 1.54471521,
      "memory(GiB)": 128.51,
      "step": 65210,
      "train_speed(iter/s)": 1.63597
    },
    {
      "acc": 0.64623723,
      "epoch": 1.654363267376966,
      "grad_norm": 6.125,
      "learning_rate": 7.945468425429525e-07,
      "loss": 1.64616013,
      "memory(GiB)": 128.51,
      "step": 65215,
      "train_speed(iter/s)": 1.635983
    },
    {
      "acc": 0.63586564,
      "epoch": 1.654490106544901,
      "grad_norm": 5.78125,
      "learning_rate": 7.939797405546496e-07,
      "loss": 1.67903404,
      "memory(GiB)": 128.51,
      "step": 65220,
      "train_speed(iter/s)": 1.635997
    },
    {
      "acc": 0.66955652,
      "epoch": 1.6546169457128361,
      "grad_norm": 7.65625,
      "learning_rate": 7.934128235658955e-07,
      "loss": 1.58389778,
      "memory(GiB)": 128.51,
      "step": 65225,
      "train_speed(iter/s)": 1.63601
    },
    {
      "acc": 0.6645081,
      "epoch": 1.6547437848807713,
      "grad_norm": 4.9375,
      "learning_rate": 7.928460916016272e-07,
      "loss": 1.56747551,
      "memory(GiB)": 128.51,
      "step": 65230,
      "train_speed(iter/s)": 1.636022
    },
    {
      "acc": 0.64479227,
      "epoch": 1.6548706240487063,
      "grad_norm": 5.59375,
      "learning_rate": 7.92279544686771e-07,
      "loss": 1.67095509,
      "memory(GiB)": 128.51,
      "step": 65235,
      "train_speed(iter/s)": 1.636035
    },
    {
      "acc": 0.64896307,
      "epoch": 1.6549974632166413,
      "grad_norm": 5.6875,
      "learning_rate": 7.917131828462465e-07,
      "loss": 1.58562851,
      "memory(GiB)": 128.51,
      "step": 65240,
      "train_speed(iter/s)": 1.636047
    },
    {
      "acc": 0.65538549,
      "epoch": 1.6551243023845763,
      "grad_norm": 5.96875,
      "learning_rate": 7.911470061049647e-07,
      "loss": 1.59046879,
      "memory(GiB)": 128.51,
      "step": 65245,
      "train_speed(iter/s)": 1.63606
    },
    {
      "acc": 0.64248624,
      "epoch": 1.6552511415525113,
      "grad_norm": 6.25,
      "learning_rate": 7.90581014487829e-07,
      "loss": 1.62341557,
      "memory(GiB)": 128.51,
      "step": 65250,
      "train_speed(iter/s)": 1.636073
    },
    {
      "acc": 0.65540371,
      "epoch": 1.6553779807204465,
      "grad_norm": 7.09375,
      "learning_rate": 7.900152080197337e-07,
      "loss": 1.68097935,
      "memory(GiB)": 128.51,
      "step": 65255,
      "train_speed(iter/s)": 1.636086
    },
    {
      "acc": 0.6656146,
      "epoch": 1.6555048198883815,
      "grad_norm": 5.8125,
      "learning_rate": 7.894495867255664e-07,
      "loss": 1.60883026,
      "memory(GiB)": 128.51,
      "step": 65260,
      "train_speed(iter/s)": 1.6361
    },
    {
      "acc": 0.65339518,
      "epoch": 1.6556316590563167,
      "grad_norm": 6.28125,
      "learning_rate": 7.888841506302048e-07,
      "loss": 1.6685379,
      "memory(GiB)": 128.51,
      "step": 65265,
      "train_speed(iter/s)": 1.636113
    },
    {
      "acc": 0.67134261,
      "epoch": 1.6557584982242517,
      "grad_norm": 5.3125,
      "learning_rate": 7.883188997585189e-07,
      "loss": 1.4518425,
      "memory(GiB)": 128.51,
      "step": 65270,
      "train_speed(iter/s)": 1.636126
    },
    {
      "acc": 0.64956789,
      "epoch": 1.6558853373921867,
      "grad_norm": 5.34375,
      "learning_rate": 7.877538341353724e-07,
      "loss": 1.66238289,
      "memory(GiB)": 128.51,
      "step": 65275,
      "train_speed(iter/s)": 1.63614
    },
    {
      "acc": 0.67035651,
      "epoch": 1.6560121765601217,
      "grad_norm": 5.09375,
      "learning_rate": 7.871889537856181e-07,
      "loss": 1.58492785,
      "memory(GiB)": 128.51,
      "step": 65280,
      "train_speed(iter/s)": 1.636152
    },
    {
      "acc": 0.64411297,
      "epoch": 1.6561390157280567,
      "grad_norm": 6.875,
      "learning_rate": 7.866242587341017e-07,
      "loss": 1.61599236,
      "memory(GiB)": 128.51,
      "step": 65285,
      "train_speed(iter/s)": 1.636164
    },
    {
      "acc": 0.65285897,
      "epoch": 1.6562658548959919,
      "grad_norm": 6.40625,
      "learning_rate": 7.860597490056638e-07,
      "loss": 1.59369917,
      "memory(GiB)": 128.51,
      "step": 65290,
      "train_speed(iter/s)": 1.636178
    },
    {
      "acc": 0.66462326,
      "epoch": 1.656392694063927,
      "grad_norm": 5.15625,
      "learning_rate": 7.854954246251306e-07,
      "loss": 1.5429512,
      "memory(GiB)": 128.51,
      "step": 65295,
      "train_speed(iter/s)": 1.636191
    },
    {
      "acc": 0.66352501,
      "epoch": 1.656519533231862,
      "grad_norm": 6.0625,
      "learning_rate": 7.849312856173242e-07,
      "loss": 1.57256432,
      "memory(GiB)": 128.51,
      "step": 65300,
      "train_speed(iter/s)": 1.636205
    },
    {
      "acc": 0.65696092,
      "epoch": 1.656646372399797,
      "grad_norm": 5.21875,
      "learning_rate": 7.8436733200706e-07,
      "loss": 1.60008869,
      "memory(GiB)": 128.51,
      "step": 65305,
      "train_speed(iter/s)": 1.636218
    },
    {
      "acc": 0.64650974,
      "epoch": 1.656773211567732,
      "grad_norm": 5.90625,
      "learning_rate": 7.838035638191432e-07,
      "loss": 1.64540615,
      "memory(GiB)": 128.51,
      "step": 65310,
      "train_speed(iter/s)": 1.63623
    },
    {
      "acc": 0.65073004,
      "epoch": 1.656900050735667,
      "grad_norm": 6.25,
      "learning_rate": 7.832399810783686e-07,
      "loss": 1.62034454,
      "memory(GiB)": 128.51,
      "step": 65315,
      "train_speed(iter/s)": 1.636244
    },
    {
      "acc": 0.64747396,
      "epoch": 1.6570268899036023,
      "grad_norm": 4.90625,
      "learning_rate": 7.826765838095246e-07,
      "loss": 1.58713512,
      "memory(GiB)": 128.51,
      "step": 65320,
      "train_speed(iter/s)": 1.636257
    },
    {
      "acc": 0.67192745,
      "epoch": 1.6571537290715372,
      "grad_norm": 6.21875,
      "learning_rate": 7.821133720373947e-07,
      "loss": 1.60663223,
      "memory(GiB)": 128.51,
      "step": 65325,
      "train_speed(iter/s)": 1.636271
    },
    {
      "acc": 0.66326227,
      "epoch": 1.6572805682394725,
      "grad_norm": 5.875,
      "learning_rate": 7.815503457867512e-07,
      "loss": 1.6469099,
      "memory(GiB)": 128.51,
      "step": 65330,
      "train_speed(iter/s)": 1.636284
    },
    {
      "acc": 0.66206222,
      "epoch": 1.6574074074074074,
      "grad_norm": 6.03125,
      "learning_rate": 7.809875050823556e-07,
      "loss": 1.54736423,
      "memory(GiB)": 128.51,
      "step": 65335,
      "train_speed(iter/s)": 1.636297
    },
    {
      "acc": 0.64818468,
      "epoch": 1.6575342465753424,
      "grad_norm": 5.375,
      "learning_rate": 7.804248499489669e-07,
      "loss": 1.65572701,
      "memory(GiB)": 128.51,
      "step": 65340,
      "train_speed(iter/s)": 1.636311
    },
    {
      "acc": 0.66980877,
      "epoch": 1.6576610857432774,
      "grad_norm": 5.375,
      "learning_rate": 7.798623804113326e-07,
      "loss": 1.59281693,
      "memory(GiB)": 128.51,
      "step": 65345,
      "train_speed(iter/s)": 1.636322
    },
    {
      "acc": 0.65084968,
      "epoch": 1.6577879249112126,
      "grad_norm": 4.6875,
      "learning_rate": 7.793000964941932e-07,
      "loss": 1.59598341,
      "memory(GiB)": 128.51,
      "step": 65350,
      "train_speed(iter/s)": 1.636335
    },
    {
      "acc": 0.65423284,
      "epoch": 1.6579147640791476,
      "grad_norm": 5.53125,
      "learning_rate": 7.787379982222776e-07,
      "loss": 1.60357094,
      "memory(GiB)": 128.51,
      "step": 65355,
      "train_speed(iter/s)": 1.636347
    },
    {
      "acc": 0.65282068,
      "epoch": 1.6580416032470828,
      "grad_norm": 5.28125,
      "learning_rate": 7.781760856203124e-07,
      "loss": 1.56206036,
      "memory(GiB)": 128.51,
      "step": 65360,
      "train_speed(iter/s)": 1.636359
    },
    {
      "acc": 0.68612952,
      "epoch": 1.6581684424150178,
      "grad_norm": 5.5625,
      "learning_rate": 7.77614358713012e-07,
      "loss": 1.52576618,
      "memory(GiB)": 128.51,
      "step": 65365,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.66813354,
      "epoch": 1.6582952815829528,
      "grad_norm": 6.125,
      "learning_rate": 7.770528175250835e-07,
      "loss": 1.57223053,
      "memory(GiB)": 128.51,
      "step": 65370,
      "train_speed(iter/s)": 1.636385
    },
    {
      "acc": 0.64530592,
      "epoch": 1.6584221207508878,
      "grad_norm": 6.25,
      "learning_rate": 7.764914620812269e-07,
      "loss": 1.65743713,
      "memory(GiB)": 128.51,
      "step": 65375,
      "train_speed(iter/s)": 1.636397
    },
    {
      "acc": 0.64771333,
      "epoch": 1.6585489599188228,
      "grad_norm": 7.125,
      "learning_rate": 7.759302924061318e-07,
      "loss": 1.60482826,
      "memory(GiB)": 128.51,
      "step": 65380,
      "train_speed(iter/s)": 1.63641
    },
    {
      "acc": 0.64264994,
      "epoch": 1.658675799086758,
      "grad_norm": 5.34375,
      "learning_rate": 7.753693085244818e-07,
      "loss": 1.64855995,
      "memory(GiB)": 128.51,
      "step": 65385,
      "train_speed(iter/s)": 1.636423
    },
    {
      "acc": 0.65659723,
      "epoch": 1.6588026382546932,
      "grad_norm": 5.28125,
      "learning_rate": 7.748085104609509e-07,
      "loss": 1.66198273,
      "memory(GiB)": 128.51,
      "step": 65390,
      "train_speed(iter/s)": 1.636436
    },
    {
      "acc": 0.66503048,
      "epoch": 1.6589294774226282,
      "grad_norm": 7.0625,
      "learning_rate": 7.742478982402063e-07,
      "loss": 1.55654812,
      "memory(GiB)": 128.51,
      "step": 65395,
      "train_speed(iter/s)": 1.636449
    },
    {
      "acc": 0.65070424,
      "epoch": 1.6590563165905632,
      "grad_norm": 4.78125,
      "learning_rate": 7.736874718869053e-07,
      "loss": 1.66279659,
      "memory(GiB)": 128.51,
      "step": 65400,
      "train_speed(iter/s)": 1.636462
    },
    {
      "acc": 0.65407863,
      "epoch": 1.6591831557584982,
      "grad_norm": 6.28125,
      "learning_rate": 7.731272314256988e-07,
      "loss": 1.6149765,
      "memory(GiB)": 128.51,
      "step": 65405,
      "train_speed(iter/s)": 1.636476
    },
    {
      "acc": 0.65497122,
      "epoch": 1.6593099949264332,
      "grad_norm": 8.125,
      "learning_rate": 7.725671768812282e-07,
      "loss": 1.58684559,
      "memory(GiB)": 128.51,
      "step": 65410,
      "train_speed(iter/s)": 1.636489
    },
    {
      "acc": 0.64559765,
      "epoch": 1.6594368340943684,
      "grad_norm": 6.0625,
      "learning_rate": 7.720073082781271e-07,
      "loss": 1.61308193,
      "memory(GiB)": 128.51,
      "step": 65415,
      "train_speed(iter/s)": 1.636503
    },
    {
      "acc": 0.66281614,
      "epoch": 1.6595636732623034,
      "grad_norm": 4.6875,
      "learning_rate": 7.714476256410214e-07,
      "loss": 1.59173279,
      "memory(GiB)": 128.51,
      "step": 65420,
      "train_speed(iter/s)": 1.636516
    },
    {
      "acc": 0.6603837,
      "epoch": 1.6596905124302386,
      "grad_norm": 5.3125,
      "learning_rate": 7.708881289945275e-07,
      "loss": 1.58160429,
      "memory(GiB)": 128.51,
      "step": 65425,
      "train_speed(iter/s)": 1.636529
    },
    {
      "acc": 0.66036453,
      "epoch": 1.6598173515981736,
      "grad_norm": 5.15625,
      "learning_rate": 7.703288183632567e-07,
      "loss": 1.58553047,
      "memory(GiB)": 128.51,
      "step": 65430,
      "train_speed(iter/s)": 1.636542
    },
    {
      "acc": 0.64074917,
      "epoch": 1.6599441907661086,
      "grad_norm": 5.84375,
      "learning_rate": 7.697696937718079e-07,
      "loss": 1.70121021,
      "memory(GiB)": 128.51,
      "step": 65435,
      "train_speed(iter/s)": 1.636554
    },
    {
      "acc": 0.65881038,
      "epoch": 1.6600710299340435,
      "grad_norm": 6.34375,
      "learning_rate": 7.692107552447748e-07,
      "loss": 1.58281765,
      "memory(GiB)": 128.51,
      "step": 65440,
      "train_speed(iter/s)": 1.636566
    },
    {
      "acc": 0.67516508,
      "epoch": 1.6601978691019785,
      "grad_norm": 6.0,
      "learning_rate": 7.686520028067406e-07,
      "loss": 1.54906082,
      "memory(GiB)": 128.51,
      "step": 65445,
      "train_speed(iter/s)": 1.63658
    },
    {
      "acc": 0.65797296,
      "epoch": 1.6603247082699137,
      "grad_norm": 7.0,
      "learning_rate": 7.680934364822851e-07,
      "loss": 1.68625107,
      "memory(GiB)": 128.51,
      "step": 65450,
      "train_speed(iter/s)": 1.636595
    },
    {
      "acc": 0.66376686,
      "epoch": 1.660451547437849,
      "grad_norm": 6.125,
      "learning_rate": 7.675350562959733e-07,
      "loss": 1.60438538,
      "memory(GiB)": 128.51,
      "step": 65455,
      "train_speed(iter/s)": 1.636609
    },
    {
      "acc": 0.66321735,
      "epoch": 1.660578386605784,
      "grad_norm": 5.5,
      "learning_rate": 7.66976862272365e-07,
      "loss": 1.55235806,
      "memory(GiB)": 128.51,
      "step": 65460,
      "train_speed(iter/s)": 1.636624
    },
    {
      "acc": 0.6486279,
      "epoch": 1.660705225773719,
      "grad_norm": 5.6875,
      "learning_rate": 7.664188544360146e-07,
      "loss": 1.61041298,
      "memory(GiB)": 128.51,
      "step": 65465,
      "train_speed(iter/s)": 1.636637
    },
    {
      "acc": 0.67162523,
      "epoch": 1.660832064941654,
      "grad_norm": 7.3125,
      "learning_rate": 7.658610328114658e-07,
      "loss": 1.52537556,
      "memory(GiB)": 128.51,
      "step": 65470,
      "train_speed(iter/s)": 1.63665
    },
    {
      "acc": 0.65377564,
      "epoch": 1.660958904109589,
      "grad_norm": 6.03125,
      "learning_rate": 7.653033974232504e-07,
      "loss": 1.65487843,
      "memory(GiB)": 128.51,
      "step": 65475,
      "train_speed(iter/s)": 1.636664
    },
    {
      "acc": 0.65147114,
      "epoch": 1.6610857432775241,
      "grad_norm": 5.28125,
      "learning_rate": 7.647459482958991e-07,
      "loss": 1.6890976,
      "memory(GiB)": 128.51,
      "step": 65480,
      "train_speed(iter/s)": 1.636676
    },
    {
      "acc": 0.65547643,
      "epoch": 1.661212582445459,
      "grad_norm": 5.03125,
      "learning_rate": 7.641886854539304e-07,
      "loss": 1.61011543,
      "memory(GiB)": 128.51,
      "step": 65485,
      "train_speed(iter/s)": 1.63669
    },
    {
      "acc": 0.6762126,
      "epoch": 1.6613394216133943,
      "grad_norm": 4.9375,
      "learning_rate": 7.63631608921856e-07,
      "loss": 1.5608264,
      "memory(GiB)": 128.51,
      "step": 65490,
      "train_speed(iter/s)": 1.636704
    },
    {
      "acc": 0.65644288,
      "epoch": 1.6614662607813293,
      "grad_norm": 5.1875,
      "learning_rate": 7.63074718724175e-07,
      "loss": 1.57908649,
      "memory(GiB)": 128.51,
      "step": 65495,
      "train_speed(iter/s)": 1.636717
    },
    {
      "acc": 0.65521169,
      "epoch": 1.6615930999492643,
      "grad_norm": 5.125,
      "learning_rate": 7.625180148853856e-07,
      "loss": 1.5827076,
      "memory(GiB)": 128.51,
      "step": 65500,
      "train_speed(iter/s)": 1.636731
    },
    {
      "acc": 0.65709448,
      "epoch": 1.6617199391171993,
      "grad_norm": 6.25,
      "learning_rate": 7.619614974299727e-07,
      "loss": 1.59858389,
      "memory(GiB)": 128.51,
      "step": 65505,
      "train_speed(iter/s)": 1.636745
    },
    {
      "acc": 0.67462406,
      "epoch": 1.6618467782851345,
      "grad_norm": 6.25,
      "learning_rate": 7.614051663824152e-07,
      "loss": 1.50673265,
      "memory(GiB)": 128.51,
      "step": 65510,
      "train_speed(iter/s)": 1.636758
    },
    {
      "acc": 0.67008948,
      "epoch": 1.6619736174530695,
      "grad_norm": 5.75,
      "learning_rate": 7.608490217671821e-07,
      "loss": 1.52118893,
      "memory(GiB)": 128.51,
      "step": 65515,
      "train_speed(iter/s)": 1.636771
    },
    {
      "acc": 0.64825025,
      "epoch": 1.6621004566210047,
      "grad_norm": 6.03125,
      "learning_rate": 7.602930636087352e-07,
      "loss": 1.63918686,
      "memory(GiB)": 128.51,
      "step": 65520,
      "train_speed(iter/s)": 1.636785
    },
    {
      "acc": 0.66683917,
      "epoch": 1.6622272957889397,
      "grad_norm": 5.125,
      "learning_rate": 7.597372919315288e-07,
      "loss": 1.55445423,
      "memory(GiB)": 128.51,
      "step": 65525,
      "train_speed(iter/s)": 1.636797
    },
    {
      "acc": 0.65495896,
      "epoch": 1.6623541349568747,
      "grad_norm": 6.03125,
      "learning_rate": 7.591817067600071e-07,
      "loss": 1.63891392,
      "memory(GiB)": 128.51,
      "step": 65530,
      "train_speed(iter/s)": 1.636811
    },
    {
      "acc": 0.63826151,
      "epoch": 1.6624809741248097,
      "grad_norm": 6.03125,
      "learning_rate": 7.586263081186085e-07,
      "loss": 1.69268265,
      "memory(GiB)": 128.51,
      "step": 65535,
      "train_speed(iter/s)": 1.636824
    },
    {
      "acc": 0.6428556,
      "epoch": 1.6626078132927447,
      "grad_norm": 5.40625,
      "learning_rate": 7.580710960317605e-07,
      "loss": 1.65391922,
      "memory(GiB)": 128.51,
      "step": 65540,
      "train_speed(iter/s)": 1.636838
    },
    {
      "acc": 0.65940886,
      "epoch": 1.6627346524606799,
      "grad_norm": 6.25,
      "learning_rate": 7.575160705238854e-07,
      "loss": 1.59410992,
      "memory(GiB)": 128.51,
      "step": 65545,
      "train_speed(iter/s)": 1.636852
    },
    {
      "acc": 0.65870733,
      "epoch": 1.662861491628615,
      "grad_norm": 5.0,
      "learning_rate": 7.569612316193942e-07,
      "loss": 1.60653038,
      "memory(GiB)": 128.51,
      "step": 65550,
      "train_speed(iter/s)": 1.636865
    },
    {
      "acc": 0.65938873,
      "epoch": 1.66298833079655,
      "grad_norm": 5.5625,
      "learning_rate": 7.564065793426923e-07,
      "loss": 1.58472986,
      "memory(GiB)": 128.51,
      "step": 65555,
      "train_speed(iter/s)": 1.636879
    },
    {
      "acc": 0.65987911,
      "epoch": 1.663115169964485,
      "grad_norm": 5.4375,
      "learning_rate": 7.558521137181752e-07,
      "loss": 1.61114902,
      "memory(GiB)": 128.51,
      "step": 65560,
      "train_speed(iter/s)": 1.636893
    },
    {
      "acc": 0.66059217,
      "epoch": 1.66324200913242,
      "grad_norm": 7.625,
      "learning_rate": 7.552978347702295e-07,
      "loss": 1.53199158,
      "memory(GiB)": 128.51,
      "step": 65565,
      "train_speed(iter/s)": 1.636906
    },
    {
      "acc": 0.6578804,
      "epoch": 1.663368848300355,
      "grad_norm": 6.375,
      "learning_rate": 7.547437425232384e-07,
      "loss": 1.56972036,
      "memory(GiB)": 128.51,
      "step": 65570,
      "train_speed(iter/s)": 1.636919
    },
    {
      "acc": 0.65196123,
      "epoch": 1.6634956874682902,
      "grad_norm": 5.6875,
      "learning_rate": 7.541898370015704e-07,
      "loss": 1.67188797,
      "memory(GiB)": 128.51,
      "step": 65575,
      "train_speed(iter/s)": 1.636933
    },
    {
      "acc": 0.6683013,
      "epoch": 1.6636225266362252,
      "grad_norm": 5.84375,
      "learning_rate": 7.536361182295893e-07,
      "loss": 1.5742116,
      "memory(GiB)": 128.51,
      "step": 65580,
      "train_speed(iter/s)": 1.636946
    },
    {
      "acc": 0.65599442,
      "epoch": 1.6637493658041604,
      "grad_norm": 5.03125,
      "learning_rate": 7.530825862316493e-07,
      "loss": 1.6119175,
      "memory(GiB)": 128.51,
      "step": 65585,
      "train_speed(iter/s)": 1.636959
    },
    {
      "acc": 0.66936278,
      "epoch": 1.6638762049720954,
      "grad_norm": 5.1875,
      "learning_rate": 7.525292410321011e-07,
      "loss": 1.50604992,
      "memory(GiB)": 128.51,
      "step": 65590,
      "train_speed(iter/s)": 1.636971
    },
    {
      "acc": 0.6632875,
      "epoch": 1.6640030441400304,
      "grad_norm": 5.46875,
      "learning_rate": 7.519760826552786e-07,
      "loss": 1.60509377,
      "memory(GiB)": 128.51,
      "step": 65595,
      "train_speed(iter/s)": 1.636984
    },
    {
      "acc": 0.65209823,
      "epoch": 1.6641298833079654,
      "grad_norm": 5.375,
      "learning_rate": 7.514231111255132e-07,
      "loss": 1.57483101,
      "memory(GiB)": 128.51,
      "step": 65600,
      "train_speed(iter/s)": 1.636996
    },
    {
      "acc": 0.66512256,
      "epoch": 1.6642567224759004,
      "grad_norm": 5.15625,
      "learning_rate": 7.508703264671291e-07,
      "loss": 1.53045959,
      "memory(GiB)": 128.51,
      "step": 65605,
      "train_speed(iter/s)": 1.637008
    },
    {
      "acc": 0.64667196,
      "epoch": 1.6643835616438356,
      "grad_norm": 6.09375,
      "learning_rate": 7.5031772870444e-07,
      "loss": 1.62335701,
      "memory(GiB)": 128.51,
      "step": 65610,
      "train_speed(iter/s)": 1.637021
    },
    {
      "acc": 0.65590401,
      "epoch": 1.6645104008117708,
      "grad_norm": 5.5,
      "learning_rate": 7.497653178617498e-07,
      "loss": 1.62596111,
      "memory(GiB)": 128.51,
      "step": 65615,
      "train_speed(iter/s)": 1.637034
    },
    {
      "acc": 0.64788008,
      "epoch": 1.6646372399797058,
      "grad_norm": 5.0625,
      "learning_rate": 7.492130939633557e-07,
      "loss": 1.6896162,
      "memory(GiB)": 128.51,
      "step": 65620,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.63460608,
      "epoch": 1.6647640791476408,
      "grad_norm": 5.3125,
      "learning_rate": 7.48661057033549e-07,
      "loss": 1.6710701,
      "memory(GiB)": 128.51,
      "step": 65625,
      "train_speed(iter/s)": 1.63706
    },
    {
      "acc": 0.65772924,
      "epoch": 1.6648909183155758,
      "grad_norm": 4.8125,
      "learning_rate": 7.481092070966111e-07,
      "loss": 1.64268894,
      "memory(GiB)": 128.51,
      "step": 65630,
      "train_speed(iter/s)": 1.637072
    },
    {
      "acc": 0.65640893,
      "epoch": 1.6650177574835108,
      "grad_norm": 6.5625,
      "learning_rate": 7.475575441768112e-07,
      "loss": 1.60860653,
      "memory(GiB)": 128.51,
      "step": 65635,
      "train_speed(iter/s)": 1.637085
    },
    {
      "acc": 0.65400248,
      "epoch": 1.665144596651446,
      "grad_norm": 5.75,
      "learning_rate": 7.470060682984176e-07,
      "loss": 1.56483631,
      "memory(GiB)": 128.51,
      "step": 65640,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.65104542,
      "epoch": 1.665271435819381,
      "grad_norm": 5.4375,
      "learning_rate": 7.464547794856858e-07,
      "loss": 1.58781214,
      "memory(GiB)": 128.51,
      "step": 65645,
      "train_speed(iter/s)": 1.637109
    },
    {
      "acc": 0.65274248,
      "epoch": 1.6653982749873162,
      "grad_norm": 6.03125,
      "learning_rate": 7.459036777628631e-07,
      "loss": 1.55972958,
      "memory(GiB)": 128.51,
      "step": 65650,
      "train_speed(iter/s)": 1.637122
    },
    {
      "acc": 0.6407217,
      "epoch": 1.6655251141552512,
      "grad_norm": 6.375,
      "learning_rate": 7.453527631541896e-07,
      "loss": 1.61590977,
      "memory(GiB)": 128.51,
      "step": 65655,
      "train_speed(iter/s)": 1.637135
    },
    {
      "acc": 0.66780691,
      "epoch": 1.6656519533231862,
      "grad_norm": 6.28125,
      "learning_rate": 7.448020356838975e-07,
      "loss": 1.58678398,
      "memory(GiB)": 128.51,
      "step": 65660,
      "train_speed(iter/s)": 1.637148
    },
    {
      "acc": 0.65969954,
      "epoch": 1.6657787924911212,
      "grad_norm": 5.46875,
      "learning_rate": 7.442514953762098e-07,
      "loss": 1.58708534,
      "memory(GiB)": 128.51,
      "step": 65665,
      "train_speed(iter/s)": 1.63716
    },
    {
      "acc": 0.64949865,
      "epoch": 1.6659056316590564,
      "grad_norm": 5.125,
      "learning_rate": 7.43701142255342e-07,
      "loss": 1.61301956,
      "memory(GiB)": 128.51,
      "step": 65670,
      "train_speed(iter/s)": 1.637173
    },
    {
      "acc": 0.6605875,
      "epoch": 1.6660324708269914,
      "grad_norm": 5.75,
      "learning_rate": 7.431509763455008e-07,
      "loss": 1.53994608,
      "memory(GiB)": 128.51,
      "step": 65675,
      "train_speed(iter/s)": 1.637186
    },
    {
      "acc": 0.65858688,
      "epoch": 1.6661593099949266,
      "grad_norm": 6.46875,
      "learning_rate": 7.426009976708854e-07,
      "loss": 1.56550102,
      "memory(GiB)": 128.51,
      "step": 65680,
      "train_speed(iter/s)": 1.637198
    },
    {
      "acc": 0.64349213,
      "epoch": 1.6662861491628616,
      "grad_norm": 5.09375,
      "learning_rate": 7.420512062556856e-07,
      "loss": 1.65120335,
      "memory(GiB)": 128.51,
      "step": 65685,
      "train_speed(iter/s)": 1.637211
    },
    {
      "acc": 0.66316233,
      "epoch": 1.6664129883307965,
      "grad_norm": 6.34375,
      "learning_rate": 7.415016021240845e-07,
      "loss": 1.55864029,
      "memory(GiB)": 128.51,
      "step": 65690,
      "train_speed(iter/s)": 1.637223
    },
    {
      "acc": 0.67226272,
      "epoch": 1.6665398274987315,
      "grad_norm": 6.03125,
      "learning_rate": 7.40952185300256e-07,
      "loss": 1.56930151,
      "memory(GiB)": 128.51,
      "step": 65695,
      "train_speed(iter/s)": 1.637236
    },
    {
      "acc": 0.67184854,
      "epoch": 1.6666666666666665,
      "grad_norm": 5.0625,
      "learning_rate": 7.404029558083653e-07,
      "loss": 1.53423843,
      "memory(GiB)": 128.51,
      "step": 65700,
      "train_speed(iter/s)": 1.637249
    },
    {
      "acc": 0.6415741,
      "epoch": 1.6667935058346017,
      "grad_norm": 6.8125,
      "learning_rate": 7.398539136725702e-07,
      "loss": 1.65635262,
      "memory(GiB)": 128.51,
      "step": 65705,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.64650521,
      "epoch": 1.666920345002537,
      "grad_norm": 5.5625,
      "learning_rate": 7.393050589170203e-07,
      "loss": 1.61602917,
      "memory(GiB)": 128.51,
      "step": 65710,
      "train_speed(iter/s)": 1.637275
    },
    {
      "acc": 0.65448236,
      "epoch": 1.667047184170472,
      "grad_norm": 4.9375,
      "learning_rate": 7.387563915658569e-07,
      "loss": 1.57407494,
      "memory(GiB)": 128.51,
      "step": 65715,
      "train_speed(iter/s)": 1.637287
    },
    {
      "acc": 0.64957504,
      "epoch": 1.667174023338407,
      "grad_norm": 5.28125,
      "learning_rate": 7.382079116432117e-07,
      "loss": 1.61673355,
      "memory(GiB)": 128.51,
      "step": 65720,
      "train_speed(iter/s)": 1.637299
    },
    {
      "acc": 0.65638967,
      "epoch": 1.667300862506342,
      "grad_norm": 5.6875,
      "learning_rate": 7.376596191732093e-07,
      "loss": 1.58749752,
      "memory(GiB)": 128.51,
      "step": 65725,
      "train_speed(iter/s)": 1.637313
    },
    {
      "acc": 0.63733754,
      "epoch": 1.667427701674277,
      "grad_norm": 5.96875,
      "learning_rate": 7.371115141799695e-07,
      "loss": 1.65910378,
      "memory(GiB)": 128.51,
      "step": 65730,
      "train_speed(iter/s)": 1.637327
    },
    {
      "acc": 0.66136904,
      "epoch": 1.667554540842212,
      "grad_norm": 5.8125,
      "learning_rate": 7.365635966875961e-07,
      "loss": 1.60378284,
      "memory(GiB)": 128.51,
      "step": 65735,
      "train_speed(iter/s)": 1.63734
    },
    {
      "acc": 0.65051584,
      "epoch": 1.667681380010147,
      "grad_norm": 5.375,
      "learning_rate": 7.3601586672019e-07,
      "loss": 1.60557194,
      "memory(GiB)": 128.51,
      "step": 65740,
      "train_speed(iter/s)": 1.637354
    },
    {
      "acc": 0.66332998,
      "epoch": 1.6678082191780823,
      "grad_norm": 7.28125,
      "learning_rate": 7.35468324301844e-07,
      "loss": 1.58791981,
      "memory(GiB)": 128.51,
      "step": 65745,
      "train_speed(iter/s)": 1.637366
    },
    {
      "acc": 0.66530094,
      "epoch": 1.6679350583460173,
      "grad_norm": 7.5625,
      "learning_rate": 7.349209694566422e-07,
      "loss": 1.56388416,
      "memory(GiB)": 128.51,
      "step": 65750,
      "train_speed(iter/s)": 1.63738
    },
    {
      "acc": 0.66008701,
      "epoch": 1.6680618975139523,
      "grad_norm": 5.71875,
      "learning_rate": 7.343738022086572e-07,
      "loss": 1.58604755,
      "memory(GiB)": 128.51,
      "step": 65755,
      "train_speed(iter/s)": 1.637393
    },
    {
      "acc": 0.65272512,
      "epoch": 1.6681887366818873,
      "grad_norm": 5.59375,
      "learning_rate": 7.338268225819562e-07,
      "loss": 1.64771614,
      "memory(GiB)": 128.51,
      "step": 65760,
      "train_speed(iter/s)": 1.637407
    },
    {
      "acc": 0.63742986,
      "epoch": 1.6683155758498223,
      "grad_norm": 5.65625,
      "learning_rate": 7.332800306005994e-07,
      "loss": 1.6484024,
      "memory(GiB)": 128.51,
      "step": 65765,
      "train_speed(iter/s)": 1.63742
    },
    {
      "acc": 0.65031991,
      "epoch": 1.6684424150177575,
      "grad_norm": 5.375,
      "learning_rate": 7.327334262886376e-07,
      "loss": 1.68305264,
      "memory(GiB)": 128.51,
      "step": 65770,
      "train_speed(iter/s)": 1.637434
    },
    {
      "acc": 0.66134796,
      "epoch": 1.6685692541856927,
      "grad_norm": 5.59375,
      "learning_rate": 7.321870096701095e-07,
      "loss": 1.61196709,
      "memory(GiB)": 128.51,
      "step": 65775,
      "train_speed(iter/s)": 1.637447
    },
    {
      "acc": 0.66320267,
      "epoch": 1.6686960933536277,
      "grad_norm": 5.625,
      "learning_rate": 7.316407807690523e-07,
      "loss": 1.51767149,
      "memory(GiB)": 128.51,
      "step": 65780,
      "train_speed(iter/s)": 1.63746
    },
    {
      "acc": 0.6484231,
      "epoch": 1.6688229325215627,
      "grad_norm": 7.0625,
      "learning_rate": 7.310947396094903e-07,
      "loss": 1.65481453,
      "memory(GiB)": 128.51,
      "step": 65785,
      "train_speed(iter/s)": 1.637473
    },
    {
      "acc": 0.65878487,
      "epoch": 1.6689497716894977,
      "grad_norm": 5.8125,
      "learning_rate": 7.30548886215442e-07,
      "loss": 1.62875786,
      "memory(GiB)": 128.51,
      "step": 65790,
      "train_speed(iter/s)": 1.637486
    },
    {
      "acc": 0.66949759,
      "epoch": 1.6690766108574326,
      "grad_norm": 5.53125,
      "learning_rate": 7.300032206109131e-07,
      "loss": 1.53504972,
      "memory(GiB)": 128.51,
      "step": 65795,
      "train_speed(iter/s)": 1.6375
    },
    {
      "acc": 0.66699028,
      "epoch": 1.6692034500253679,
      "grad_norm": 5.34375,
      "learning_rate": 7.294577428199084e-07,
      "loss": 1.54227524,
      "memory(GiB)": 128.51,
      "step": 65800,
      "train_speed(iter/s)": 1.637514
    },
    {
      "acc": 0.64892025,
      "epoch": 1.6693302891933028,
      "grad_norm": 5.15625,
      "learning_rate": 7.289124528664182e-07,
      "loss": 1.70395527,
      "memory(GiB)": 128.51,
      "step": 65805,
      "train_speed(iter/s)": 1.637528
    },
    {
      "acc": 0.6476965,
      "epoch": 1.669457128361238,
      "grad_norm": 5.125,
      "learning_rate": 7.283673507744276e-07,
      "loss": 1.64203892,
      "memory(GiB)": 128.51,
      "step": 65810,
      "train_speed(iter/s)": 1.637541
    },
    {
      "acc": 0.64554396,
      "epoch": 1.669583967529173,
      "grad_norm": 7.0625,
      "learning_rate": 7.278224365679121e-07,
      "loss": 1.59112759,
      "memory(GiB)": 128.51,
      "step": 65815,
      "train_speed(iter/s)": 1.637553
    },
    {
      "acc": 0.6584939,
      "epoch": 1.669710806697108,
      "grad_norm": 7.09375,
      "learning_rate": 7.2727771027084e-07,
      "loss": 1.61457024,
      "memory(GiB)": 128.51,
      "step": 65820,
      "train_speed(iter/s)": 1.637567
    },
    {
      "acc": 0.65968752,
      "epoch": 1.669837645865043,
      "grad_norm": 5.59375,
      "learning_rate": 7.267331719071707e-07,
      "loss": 1.6631279,
      "memory(GiB)": 128.51,
      "step": 65825,
      "train_speed(iter/s)": 1.637581
    },
    {
      "acc": 0.66390309,
      "epoch": 1.6699644850329782,
      "grad_norm": 5.6875,
      "learning_rate": 7.261888215008551e-07,
      "loss": 1.56354103,
      "memory(GiB)": 128.51,
      "step": 65830,
      "train_speed(iter/s)": 1.637595
    },
    {
      "acc": 0.64683018,
      "epoch": 1.6700913242009132,
      "grad_norm": 5.25,
      "learning_rate": 7.256446590758359e-07,
      "loss": 1.67115707,
      "memory(GiB)": 128.51,
      "step": 65835,
      "train_speed(iter/s)": 1.637608
    },
    {
      "acc": 0.67179337,
      "epoch": 1.6702181633688484,
      "grad_norm": 7.625,
      "learning_rate": 7.251006846560487e-07,
      "loss": 1.51348686,
      "memory(GiB)": 128.51,
      "step": 65840,
      "train_speed(iter/s)": 1.637621
    },
    {
      "acc": 0.64702387,
      "epoch": 1.6703450025367834,
      "grad_norm": 5.875,
      "learning_rate": 7.245568982654194e-07,
      "loss": 1.64454041,
      "memory(GiB)": 128.51,
      "step": 65845,
      "train_speed(iter/s)": 1.637634
    },
    {
      "acc": 0.6678637,
      "epoch": 1.6704718417047184,
      "grad_norm": 6.15625,
      "learning_rate": 7.24013299927866e-07,
      "loss": 1.63014526,
      "memory(GiB)": 128.51,
      "step": 65850,
      "train_speed(iter/s)": 1.637647
    },
    {
      "acc": 0.6566741,
      "epoch": 1.6705986808726534,
      "grad_norm": 6.71875,
      "learning_rate": 7.234698896672987e-07,
      "loss": 1.64499893,
      "memory(GiB)": 128.51,
      "step": 65855,
      "train_speed(iter/s)": 1.63766
    },
    {
      "acc": 0.65980058,
      "epoch": 1.6707255200405884,
      "grad_norm": 5.34375,
      "learning_rate": 7.229266675076191e-07,
      "loss": 1.59843178,
      "memory(GiB)": 128.51,
      "step": 65860,
      "train_speed(iter/s)": 1.637672
    },
    {
      "acc": 0.66773119,
      "epoch": 1.6708523592085236,
      "grad_norm": 6.03125,
      "learning_rate": 7.223836334727191e-07,
      "loss": 1.51826258,
      "memory(GiB)": 128.51,
      "step": 65865,
      "train_speed(iter/s)": 1.637685
    },
    {
      "acc": 0.65239058,
      "epoch": 1.6709791983764588,
      "grad_norm": 7.8125,
      "learning_rate": 7.218407875864875e-07,
      "loss": 1.591014,
      "memory(GiB)": 128.51,
      "step": 65870,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.65236731,
      "epoch": 1.6711060375443938,
      "grad_norm": 6.375,
      "learning_rate": 7.212981298727972e-07,
      "loss": 1.6539753,
      "memory(GiB)": 128.51,
      "step": 65875,
      "train_speed(iter/s)": 1.63771
    },
    {
      "acc": 0.64987974,
      "epoch": 1.6712328767123288,
      "grad_norm": 5.8125,
      "learning_rate": 7.207556603555188e-07,
      "loss": 1.66034451,
      "memory(GiB)": 128.51,
      "step": 65880,
      "train_speed(iter/s)": 1.637724
    },
    {
      "acc": 0.63892794,
      "epoch": 1.6713597158802638,
      "grad_norm": 5.84375,
      "learning_rate": 7.202133790585103e-07,
      "loss": 1.62107201,
      "memory(GiB)": 128.51,
      "step": 65885,
      "train_speed(iter/s)": 1.637737
    },
    {
      "acc": 0.65837064,
      "epoch": 1.6714865550481988,
      "grad_norm": 7.09375,
      "learning_rate": 7.196712860056277e-07,
      "loss": 1.58579884,
      "memory(GiB)": 128.51,
      "step": 65890,
      "train_speed(iter/s)": 1.637749
    },
    {
      "acc": 0.65704942,
      "epoch": 1.671613394216134,
      "grad_norm": 4.53125,
      "learning_rate": 7.191293812207111e-07,
      "loss": 1.63663578,
      "memory(GiB)": 128.51,
      "step": 65895,
      "train_speed(iter/s)": 1.637761
    },
    {
      "acc": 0.6690927,
      "epoch": 1.671740233384069,
      "grad_norm": 5.46875,
      "learning_rate": 7.185876647275958e-07,
      "loss": 1.53855295,
      "memory(GiB)": 128.51,
      "step": 65900,
      "train_speed(iter/s)": 1.637773
    },
    {
      "acc": 0.65507488,
      "epoch": 1.6718670725520042,
      "grad_norm": 6.6875,
      "learning_rate": 7.180461365501113e-07,
      "loss": 1.59777317,
      "memory(GiB)": 128.51,
      "step": 65905,
      "train_speed(iter/s)": 1.637785
    },
    {
      "acc": 0.64536839,
      "epoch": 1.6719939117199392,
      "grad_norm": 5.65625,
      "learning_rate": 7.175047967120763e-07,
      "loss": 1.5690485,
      "memory(GiB)": 128.51,
      "step": 65910,
      "train_speed(iter/s)": 1.637798
    },
    {
      "acc": 0.66415539,
      "epoch": 1.6721207508878742,
      "grad_norm": 5.875,
      "learning_rate": 7.16963645237298e-07,
      "loss": 1.56981068,
      "memory(GiB)": 128.51,
      "step": 65915,
      "train_speed(iter/s)": 1.637811
    },
    {
      "acc": 0.67045584,
      "epoch": 1.6722475900558091,
      "grad_norm": 7.0625,
      "learning_rate": 7.164226821495824e-07,
      "loss": 1.49785137,
      "memory(GiB)": 128.51,
      "step": 65920,
      "train_speed(iter/s)": 1.637824
    },
    {
      "acc": 0.65235615,
      "epoch": 1.6723744292237441,
      "grad_norm": 5.46875,
      "learning_rate": 7.158819074727219e-07,
      "loss": 1.63071136,
      "memory(GiB)": 128.51,
      "step": 65925,
      "train_speed(iter/s)": 1.637837
    },
    {
      "acc": 0.65252352,
      "epoch": 1.6725012683916793,
      "grad_norm": 9.0625,
      "learning_rate": 7.153413212305032e-07,
      "loss": 1.59601116,
      "memory(GiB)": 128.51,
      "step": 65930,
      "train_speed(iter/s)": 1.637849
    },
    {
      "acc": 0.64923925,
      "epoch": 1.6726281075596146,
      "grad_norm": 7.5,
      "learning_rate": 7.148009234467007e-07,
      "loss": 1.63145771,
      "memory(GiB)": 128.51,
      "step": 65935,
      "train_speed(iter/s)": 1.637862
    },
    {
      "acc": 0.64023819,
      "epoch": 1.6727549467275495,
      "grad_norm": 5.6875,
      "learning_rate": 7.142607141450869e-07,
      "loss": 1.64850807,
      "memory(GiB)": 128.51,
      "step": 65940,
      "train_speed(iter/s)": 1.637875
    },
    {
      "acc": 0.6492919,
      "epoch": 1.6728817858954845,
      "grad_norm": 6.875,
      "learning_rate": 7.137206933494211e-07,
      "loss": 1.61210518,
      "memory(GiB)": 128.51,
      "step": 65945,
      "train_speed(iter/s)": 1.637888
    },
    {
      "acc": 0.6444777,
      "epoch": 1.6730086250634195,
      "grad_norm": 5.59375,
      "learning_rate": 7.131808610834567e-07,
      "loss": 1.63769684,
      "memory(GiB)": 128.51,
      "step": 65950,
      "train_speed(iter/s)": 1.637899
    },
    {
      "acc": 0.65460792,
      "epoch": 1.6731354642313545,
      "grad_norm": 5.28125,
      "learning_rate": 7.126412173709369e-07,
      "loss": 1.57896786,
      "memory(GiB)": 128.51,
      "step": 65955,
      "train_speed(iter/s)": 1.637912
    },
    {
      "acc": 0.65432692,
      "epoch": 1.6732623033992897,
      "grad_norm": 5.59375,
      "learning_rate": 7.12101762235598e-07,
      "loss": 1.6123373,
      "memory(GiB)": 128.51,
      "step": 65960,
      "train_speed(iter/s)": 1.637924
    },
    {
      "acc": 0.65945477,
      "epoch": 1.6733891425672247,
      "grad_norm": 6.09375,
      "learning_rate": 7.115624957011674e-07,
      "loss": 1.55556021,
      "memory(GiB)": 128.51,
      "step": 65965,
      "train_speed(iter/s)": 1.637937
    },
    {
      "acc": 0.65260043,
      "epoch": 1.67351598173516,
      "grad_norm": 5.96875,
      "learning_rate": 7.11023417791365e-07,
      "loss": 1.55330524,
      "memory(GiB)": 128.51,
      "step": 65970,
      "train_speed(iter/s)": 1.637949
    },
    {
      "acc": 0.65940013,
      "epoch": 1.673642820903095,
      "grad_norm": 4.75,
      "learning_rate": 7.104845285299017e-07,
      "loss": 1.6362566,
      "memory(GiB)": 128.51,
      "step": 65975,
      "train_speed(iter/s)": 1.637961
    },
    {
      "acc": 0.65371928,
      "epoch": 1.67376966007103,
      "grad_norm": 5.625,
      "learning_rate": 7.099458279404797e-07,
      "loss": 1.6317297,
      "memory(GiB)": 128.51,
      "step": 65980,
      "train_speed(iter/s)": 1.637974
    },
    {
      "acc": 0.66153316,
      "epoch": 1.6738964992389649,
      "grad_norm": 5.375,
      "learning_rate": 7.094073160467945e-07,
      "loss": 1.57646141,
      "memory(GiB)": 128.51,
      "step": 65985,
      "train_speed(iter/s)": 1.637985
    },
    {
      "acc": 0.6506083,
      "epoch": 1.6740233384069,
      "grad_norm": 5.5,
      "learning_rate": 7.088689928725311e-07,
      "loss": 1.61677818,
      "memory(GiB)": 128.51,
      "step": 65990,
      "train_speed(iter/s)": 1.637997
    },
    {
      "acc": 0.65889263,
      "epoch": 1.674150177574835,
      "grad_norm": 5.5,
      "learning_rate": 7.083308584413684e-07,
      "loss": 1.63257027,
      "memory(GiB)": 128.51,
      "step": 65995,
      "train_speed(iter/s)": 1.63801
    },
    {
      "acc": 0.65007911,
      "epoch": 1.6742770167427703,
      "grad_norm": 5.59375,
      "learning_rate": 7.077929127769756e-07,
      "loss": 1.58507204,
      "memory(GiB)": 128.51,
      "step": 66000,
      "train_speed(iter/s)": 1.638023
    },
    {
      "epoch": 1.6742770167427703,
      "eval_acc": 0.6463461952106787,
      "eval_loss": 1.5732182264328003,
      "eval_runtime": 58.7129,
      "eval_samples_per_second": 108.494,
      "eval_steps_per_second": 27.132,
      "step": 66000
    },
    {
      "acc": 0.65895338,
      "epoch": 1.6744038559107053,
      "grad_norm": 6.46875,
      "learning_rate": 7.072551559030122e-07,
      "loss": 1.56474609,
      "memory(GiB)": 128.51,
      "step": 66005,
      "train_speed(iter/s)": 1.635478
    },
    {
      "acc": 0.65693903,
      "epoch": 1.6745306950786403,
      "grad_norm": 5.65625,
      "learning_rate": 7.067175878431353e-07,
      "loss": 1.54142103,
      "memory(GiB)": 128.51,
      "step": 66010,
      "train_speed(iter/s)": 1.635487
    },
    {
      "acc": 0.65787177,
      "epoch": 1.6746575342465753,
      "grad_norm": 5.6875,
      "learning_rate": 7.061802086209857e-07,
      "loss": 1.61344299,
      "memory(GiB)": 128.51,
      "step": 66015,
      "train_speed(iter/s)": 1.6355
    },
    {
      "acc": 0.64370451,
      "epoch": 1.6747843734145103,
      "grad_norm": 4.75,
      "learning_rate": 7.056430182602008e-07,
      "loss": 1.6362175,
      "memory(GiB)": 128.51,
      "step": 66020,
      "train_speed(iter/s)": 1.635513
    },
    {
      "acc": 0.66851707,
      "epoch": 1.6749112125824455,
      "grad_norm": 5.9375,
      "learning_rate": 7.051060167844081e-07,
      "loss": 1.57702732,
      "memory(GiB)": 128.51,
      "step": 66025,
      "train_speed(iter/s)": 1.635526
    },
    {
      "acc": 0.66495824,
      "epoch": 1.6750380517503807,
      "grad_norm": 6.90625,
      "learning_rate": 7.045692042172309e-07,
      "loss": 1.59467278,
      "memory(GiB)": 128.51,
      "step": 66030,
      "train_speed(iter/s)": 1.635541
    },
    {
      "acc": 0.6661654,
      "epoch": 1.6751648909183157,
      "grad_norm": 6.34375,
      "learning_rate": 7.040325805822756e-07,
      "loss": 1.54114189,
      "memory(GiB)": 128.51,
      "step": 66035,
      "train_speed(iter/s)": 1.635554
    },
    {
      "acc": 0.64201546,
      "epoch": 1.6752917300862507,
      "grad_norm": 4.875,
      "learning_rate": 7.03496145903147e-07,
      "loss": 1.6417347,
      "memory(GiB)": 128.51,
      "step": 66040,
      "train_speed(iter/s)": 1.635568
    },
    {
      "acc": 0.66140008,
      "epoch": 1.6754185692541856,
      "grad_norm": 5.15625,
      "learning_rate": 7.029599002034415e-07,
      "loss": 1.5959465,
      "memory(GiB)": 128.51,
      "step": 66045,
      "train_speed(iter/s)": 1.635581
    },
    {
      "acc": 0.653299,
      "epoch": 1.6755454084221206,
      "grad_norm": 5.8125,
      "learning_rate": 7.02423843506746e-07,
      "loss": 1.59394512,
      "memory(GiB)": 128.51,
      "step": 66050,
      "train_speed(iter/s)": 1.635595
    },
    {
      "acc": 0.65036874,
      "epoch": 1.6756722475900558,
      "grad_norm": 7.25,
      "learning_rate": 7.018879758366354e-07,
      "loss": 1.59594193,
      "memory(GiB)": 128.51,
      "step": 66055,
      "train_speed(iter/s)": 1.635608
    },
    {
      "acc": 0.66116576,
      "epoch": 1.6757990867579908,
      "grad_norm": 6.03125,
      "learning_rate": 7.013522972166803e-07,
      "loss": 1.56445589,
      "memory(GiB)": 128.51,
      "step": 66060,
      "train_speed(iter/s)": 1.635621
    },
    {
      "acc": 0.65793242,
      "epoch": 1.675925925925926,
      "grad_norm": 8.125,
      "learning_rate": 7.008168076704447e-07,
      "loss": 1.57527142,
      "memory(GiB)": 128.51,
      "step": 66065,
      "train_speed(iter/s)": 1.635633
    },
    {
      "acc": 0.65877395,
      "epoch": 1.676052765093861,
      "grad_norm": 6.21875,
      "learning_rate": 7.002815072214814e-07,
      "loss": 1.59430256,
      "memory(GiB)": 128.51,
      "step": 66070,
      "train_speed(iter/s)": 1.635646
    },
    {
      "acc": 0.66328082,
      "epoch": 1.676179604261796,
      "grad_norm": 4.96875,
      "learning_rate": 6.997463958933315e-07,
      "loss": 1.56871204,
      "memory(GiB)": 128.51,
      "step": 66075,
      "train_speed(iter/s)": 1.63566
    },
    {
      "acc": 0.66457644,
      "epoch": 1.676306443429731,
      "grad_norm": 4.84375,
      "learning_rate": 6.992114737095362e-07,
      "loss": 1.58344975,
      "memory(GiB)": 128.51,
      "step": 66080,
      "train_speed(iter/s)": 1.635674
    },
    {
      "acc": 0.65482426,
      "epoch": 1.676433282597666,
      "grad_norm": 5.4375,
      "learning_rate": 6.986767406936212e-07,
      "loss": 1.5526042,
      "memory(GiB)": 128.51,
      "step": 66085,
      "train_speed(iter/s)": 1.635687
    },
    {
      "acc": 0.65471187,
      "epoch": 1.6765601217656012,
      "grad_norm": 5.65625,
      "learning_rate": 6.981421968691077e-07,
      "loss": 1.62837772,
      "memory(GiB)": 128.51,
      "step": 66090,
      "train_speed(iter/s)": 1.635702
    },
    {
      "acc": 0.64523411,
      "epoch": 1.6766869609335364,
      "grad_norm": 6.875,
      "learning_rate": 6.976078422595067e-07,
      "loss": 1.65894051,
      "memory(GiB)": 128.51,
      "step": 66095,
      "train_speed(iter/s)": 1.635715
    },
    {
      "acc": 0.66164365,
      "epoch": 1.6768138001014714,
      "grad_norm": 4.6875,
      "learning_rate": 6.970736768883219e-07,
      "loss": 1.59820232,
      "memory(GiB)": 128.51,
      "step": 66100,
      "train_speed(iter/s)": 1.635729
    },
    {
      "acc": 0.66650925,
      "epoch": 1.6769406392694064,
      "grad_norm": 5.4375,
      "learning_rate": 6.965397007790476e-07,
      "loss": 1.59362192,
      "memory(GiB)": 128.51,
      "step": 66105,
      "train_speed(iter/s)": 1.635742
    },
    {
      "acc": 0.68512206,
      "epoch": 1.6770674784373414,
      "grad_norm": 5.375,
      "learning_rate": 6.960059139551706e-07,
      "loss": 1.49706717,
      "memory(GiB)": 128.51,
      "step": 66110,
      "train_speed(iter/s)": 1.635755
    },
    {
      "acc": 0.65982838,
      "epoch": 1.6771943176052764,
      "grad_norm": 7.21875,
      "learning_rate": 6.9547231644017e-07,
      "loss": 1.62524433,
      "memory(GiB)": 128.51,
      "step": 66115,
      "train_speed(iter/s)": 1.635769
    },
    {
      "acc": 0.64518566,
      "epoch": 1.6773211567732116,
      "grad_norm": 6.59375,
      "learning_rate": 6.949389082575148e-07,
      "loss": 1.63537827,
      "memory(GiB)": 128.51,
      "step": 66120,
      "train_speed(iter/s)": 1.635783
    },
    {
      "acc": 0.65603218,
      "epoch": 1.6774479959411466,
      "grad_norm": 6.84375,
      "learning_rate": 6.944056894306672e-07,
      "loss": 1.57258205,
      "memory(GiB)": 128.51,
      "step": 66125,
      "train_speed(iter/s)": 1.635796
    },
    {
      "acc": 0.6519145,
      "epoch": 1.6775748351090818,
      "grad_norm": 4.78125,
      "learning_rate": 6.938726599830808e-07,
      "loss": 1.70210648,
      "memory(GiB)": 128.51,
      "step": 66130,
      "train_speed(iter/s)": 1.635808
    },
    {
      "acc": 0.64451842,
      "epoch": 1.6777016742770168,
      "grad_norm": 5.8125,
      "learning_rate": 6.933398199382002e-07,
      "loss": 1.6237957,
      "memory(GiB)": 128.51,
      "step": 66135,
      "train_speed(iter/s)": 1.635821
    },
    {
      "acc": 0.66308241,
      "epoch": 1.6778285134449518,
      "grad_norm": 8.1875,
      "learning_rate": 6.928071693194616e-07,
      "loss": 1.62530499,
      "memory(GiB)": 128.51,
      "step": 66140,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.65866814,
      "epoch": 1.6779553526128868,
      "grad_norm": 4.96875,
      "learning_rate": 6.922747081502945e-07,
      "loss": 1.60301971,
      "memory(GiB)": 128.51,
      "step": 66145,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.67551317,
      "epoch": 1.678082191780822,
      "grad_norm": 5.8125,
      "learning_rate": 6.91742436454118e-07,
      "loss": 1.5376049,
      "memory(GiB)": 128.51,
      "step": 66150,
      "train_speed(iter/s)": 1.635859
    },
    {
      "acc": 0.66165857,
      "epoch": 1.678209030948757,
      "grad_norm": 5.5,
      "learning_rate": 6.912103542543446e-07,
      "loss": 1.6255394,
      "memory(GiB)": 128.51,
      "step": 66155,
      "train_speed(iter/s)": 1.635872
    },
    {
      "acc": 0.6597002,
      "epoch": 1.6783358701166922,
      "grad_norm": 5.71875,
      "learning_rate": 6.906784615743772e-07,
      "loss": 1.5963603,
      "memory(GiB)": 128.51,
      "step": 66160,
      "train_speed(iter/s)": 1.635885
    },
    {
      "acc": 0.66195235,
      "epoch": 1.6784627092846272,
      "grad_norm": 6.0625,
      "learning_rate": 6.901467584376093e-07,
      "loss": 1.5107254,
      "memory(GiB)": 128.51,
      "step": 66165,
      "train_speed(iter/s)": 1.635898
    },
    {
      "acc": 0.64265833,
      "epoch": 1.6785895484525621,
      "grad_norm": 5.75,
      "learning_rate": 6.896152448674315e-07,
      "loss": 1.59157715,
      "memory(GiB)": 128.51,
      "step": 66170,
      "train_speed(iter/s)": 1.635911
    },
    {
      "acc": 0.6388649,
      "epoch": 1.6787163876204971,
      "grad_norm": 5.96875,
      "learning_rate": 6.890839208872185e-07,
      "loss": 1.64452457,
      "memory(GiB)": 128.51,
      "step": 66175,
      "train_speed(iter/s)": 1.635923
    },
    {
      "acc": 0.67246189,
      "epoch": 1.6788432267884321,
      "grad_norm": 6.75,
      "learning_rate": 6.885527865203401e-07,
      "loss": 1.55935669,
      "memory(GiB)": 128.51,
      "step": 66180,
      "train_speed(iter/s)": 1.635936
    },
    {
      "acc": 0.66789808,
      "epoch": 1.6789700659563673,
      "grad_norm": 5.34375,
      "learning_rate": 6.880218417901608e-07,
      "loss": 1.59915428,
      "memory(GiB)": 128.51,
      "step": 66185,
      "train_speed(iter/s)": 1.635949
    },
    {
      "acc": 0.6600852,
      "epoch": 1.6790969051243025,
      "grad_norm": 5.8125,
      "learning_rate": 6.874910867200341e-07,
      "loss": 1.62683716,
      "memory(GiB)": 128.51,
      "step": 66190,
      "train_speed(iter/s)": 1.635962
    },
    {
      "acc": 0.68290262,
      "epoch": 1.6792237442922375,
      "grad_norm": 5.78125,
      "learning_rate": 6.869605213333014e-07,
      "loss": 1.51938848,
      "memory(GiB)": 128.51,
      "step": 66195,
      "train_speed(iter/s)": 1.635973
    },
    {
      "acc": 0.66223049,
      "epoch": 1.6793505834601725,
      "grad_norm": 4.71875,
      "learning_rate": 6.864301456533007e-07,
      "loss": 1.58650331,
      "memory(GiB)": 128.51,
      "step": 66200,
      "train_speed(iter/s)": 1.635986
    },
    {
      "acc": 0.64770937,
      "epoch": 1.6794774226281075,
      "grad_norm": 5.875,
      "learning_rate": 6.858999597033617e-07,
      "loss": 1.65537148,
      "memory(GiB)": 128.51,
      "step": 66205,
      "train_speed(iter/s)": 1.635999
    },
    {
      "acc": 0.64754248,
      "epoch": 1.6796042617960425,
      "grad_norm": 5.21875,
      "learning_rate": 6.85369963506805e-07,
      "loss": 1.69379807,
      "memory(GiB)": 128.51,
      "step": 66210,
      "train_speed(iter/s)": 1.636012
    },
    {
      "acc": 0.63429809,
      "epoch": 1.6797311009639777,
      "grad_norm": 5.0625,
      "learning_rate": 6.848401570869384e-07,
      "loss": 1.69971447,
      "memory(GiB)": 128.51,
      "step": 66215,
      "train_speed(iter/s)": 1.636025
    },
    {
      "acc": 0.65975738,
      "epoch": 1.6798579401319127,
      "grad_norm": 5.875,
      "learning_rate": 6.843105404670685e-07,
      "loss": 1.54582596,
      "memory(GiB)": 128.51,
      "step": 66220,
      "train_speed(iter/s)": 1.636038
    },
    {
      "acc": 0.64358845,
      "epoch": 1.679984779299848,
      "grad_norm": 6.09375,
      "learning_rate": 6.837811136704892e-07,
      "loss": 1.63169899,
      "memory(GiB)": 128.51,
      "step": 66225,
      "train_speed(iter/s)": 1.63605
    },
    {
      "acc": 0.64815269,
      "epoch": 1.680111618467783,
      "grad_norm": 4.9375,
      "learning_rate": 6.832518767204882e-07,
      "loss": 1.58104324,
      "memory(GiB)": 128.51,
      "step": 66230,
      "train_speed(iter/s)": 1.636062
    },
    {
      "acc": 0.64621162,
      "epoch": 1.6802384576357179,
      "grad_norm": 5.3125,
      "learning_rate": 6.827228296403405e-07,
      "loss": 1.67636414,
      "memory(GiB)": 128.51,
      "step": 66235,
      "train_speed(iter/s)": 1.636076
    },
    {
      "acc": 0.65409412,
      "epoch": 1.6803652968036529,
      "grad_norm": 6.34375,
      "learning_rate": 6.821939724533189e-07,
      "loss": 1.63334789,
      "memory(GiB)": 128.51,
      "step": 66240,
      "train_speed(iter/s)": 1.636088
    },
    {
      "acc": 0.67309103,
      "epoch": 1.6804921359715879,
      "grad_norm": 6.0,
      "learning_rate": 6.816653051826838e-07,
      "loss": 1.57870569,
      "memory(GiB)": 128.51,
      "step": 66245,
      "train_speed(iter/s)": 1.636102
    },
    {
      "acc": 0.66775198,
      "epoch": 1.680618975139523,
      "grad_norm": 5.6875,
      "learning_rate": 6.811368278516889e-07,
      "loss": 1.58345413,
      "memory(GiB)": 128.51,
      "step": 66250,
      "train_speed(iter/s)": 1.636115
    },
    {
      "acc": 0.66788878,
      "epoch": 1.6807458143074583,
      "grad_norm": 8.75,
      "learning_rate": 6.806085404835788e-07,
      "loss": 1.59038954,
      "memory(GiB)": 128.51,
      "step": 66255,
      "train_speed(iter/s)": 1.63613
    },
    {
      "acc": 0.6525846,
      "epoch": 1.6808726534753933,
      "grad_norm": 8.1875,
      "learning_rate": 6.800804431015895e-07,
      "loss": 1.6459919,
      "memory(GiB)": 128.51,
      "step": 66260,
      "train_speed(iter/s)": 1.636143
    },
    {
      "acc": 0.6666266,
      "epoch": 1.6809994926433283,
      "grad_norm": 6.4375,
      "learning_rate": 6.795525357289496e-07,
      "loss": 1.5648243,
      "memory(GiB)": 128.51,
      "step": 66265,
      "train_speed(iter/s)": 1.636156
    },
    {
      "acc": 0.64701176,
      "epoch": 1.6811263318112633,
      "grad_norm": 5.34375,
      "learning_rate": 6.790248183888781e-07,
      "loss": 1.63595924,
      "memory(GiB)": 128.51,
      "step": 66270,
      "train_speed(iter/s)": 1.636169
    },
    {
      "acc": 0.66199794,
      "epoch": 1.6812531709791982,
      "grad_norm": 5.15625,
      "learning_rate": 6.784972911045872e-07,
      "loss": 1.61182404,
      "memory(GiB)": 128.51,
      "step": 66275,
      "train_speed(iter/s)": 1.636182
    },
    {
      "acc": 0.65428448,
      "epoch": 1.6813800101471335,
      "grad_norm": 5.90625,
      "learning_rate": 6.779699538992796e-07,
      "loss": 1.5914463,
      "memory(GiB)": 128.51,
      "step": 66280,
      "train_speed(iter/s)": 1.636195
    },
    {
      "acc": 0.65789757,
      "epoch": 1.6815068493150684,
      "grad_norm": 6.4375,
      "learning_rate": 6.774428067961502e-07,
      "loss": 1.65470734,
      "memory(GiB)": 128.51,
      "step": 66285,
      "train_speed(iter/s)": 1.636208
    },
    {
      "acc": 0.64024372,
      "epoch": 1.6816336884830037,
      "grad_norm": 6.09375,
      "learning_rate": 6.769158498183842e-07,
      "loss": 1.68790245,
      "memory(GiB)": 128.51,
      "step": 66290,
      "train_speed(iter/s)": 1.636222
    },
    {
      "acc": 0.67159791,
      "epoch": 1.6817605276509386,
      "grad_norm": 5.21875,
      "learning_rate": 6.763890829891611e-07,
      "loss": 1.58680639,
      "memory(GiB)": 128.51,
      "step": 66295,
      "train_speed(iter/s)": 1.636234
    },
    {
      "acc": 0.64143753,
      "epoch": 1.6818873668188736,
      "grad_norm": 5.28125,
      "learning_rate": 6.758625063316493e-07,
      "loss": 1.63990402,
      "memory(GiB)": 128.51,
      "step": 66300,
      "train_speed(iter/s)": 1.636246
    },
    {
      "acc": 0.67320423,
      "epoch": 1.6820142059868086,
      "grad_norm": 6.09375,
      "learning_rate": 6.75336119869009e-07,
      "loss": 1.5223978,
      "memory(GiB)": 128.51,
      "step": 66305,
      "train_speed(iter/s)": 1.636258
    },
    {
      "acc": 0.64519243,
      "epoch": 1.6821410451547438,
      "grad_norm": 5.75,
      "learning_rate": 6.748099236243971e-07,
      "loss": 1.6092411,
      "memory(GiB)": 128.51,
      "step": 66310,
      "train_speed(iter/s)": 1.63627
    },
    {
      "acc": 0.64731922,
      "epoch": 1.6822678843226788,
      "grad_norm": 6.0,
      "learning_rate": 6.742839176209537e-07,
      "loss": 1.64334335,
      "memory(GiB)": 128.51,
      "step": 66315,
      "train_speed(iter/s)": 1.636283
    },
    {
      "acc": 0.64213443,
      "epoch": 1.682394723490614,
      "grad_norm": 5.875,
      "learning_rate": 6.737581018818167e-07,
      "loss": 1.62842293,
      "memory(GiB)": 128.51,
      "step": 66320,
      "train_speed(iter/s)": 1.636296
    },
    {
      "acc": 0.65887785,
      "epoch": 1.682521562658549,
      "grad_norm": 7.78125,
      "learning_rate": 6.732324764301129e-07,
      "loss": 1.62973042,
      "memory(GiB)": 128.51,
      "step": 66325,
      "train_speed(iter/s)": 1.636308
    },
    {
      "acc": 0.65739026,
      "epoch": 1.682648401826484,
      "grad_norm": 5.15625,
      "learning_rate": 6.727070412889647e-07,
      "loss": 1.61480217,
      "memory(GiB)": 128.51,
      "step": 66330,
      "train_speed(iter/s)": 1.636321
    },
    {
      "acc": 0.66327405,
      "epoch": 1.682775240994419,
      "grad_norm": 6.15625,
      "learning_rate": 6.721817964814792e-07,
      "loss": 1.53326387,
      "memory(GiB)": 128.51,
      "step": 66335,
      "train_speed(iter/s)": 1.636334
    },
    {
      "acc": 0.67155638,
      "epoch": 1.682902080162354,
      "grad_norm": 5.65625,
      "learning_rate": 6.716567420307596e-07,
      "loss": 1.54714222,
      "memory(GiB)": 128.51,
      "step": 66340,
      "train_speed(iter/s)": 1.636348
    },
    {
      "acc": 0.65686874,
      "epoch": 1.6830289193302892,
      "grad_norm": 5.25,
      "learning_rate": 6.711318779599025e-07,
      "loss": 1.50884895,
      "memory(GiB)": 128.51,
      "step": 66345,
      "train_speed(iter/s)": 1.63636
    },
    {
      "acc": 0.67943659,
      "epoch": 1.6831557584982244,
      "grad_norm": 7.8125,
      "learning_rate": 6.706072042919931e-07,
      "loss": 1.48498478,
      "memory(GiB)": 128.51,
      "step": 66350,
      "train_speed(iter/s)": 1.636373
    },
    {
      "acc": 0.64314775,
      "epoch": 1.6832825976661594,
      "grad_norm": 5.9375,
      "learning_rate": 6.700827210501065e-07,
      "loss": 1.63353691,
      "memory(GiB)": 128.51,
      "step": 66355,
      "train_speed(iter/s)": 1.636387
    },
    {
      "acc": 0.66073294,
      "epoch": 1.6834094368340944,
      "grad_norm": 5.625,
      "learning_rate": 6.695584282573142e-07,
      "loss": 1.59654312,
      "memory(GiB)": 128.51,
      "step": 66360,
      "train_speed(iter/s)": 1.636399
    },
    {
      "acc": 0.65005159,
      "epoch": 1.6835362760020294,
      "grad_norm": 5.34375,
      "learning_rate": 6.690343259366766e-07,
      "loss": 1.5954752,
      "memory(GiB)": 128.51,
      "step": 66365,
      "train_speed(iter/s)": 1.636413
    },
    {
      "acc": 0.64390888,
      "epoch": 1.6836631151699644,
      "grad_norm": 4.96875,
      "learning_rate": 6.685104141112464e-07,
      "loss": 1.68616505,
      "memory(GiB)": 128.51,
      "step": 66370,
      "train_speed(iter/s)": 1.636426
    },
    {
      "acc": 0.67171893,
      "epoch": 1.6837899543378996,
      "grad_norm": 5.75,
      "learning_rate": 6.679866928040651e-07,
      "loss": 1.53451977,
      "memory(GiB)": 128.51,
      "step": 66375,
      "train_speed(iter/s)": 1.636438
    },
    {
      "acc": 0.6437346,
      "epoch": 1.6839167935058346,
      "grad_norm": 6.5,
      "learning_rate": 6.674631620381711e-07,
      "loss": 1.68510971,
      "memory(GiB)": 128.51,
      "step": 66380,
      "train_speed(iter/s)": 1.636451
    },
    {
      "acc": 0.6613471,
      "epoch": 1.6840436326737698,
      "grad_norm": 5.5,
      "learning_rate": 6.669398218365902e-07,
      "loss": 1.52771311,
      "memory(GiB)": 128.51,
      "step": 66385,
      "train_speed(iter/s)": 1.636464
    },
    {
      "acc": 0.66040182,
      "epoch": 1.6841704718417048,
      "grad_norm": 5.71875,
      "learning_rate": 6.664166722223426e-07,
      "loss": 1.57890501,
      "memory(GiB)": 128.51,
      "step": 66390,
      "train_speed(iter/s)": 1.636477
    },
    {
      "acc": 0.65939522,
      "epoch": 1.6842973110096398,
      "grad_norm": 4.8125,
      "learning_rate": 6.658937132184368e-07,
      "loss": 1.61428986,
      "memory(GiB)": 128.51,
      "step": 66395,
      "train_speed(iter/s)": 1.63649
    },
    {
      "acc": 0.66360259,
      "epoch": 1.6844241501775747,
      "grad_norm": 6.09375,
      "learning_rate": 6.653709448478762e-07,
      "loss": 1.56844444,
      "memory(GiB)": 128.51,
      "step": 66400,
      "train_speed(iter/s)": 1.636504
    },
    {
      "acc": 0.65322342,
      "epoch": 1.6845509893455097,
      "grad_norm": 6.46875,
      "learning_rate": 6.648483671336548e-07,
      "loss": 1.62249107,
      "memory(GiB)": 128.51,
      "step": 66405,
      "train_speed(iter/s)": 1.636519
    },
    {
      "acc": 0.63693781,
      "epoch": 1.684677828513445,
      "grad_norm": 6.4375,
      "learning_rate": 6.643259800987567e-07,
      "loss": 1.66031399,
      "memory(GiB)": 128.51,
      "step": 66410,
      "train_speed(iter/s)": 1.636531
    },
    {
      "acc": 0.65126677,
      "epoch": 1.6848046676813802,
      "grad_norm": 6.90625,
      "learning_rate": 6.638037837661593e-07,
      "loss": 1.58176994,
      "memory(GiB)": 128.51,
      "step": 66415,
      "train_speed(iter/s)": 1.636546
    },
    {
      "acc": 0.65425711,
      "epoch": 1.6849315068493151,
      "grad_norm": 4.6875,
      "learning_rate": 6.632817781588313e-07,
      "loss": 1.68862343,
      "memory(GiB)": 128.51,
      "step": 66420,
      "train_speed(iter/s)": 1.636559
    },
    {
      "acc": 0.64426546,
      "epoch": 1.6850583460172501,
      "grad_norm": 5.90625,
      "learning_rate": 6.627599632997328e-07,
      "loss": 1.66668797,
      "memory(GiB)": 128.51,
      "step": 66425,
      "train_speed(iter/s)": 1.636572
    },
    {
      "acc": 0.67221508,
      "epoch": 1.6851851851851851,
      "grad_norm": 5.75,
      "learning_rate": 6.622383392118153e-07,
      "loss": 1.60266933,
      "memory(GiB)": 128.51,
      "step": 66430,
      "train_speed(iter/s)": 1.636584
    },
    {
      "acc": 0.6361062,
      "epoch": 1.68531202435312,
      "grad_norm": 4.96875,
      "learning_rate": 6.617169059180229e-07,
      "loss": 1.64124107,
      "memory(GiB)": 128.51,
      "step": 66435,
      "train_speed(iter/s)": 1.636598
    },
    {
      "acc": 0.65117702,
      "epoch": 1.6854388635210553,
      "grad_norm": 5.34375,
      "learning_rate": 6.611956634412897e-07,
      "loss": 1.60670547,
      "memory(GiB)": 128.51,
      "step": 66440,
      "train_speed(iter/s)": 1.636612
    },
    {
      "acc": 0.6443337,
      "epoch": 1.6855657026889903,
      "grad_norm": 5.59375,
      "learning_rate": 6.606746118045415e-07,
      "loss": 1.5744648,
      "memory(GiB)": 128.51,
      "step": 66445,
      "train_speed(iter/s)": 1.636624
    },
    {
      "acc": 0.66201048,
      "epoch": 1.6856925418569255,
      "grad_norm": 7.34375,
      "learning_rate": 6.601537510306999e-07,
      "loss": 1.58101501,
      "memory(GiB)": 128.51,
      "step": 66450,
      "train_speed(iter/s)": 1.636638
    },
    {
      "acc": 0.63887186,
      "epoch": 1.6858193810248605,
      "grad_norm": 5.9375,
      "learning_rate": 6.59633081142671e-07,
      "loss": 1.68242455,
      "memory(GiB)": 128.51,
      "step": 66455,
      "train_speed(iter/s)": 1.636652
    },
    {
      "acc": 0.659795,
      "epoch": 1.6859462201927955,
      "grad_norm": 7.0,
      "learning_rate": 6.591126021633575e-07,
      "loss": 1.55775766,
      "memory(GiB)": 128.51,
      "step": 66460,
      "train_speed(iter/s)": 1.636666
    },
    {
      "acc": 0.65130558,
      "epoch": 1.6860730593607305,
      "grad_norm": 5.71875,
      "learning_rate": 6.585923141156513e-07,
      "loss": 1.60600815,
      "memory(GiB)": 128.51,
      "step": 66465,
      "train_speed(iter/s)": 1.63668
    },
    {
      "acc": 0.6596117,
      "epoch": 1.6861998985286657,
      "grad_norm": 5.84375,
      "learning_rate": 6.580722170224408e-07,
      "loss": 1.55361948,
      "memory(GiB)": 128.51,
      "step": 66470,
      "train_speed(iter/s)": 1.636694
    },
    {
      "acc": 0.65276661,
      "epoch": 1.6863267376966007,
      "grad_norm": 6.28125,
      "learning_rate": 6.575523109065979e-07,
      "loss": 1.62090073,
      "memory(GiB)": 128.51,
      "step": 66475,
      "train_speed(iter/s)": 1.636707
    },
    {
      "acc": 0.65026875,
      "epoch": 1.686453576864536,
      "grad_norm": 5.0625,
      "learning_rate": 6.570325957909912e-07,
      "loss": 1.63086548,
      "memory(GiB)": 128.51,
      "step": 66480,
      "train_speed(iter/s)": 1.63672
    },
    {
      "acc": 0.66136537,
      "epoch": 1.6865804160324709,
      "grad_norm": 4.84375,
      "learning_rate": 6.565130716984819e-07,
      "loss": 1.57797918,
      "memory(GiB)": 128.51,
      "step": 66485,
      "train_speed(iter/s)": 1.636734
    },
    {
      "acc": 0.65649347,
      "epoch": 1.6867072552004059,
      "grad_norm": 4.90625,
      "learning_rate": 6.55993738651921e-07,
      "loss": 1.58404427,
      "memory(GiB)": 128.51,
      "step": 66490,
      "train_speed(iter/s)": 1.636747
    },
    {
      "acc": 0.65040445,
      "epoch": 1.6868340943683409,
      "grad_norm": 6.09375,
      "learning_rate": 6.554745966741488e-07,
      "loss": 1.6970562,
      "memory(GiB)": 128.51,
      "step": 66495,
      "train_speed(iter/s)": 1.63676
    },
    {
      "acc": 0.64716797,
      "epoch": 1.6869609335362759,
      "grad_norm": 6.875,
      "learning_rate": 6.549556457879996e-07,
      "loss": 1.61366329,
      "memory(GiB)": 128.51,
      "step": 66500,
      "train_speed(iter/s)": 1.636774
    },
    {
      "acc": 0.65494671,
      "epoch": 1.687087772704211,
      "grad_norm": 5.53125,
      "learning_rate": 6.54436886016302e-07,
      "loss": 1.60575371,
      "memory(GiB)": 128.51,
      "step": 66505,
      "train_speed(iter/s)": 1.636787
    },
    {
      "acc": 0.64323206,
      "epoch": 1.6872146118721463,
      "grad_norm": 6.375,
      "learning_rate": 6.539183173818725e-07,
      "loss": 1.60546474,
      "memory(GiB)": 128.51,
      "step": 66510,
      "train_speed(iter/s)": 1.6368
    },
    {
      "acc": 0.65925994,
      "epoch": 1.6873414510400813,
      "grad_norm": 5.1875,
      "learning_rate": 6.53399939907517e-07,
      "loss": 1.58527508,
      "memory(GiB)": 128.51,
      "step": 66515,
      "train_speed(iter/s)": 1.636813
    },
    {
      "acc": 0.67269168,
      "epoch": 1.6874682902080163,
      "grad_norm": 6.3125,
      "learning_rate": 6.528817536160392e-07,
      "loss": 1.5747426,
      "memory(GiB)": 128.51,
      "step": 66520,
      "train_speed(iter/s)": 1.636826
    },
    {
      "acc": 0.66061754,
      "epoch": 1.6875951293759512,
      "grad_norm": 7.03125,
      "learning_rate": 6.523637585302311e-07,
      "loss": 1.65887413,
      "memory(GiB)": 128.51,
      "step": 66525,
      "train_speed(iter/s)": 1.636841
    },
    {
      "acc": 0.65177183,
      "epoch": 1.6877219685438862,
      "grad_norm": 4.90625,
      "learning_rate": 6.51845954672875e-07,
      "loss": 1.55776482,
      "memory(GiB)": 128.51,
      "step": 66530,
      "train_speed(iter/s)": 1.636855
    },
    {
      "acc": 0.6554522,
      "epoch": 1.6878488077118214,
      "grad_norm": 8.25,
      "learning_rate": 6.513283420667471e-07,
      "loss": 1.69159813,
      "memory(GiB)": 128.51,
      "step": 66535,
      "train_speed(iter/s)": 1.636868
    },
    {
      "acc": 0.6526967,
      "epoch": 1.6879756468797564,
      "grad_norm": 6.6875,
      "learning_rate": 6.508109207346142e-07,
      "loss": 1.63218689,
      "memory(GiB)": 128.51,
      "step": 66540,
      "train_speed(iter/s)": 1.636882
    },
    {
      "acc": 0.6490098,
      "epoch": 1.6881024860476916,
      "grad_norm": 6.0,
      "learning_rate": 6.502936906992346e-07,
      "loss": 1.61840134,
      "memory(GiB)": 128.51,
      "step": 66545,
      "train_speed(iter/s)": 1.636894
    },
    {
      "acc": 0.64785519,
      "epoch": 1.6882293252156266,
      "grad_norm": 4.90625,
      "learning_rate": 6.497766519833587e-07,
      "loss": 1.52472229,
      "memory(GiB)": 128.51,
      "step": 66550,
      "train_speed(iter/s)": 1.636908
    },
    {
      "acc": 0.66280384,
      "epoch": 1.6883561643835616,
      "grad_norm": 5.0,
      "learning_rate": 6.492598046097282e-07,
      "loss": 1.57390785,
      "memory(GiB)": 128.51,
      "step": 66555,
      "train_speed(iter/s)": 1.636921
    },
    {
      "acc": 0.65174108,
      "epoch": 1.6884830035514966,
      "grad_norm": 5.65625,
      "learning_rate": 6.487431486010759e-07,
      "loss": 1.63463612,
      "memory(GiB)": 128.51,
      "step": 66560,
      "train_speed(iter/s)": 1.636936
    },
    {
      "acc": 0.64512362,
      "epoch": 1.6886098427194316,
      "grad_norm": 6.78125,
      "learning_rate": 6.482266839801265e-07,
      "loss": 1.62295952,
      "memory(GiB)": 128.51,
      "step": 66565,
      "train_speed(iter/s)": 1.636949
    },
    {
      "acc": 0.66724973,
      "epoch": 1.6887366818873668,
      "grad_norm": 4.5,
      "learning_rate": 6.477104107695975e-07,
      "loss": 1.58875866,
      "memory(GiB)": 128.51,
      "step": 66570,
      "train_speed(iter/s)": 1.636963
    },
    {
      "acc": 0.65869265,
      "epoch": 1.688863521055302,
      "grad_norm": 6.28125,
      "learning_rate": 6.471943289921955e-07,
      "loss": 1.5639576,
      "memory(GiB)": 128.51,
      "step": 66575,
      "train_speed(iter/s)": 1.636977
    },
    {
      "acc": 0.66198511,
      "epoch": 1.688990360223237,
      "grad_norm": 5.6875,
      "learning_rate": 6.466784386706215e-07,
      "loss": 1.52178745,
      "memory(GiB)": 128.51,
      "step": 66580,
      "train_speed(iter/s)": 1.63699
    },
    {
      "acc": 0.65528412,
      "epoch": 1.689117199391172,
      "grad_norm": 4.6875,
      "learning_rate": 6.461627398275655e-07,
      "loss": 1.59580669,
      "memory(GiB)": 128.51,
      "step": 66585,
      "train_speed(iter/s)": 1.637004
    },
    {
      "acc": 0.67058916,
      "epoch": 1.689244038559107,
      "grad_norm": 5.53125,
      "learning_rate": 6.456472324857111e-07,
      "loss": 1.66829681,
      "memory(GiB)": 128.51,
      "step": 66590,
      "train_speed(iter/s)": 1.637016
    },
    {
      "acc": 0.64319115,
      "epoch": 1.689370877727042,
      "grad_norm": 8.25,
      "learning_rate": 6.451319166677317e-07,
      "loss": 1.64337044,
      "memory(GiB)": 128.51,
      "step": 66595,
      "train_speed(iter/s)": 1.637029
    },
    {
      "acc": 0.66640234,
      "epoch": 1.6894977168949772,
      "grad_norm": 6.1875,
      "learning_rate": 6.446167923962943e-07,
      "loss": 1.53427582,
      "memory(GiB)": 128.51,
      "step": 66600,
      "train_speed(iter/s)": 1.637043
    },
    {
      "acc": 0.65677981,
      "epoch": 1.6896245560629122,
      "grad_norm": 5.8125,
      "learning_rate": 6.44101859694054e-07,
      "loss": 1.63551998,
      "memory(GiB)": 128.51,
      "step": 66605,
      "train_speed(iter/s)": 1.637056
    },
    {
      "acc": 0.65459723,
      "epoch": 1.6897513952308474,
      "grad_norm": 5.125,
      "learning_rate": 6.43587118583664e-07,
      "loss": 1.55850086,
      "memory(GiB)": 128.51,
      "step": 66610,
      "train_speed(iter/s)": 1.63707
    },
    {
      "acc": 0.6696631,
      "epoch": 1.6898782343987824,
      "grad_norm": 5.46875,
      "learning_rate": 6.430725690877615e-07,
      "loss": 1.53732719,
      "memory(GiB)": 128.51,
      "step": 66615,
      "train_speed(iter/s)": 1.637083
    },
    {
      "acc": 0.65037088,
      "epoch": 1.6900050735667174,
      "grad_norm": 5.28125,
      "learning_rate": 6.425582112289785e-07,
      "loss": 1.59629421,
      "memory(GiB)": 128.51,
      "step": 66620,
      "train_speed(iter/s)": 1.637097
    },
    {
      "acc": 0.6654757,
      "epoch": 1.6901319127346524,
      "grad_norm": 5.09375,
      "learning_rate": 6.420440450299414e-07,
      "loss": 1.47901516,
      "memory(GiB)": 128.51,
      "step": 66625,
      "train_speed(iter/s)": 1.63711
    },
    {
      "acc": 0.67324791,
      "epoch": 1.6902587519025876,
      "grad_norm": 6.375,
      "learning_rate": 6.415300705132648e-07,
      "loss": 1.49997272,
      "memory(GiB)": 128.51,
      "step": 66630,
      "train_speed(iter/s)": 1.637123
    },
    {
      "acc": 0.65585203,
      "epoch": 1.6903855910705226,
      "grad_norm": 5.90625,
      "learning_rate": 6.410162877015535e-07,
      "loss": 1.57841148,
      "memory(GiB)": 128.51,
      "step": 66635,
      "train_speed(iter/s)": 1.637137
    },
    {
      "acc": 0.66604357,
      "epoch": 1.6905124302384578,
      "grad_norm": 5.65625,
      "learning_rate": 6.405026966174066e-07,
      "loss": 1.58571539,
      "memory(GiB)": 128.51,
      "step": 66640,
      "train_speed(iter/s)": 1.637151
    },
    {
      "acc": 0.66064062,
      "epoch": 1.6906392694063928,
      "grad_norm": 5.46875,
      "learning_rate": 6.39989297283416e-07,
      "loss": 1.52937279,
      "memory(GiB)": 128.51,
      "step": 66645,
      "train_speed(iter/s)": 1.637164
    },
    {
      "acc": 0.64689898,
      "epoch": 1.6907661085743277,
      "grad_norm": 4.875,
      "learning_rate": 6.394760897221636e-07,
      "loss": 1.63860416,
      "memory(GiB)": 128.51,
      "step": 66650,
      "train_speed(iter/s)": 1.637177
    },
    {
      "acc": 0.62799459,
      "epoch": 1.6908929477422627,
      "grad_norm": 5.96875,
      "learning_rate": 6.389630739562186e-07,
      "loss": 1.69764309,
      "memory(GiB)": 128.51,
      "step": 66655,
      "train_speed(iter/s)": 1.63719
    },
    {
      "acc": 0.65750885,
      "epoch": 1.6910197869101977,
      "grad_norm": 5.25,
      "learning_rate": 6.38450250008149e-07,
      "loss": 1.56338425,
      "memory(GiB)": 128.51,
      "step": 66660,
      "train_speed(iter/s)": 1.637203
    },
    {
      "acc": 0.66203971,
      "epoch": 1.691146626078133,
      "grad_norm": 5.15625,
      "learning_rate": 6.379376179005103e-07,
      "loss": 1.56963091,
      "memory(GiB)": 128.51,
      "step": 66665,
      "train_speed(iter/s)": 1.637217
    },
    {
      "acc": 0.66247883,
      "epoch": 1.6912734652460681,
      "grad_norm": 6.6875,
      "learning_rate": 6.374251776558521e-07,
      "loss": 1.57038927,
      "memory(GiB)": 128.51,
      "step": 66670,
      "train_speed(iter/s)": 1.637231
    },
    {
      "acc": 0.65231047,
      "epoch": 1.6914003044140031,
      "grad_norm": 5.0625,
      "learning_rate": 6.369129292967097e-07,
      "loss": 1.55709057,
      "memory(GiB)": 128.51,
      "step": 66675,
      "train_speed(iter/s)": 1.637243
    },
    {
      "acc": 0.65538011,
      "epoch": 1.6915271435819381,
      "grad_norm": 5.4375,
      "learning_rate": 6.364008728456173e-07,
      "loss": 1.53696947,
      "memory(GiB)": 128.51,
      "step": 66680,
      "train_speed(iter/s)": 1.637257
    },
    {
      "acc": 0.65603461,
      "epoch": 1.691653982749873,
      "grad_norm": 5.8125,
      "learning_rate": 6.358890083250963e-07,
      "loss": 1.61610146,
      "memory(GiB)": 128.51,
      "step": 66685,
      "train_speed(iter/s)": 1.63727
    },
    {
      "acc": 0.6634779,
      "epoch": 1.691780821917808,
      "grad_norm": 5.8125,
      "learning_rate": 6.353773357576615e-07,
      "loss": 1.5710845,
      "memory(GiB)": 128.51,
      "step": 66690,
      "train_speed(iter/s)": 1.637283
    },
    {
      "acc": 0.6633091,
      "epoch": 1.6919076610857433,
      "grad_norm": 5.4375,
      "learning_rate": 6.348658551658182e-07,
      "loss": 1.53256826,
      "memory(GiB)": 128.51,
      "step": 66695,
      "train_speed(iter/s)": 1.637296
    },
    {
      "acc": 0.65155902,
      "epoch": 1.6920345002536783,
      "grad_norm": 6.9375,
      "learning_rate": 6.343545665720636e-07,
      "loss": 1.66027565,
      "memory(GiB)": 128.51,
      "step": 66700,
      "train_speed(iter/s)": 1.637308
    },
    {
      "acc": 0.64160042,
      "epoch": 1.6921613394216135,
      "grad_norm": 6.6875,
      "learning_rate": 6.33843469998886e-07,
      "loss": 1.68734779,
      "memory(GiB)": 128.51,
      "step": 66705,
      "train_speed(iter/s)": 1.637321
    },
    {
      "acc": 0.65719976,
      "epoch": 1.6922881785895485,
      "grad_norm": 5.34375,
      "learning_rate": 6.33332565468766e-07,
      "loss": 1.57576942,
      "memory(GiB)": 128.51,
      "step": 66710,
      "train_speed(iter/s)": 1.637334
    },
    {
      "acc": 0.64508729,
      "epoch": 1.6924150177574835,
      "grad_norm": 6.1875,
      "learning_rate": 6.328218530041757e-07,
      "loss": 1.67298508,
      "memory(GiB)": 128.51,
      "step": 66715,
      "train_speed(iter/s)": 1.637347
    },
    {
      "acc": 0.64120893,
      "epoch": 1.6925418569254185,
      "grad_norm": 5.34375,
      "learning_rate": 6.323113326275781e-07,
      "loss": 1.62973709,
      "memory(GiB)": 128.51,
      "step": 66720,
      "train_speed(iter/s)": 1.63736
    },
    {
      "acc": 0.63750439,
      "epoch": 1.6926686960933535,
      "grad_norm": 5.75,
      "learning_rate": 6.318010043614292e-07,
      "loss": 1.69637928,
      "memory(GiB)": 128.51,
      "step": 66725,
      "train_speed(iter/s)": 1.637372
    },
    {
      "acc": 0.66092091,
      "epoch": 1.6927955352612887,
      "grad_norm": 5.21875,
      "learning_rate": 6.312908682281743e-07,
      "loss": 1.547859,
      "memory(GiB)": 128.51,
      "step": 66730,
      "train_speed(iter/s)": 1.637385
    },
    {
      "acc": 0.68325677,
      "epoch": 1.6929223744292239,
      "grad_norm": 5.53125,
      "learning_rate": 6.307809242502517e-07,
      "loss": 1.51669903,
      "memory(GiB)": 128.51,
      "step": 66735,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.65210657,
      "epoch": 1.6930492135971589,
      "grad_norm": 6.03125,
      "learning_rate": 6.302711724500909e-07,
      "loss": 1.61665249,
      "memory(GiB)": 128.51,
      "step": 66740,
      "train_speed(iter/s)": 1.63741
    },
    {
      "acc": 0.66810455,
      "epoch": 1.6931760527650939,
      "grad_norm": 5.84375,
      "learning_rate": 6.297616128501133e-07,
      "loss": 1.57994995,
      "memory(GiB)": 128.51,
      "step": 66745,
      "train_speed(iter/s)": 1.637423
    },
    {
      "acc": 0.65783567,
      "epoch": 1.6933028919330289,
      "grad_norm": 10.0,
      "learning_rate": 6.292522454727329e-07,
      "loss": 1.61713924,
      "memory(GiB)": 128.51,
      "step": 66750,
      "train_speed(iter/s)": 1.637437
    },
    {
      "acc": 0.64224997,
      "epoch": 1.6934297311009638,
      "grad_norm": 7.0625,
      "learning_rate": 6.287430703403524e-07,
      "loss": 1.60276871,
      "memory(GiB)": 128.51,
      "step": 66755,
      "train_speed(iter/s)": 1.63745
    },
    {
      "acc": 0.6657033,
      "epoch": 1.693556570268899,
      "grad_norm": 6.34375,
      "learning_rate": 6.282340874753673e-07,
      "loss": 1.55315561,
      "memory(GiB)": 128.51,
      "step": 66760,
      "train_speed(iter/s)": 1.637463
    },
    {
      "acc": 0.65935316,
      "epoch": 1.693683409436834,
      "grad_norm": 5.15625,
      "learning_rate": 6.277252969001646e-07,
      "loss": 1.577668,
      "memory(GiB)": 128.51,
      "step": 66765,
      "train_speed(iter/s)": 1.637477
    },
    {
      "acc": 0.68309288,
      "epoch": 1.6938102486047693,
      "grad_norm": 6.09375,
      "learning_rate": 6.272166986371264e-07,
      "loss": 1.56526871,
      "memory(GiB)": 128.51,
      "step": 66770,
      "train_speed(iter/s)": 1.637489
    },
    {
      "acc": 0.66563978,
      "epoch": 1.6939370877727042,
      "grad_norm": 6.25,
      "learning_rate": 6.267082927086199e-07,
      "loss": 1.56169863,
      "memory(GiB)": 128.51,
      "step": 66775,
      "train_speed(iter/s)": 1.637503
    },
    {
      "acc": 0.64557629,
      "epoch": 1.6940639269406392,
      "grad_norm": 5.03125,
      "learning_rate": 6.262000791370071e-07,
      "loss": 1.61343136,
      "memory(GiB)": 128.51,
      "step": 66780,
      "train_speed(iter/s)": 1.637515
    },
    {
      "acc": 0.65218968,
      "epoch": 1.6941907661085742,
      "grad_norm": 4.8125,
      "learning_rate": 6.256920579446429e-07,
      "loss": 1.61522121,
      "memory(GiB)": 128.51,
      "step": 66785,
      "train_speed(iter/s)": 1.637527
    },
    {
      "acc": 0.63828354,
      "epoch": 1.6943176052765094,
      "grad_norm": 5.96875,
      "learning_rate": 6.251842291538734e-07,
      "loss": 1.69617958,
      "memory(GiB)": 128.51,
      "step": 66790,
      "train_speed(iter/s)": 1.637541
    },
    {
      "acc": 0.66302691,
      "epoch": 1.6944444444444444,
      "grad_norm": 5.90625,
      "learning_rate": 6.246765927870313e-07,
      "loss": 1.59541874,
      "memory(GiB)": 128.51,
      "step": 66795,
      "train_speed(iter/s)": 1.637554
    },
    {
      "acc": 0.65665331,
      "epoch": 1.6945712836123796,
      "grad_norm": 5.96875,
      "learning_rate": 6.241691488664486e-07,
      "loss": 1.62578659,
      "memory(GiB)": 128.51,
      "step": 66800,
      "train_speed(iter/s)": 1.637567
    },
    {
      "acc": 0.65172,
      "epoch": 1.6946981227803146,
      "grad_norm": 5.71875,
      "learning_rate": 6.236618974144432e-07,
      "loss": 1.63236542,
      "memory(GiB)": 128.51,
      "step": 66805,
      "train_speed(iter/s)": 1.637579
    },
    {
      "acc": 0.6651423,
      "epoch": 1.6948249619482496,
      "grad_norm": 6.0625,
      "learning_rate": 6.23154838453327e-07,
      "loss": 1.53436604,
      "memory(GiB)": 128.51,
      "step": 66810,
      "train_speed(iter/s)": 1.637591
    },
    {
      "acc": 0.65798717,
      "epoch": 1.6949518011161846,
      "grad_norm": 6.0625,
      "learning_rate": 6.226479720054007e-07,
      "loss": 1.55678043,
      "memory(GiB)": 128.51,
      "step": 66815,
      "train_speed(iter/s)": 1.637605
    },
    {
      "acc": 0.6566298,
      "epoch": 1.6950786402841196,
      "grad_norm": 8.3125,
      "learning_rate": 6.221412980929608e-07,
      "loss": 1.64250984,
      "memory(GiB)": 128.51,
      "step": 66820,
      "train_speed(iter/s)": 1.637618
    },
    {
      "acc": 0.64589119,
      "epoch": 1.6952054794520548,
      "grad_norm": 8.75,
      "learning_rate": 6.21634816738293e-07,
      "loss": 1.57805367,
      "memory(GiB)": 128.51,
      "step": 66825,
      "train_speed(iter/s)": 1.637631
    },
    {
      "acc": 0.65637679,
      "epoch": 1.69533231861999,
      "grad_norm": 5.09375,
      "learning_rate": 6.211285279636731e-07,
      "loss": 1.55608921,
      "memory(GiB)": 128.51,
      "step": 66830,
      "train_speed(iter/s)": 1.637644
    },
    {
      "acc": 0.64408159,
      "epoch": 1.695459157787925,
      "grad_norm": 6.53125,
      "learning_rate": 6.206224317913711e-07,
      "loss": 1.64528217,
      "memory(GiB)": 128.51,
      "step": 66835,
      "train_speed(iter/s)": 1.637658
    },
    {
      "acc": 0.65190306,
      "epoch": 1.69558599695586,
      "grad_norm": 5.53125,
      "learning_rate": 6.201165282436473e-07,
      "loss": 1.58617287,
      "memory(GiB)": 128.51,
      "step": 66840,
      "train_speed(iter/s)": 1.637671
    },
    {
      "acc": 0.66689792,
      "epoch": 1.695712836123795,
      "grad_norm": 5.96875,
      "learning_rate": 6.196108173427529e-07,
      "loss": 1.55203533,
      "memory(GiB)": 128.51,
      "step": 66845,
      "train_speed(iter/s)": 1.637684
    },
    {
      "acc": 0.65467238,
      "epoch": 1.69583967529173,
      "grad_norm": 5.71875,
      "learning_rate": 6.19105299110932e-07,
      "loss": 1.69402924,
      "memory(GiB)": 128.51,
      "step": 66850,
      "train_speed(iter/s)": 1.637697
    },
    {
      "acc": 0.65164242,
      "epoch": 1.6959665144596652,
      "grad_norm": 4.84375,
      "learning_rate": 6.185999735704195e-07,
      "loss": 1.56395731,
      "memory(GiB)": 128.51,
      "step": 66855,
      "train_speed(iter/s)": 1.637709
    },
    {
      "acc": 0.66126356,
      "epoch": 1.6960933536276002,
      "grad_norm": 6.4375,
      "learning_rate": 6.180948407434417e-07,
      "loss": 1.50182371,
      "memory(GiB)": 128.51,
      "step": 66860,
      "train_speed(iter/s)": 1.637722
    },
    {
      "acc": 0.66517563,
      "epoch": 1.6962201927955354,
      "grad_norm": 8.25,
      "learning_rate": 6.175899006522162e-07,
      "loss": 1.56674652,
      "memory(GiB)": 128.51,
      "step": 66865,
      "train_speed(iter/s)": 1.637736
    },
    {
      "acc": 0.64143372,
      "epoch": 1.6963470319634704,
      "grad_norm": 5.5625,
      "learning_rate": 6.170851533189537e-07,
      "loss": 1.6080389,
      "memory(GiB)": 128.51,
      "step": 66870,
      "train_speed(iter/s)": 1.637748
    },
    {
      "acc": 0.67202463,
      "epoch": 1.6964738711314054,
      "grad_norm": 5.625,
      "learning_rate": 6.165805987658535e-07,
      "loss": 1.52700891,
      "memory(GiB)": 128.51,
      "step": 66875,
      "train_speed(iter/s)": 1.637762
    },
    {
      "acc": 0.66692219,
      "epoch": 1.6966007102993403,
      "grad_norm": 4.96875,
      "learning_rate": 6.160762370151097e-07,
      "loss": 1.60886707,
      "memory(GiB)": 128.51,
      "step": 66880,
      "train_speed(iter/s)": 1.637774
    },
    {
      "acc": 0.6600338,
      "epoch": 1.6967275494672753,
      "grad_norm": 6.5,
      "learning_rate": 6.155720680889049e-07,
      "loss": 1.6563942,
      "memory(GiB)": 128.51,
      "step": 66885,
      "train_speed(iter/s)": 1.637576
    },
    {
      "acc": 0.6666203,
      "epoch": 1.6968543886352105,
      "grad_norm": 6.03125,
      "learning_rate": 6.150680920094171e-07,
      "loss": 1.6286375,
      "memory(GiB)": 128.51,
      "step": 66890,
      "train_speed(iter/s)": 1.637589
    },
    {
      "acc": 0.64762788,
      "epoch": 1.6969812278031458,
      "grad_norm": 5.4375,
      "learning_rate": 6.145643087988113e-07,
      "loss": 1.62744045,
      "memory(GiB)": 128.51,
      "step": 66895,
      "train_speed(iter/s)": 1.637384
    },
    {
      "acc": 0.6326457,
      "epoch": 1.6971080669710807,
      "grad_norm": 7.6875,
      "learning_rate": 6.140607184792469e-07,
      "loss": 1.67915802,
      "memory(GiB)": 128.51,
      "step": 66900,
      "train_speed(iter/s)": 1.637398
    },
    {
      "acc": 0.66274881,
      "epoch": 1.6972349061390157,
      "grad_norm": 7.28125,
      "learning_rate": 6.135573210728724e-07,
      "loss": 1.61145687,
      "memory(GiB)": 128.51,
      "step": 66905,
      "train_speed(iter/s)": 1.63741
    },
    {
      "acc": 0.66126814,
      "epoch": 1.6973617453069507,
      "grad_norm": 5.59375,
      "learning_rate": 6.130541166018333e-07,
      "loss": 1.50673847,
      "memory(GiB)": 128.51,
      "step": 66910,
      "train_speed(iter/s)": 1.637424
    },
    {
      "acc": 0.64857712,
      "epoch": 1.6974885844748857,
      "grad_norm": 4.8125,
      "learning_rate": 6.125511050882598e-07,
      "loss": 1.62452621,
      "memory(GiB)": 128.51,
      "step": 66915,
      "train_speed(iter/s)": 1.637436
    },
    {
      "acc": 0.65366311,
      "epoch": 1.697615423642821,
      "grad_norm": 6.625,
      "learning_rate": 6.12048286554276e-07,
      "loss": 1.56389093,
      "memory(GiB)": 128.51,
      "step": 66920,
      "train_speed(iter/s)": 1.63745
    },
    {
      "acc": 0.66152525,
      "epoch": 1.697742262810756,
      "grad_norm": 6.65625,
      "learning_rate": 6.11545661022e-07,
      "loss": 1.59166803,
      "memory(GiB)": 128.51,
      "step": 66925,
      "train_speed(iter/s)": 1.637462
    },
    {
      "acc": 0.64788628,
      "epoch": 1.6978691019786911,
      "grad_norm": 4.9375,
      "learning_rate": 6.110432285135404e-07,
      "loss": 1.604673,
      "memory(GiB)": 128.51,
      "step": 66930,
      "train_speed(iter/s)": 1.637475
    },
    {
      "acc": 0.66452541,
      "epoch": 1.697995941146626,
      "grad_norm": 5.09375,
      "learning_rate": 6.105409890509933e-07,
      "loss": 1.5599659,
      "memory(GiB)": 128.51,
      "step": 66935,
      "train_speed(iter/s)": 1.637489
    },
    {
      "acc": 0.64601746,
      "epoch": 1.698122780314561,
      "grad_norm": 5.15625,
      "learning_rate": 6.100389426564501e-07,
      "loss": 1.61214828,
      "memory(GiB)": 128.51,
      "step": 66940,
      "train_speed(iter/s)": 1.637502
    },
    {
      "acc": 0.65639439,
      "epoch": 1.698249619482496,
      "grad_norm": 5.21875,
      "learning_rate": 6.095370893519953e-07,
      "loss": 1.6274086,
      "memory(GiB)": 128.51,
      "step": 66945,
      "train_speed(iter/s)": 1.637514
    },
    {
      "acc": 0.66981788,
      "epoch": 1.6983764586504313,
      "grad_norm": 5.40625,
      "learning_rate": 6.090354291597022e-07,
      "loss": 1.5581728,
      "memory(GiB)": 128.51,
      "step": 66950,
      "train_speed(iter/s)": 1.637526
    },
    {
      "acc": 0.65510864,
      "epoch": 1.6985032978183663,
      "grad_norm": 6.5625,
      "learning_rate": 6.085339621016334e-07,
      "loss": 1.58307333,
      "memory(GiB)": 128.51,
      "step": 66955,
      "train_speed(iter/s)": 1.637538
    },
    {
      "acc": 0.66451969,
      "epoch": 1.6986301369863015,
      "grad_norm": 5.625,
      "learning_rate": 6.080326881998482e-07,
      "loss": 1.49869213,
      "memory(GiB)": 128.51,
      "step": 66960,
      "train_speed(iter/s)": 1.63755
    },
    {
      "acc": 0.66328754,
      "epoch": 1.6987569761542365,
      "grad_norm": 5.96875,
      "learning_rate": 6.075316074763943e-07,
      "loss": 1.57312756,
      "memory(GiB)": 128.51,
      "step": 66965,
      "train_speed(iter/s)": 1.637562
    },
    {
      "acc": 0.65397367,
      "epoch": 1.6988838153221715,
      "grad_norm": 5.0625,
      "learning_rate": 6.070307199533109e-07,
      "loss": 1.61045895,
      "memory(GiB)": 128.51,
      "step": 66970,
      "train_speed(iter/s)": 1.637575
    },
    {
      "acc": 0.6604332,
      "epoch": 1.6990106544901065,
      "grad_norm": 5.6875,
      "learning_rate": 6.065300256526297e-07,
      "loss": 1.59720001,
      "memory(GiB)": 128.51,
      "step": 66975,
      "train_speed(iter/s)": 1.637587
    },
    {
      "acc": 0.65703821,
      "epoch": 1.6991374936580415,
      "grad_norm": 7.125,
      "learning_rate": 6.060295245963737e-07,
      "loss": 1.61264229,
      "memory(GiB)": 128.51,
      "step": 66980,
      "train_speed(iter/s)": 1.637601
    },
    {
      "acc": 0.65314856,
      "epoch": 1.6992643328259767,
      "grad_norm": 5.625,
      "learning_rate": 6.055292168065563e-07,
      "loss": 1.53722334,
      "memory(GiB)": 128.51,
      "step": 66985,
      "train_speed(iter/s)": 1.637613
    },
    {
      "acc": 0.65193248,
      "epoch": 1.6993911719939119,
      "grad_norm": 6.53125,
      "learning_rate": 6.050291023051846e-07,
      "loss": 1.62349396,
      "memory(GiB)": 128.51,
      "step": 66990,
      "train_speed(iter/s)": 1.637626
    },
    {
      "acc": 0.6468997,
      "epoch": 1.6995180111618469,
      "grad_norm": 7.71875,
      "learning_rate": 6.045291811142545e-07,
      "loss": 1.61090851,
      "memory(GiB)": 128.51,
      "step": 66995,
      "train_speed(iter/s)": 1.637639
    },
    {
      "acc": 0.66547103,
      "epoch": 1.6996448503297819,
      "grad_norm": 6.0,
      "learning_rate": 6.040294532557555e-07,
      "loss": 1.57564735,
      "memory(GiB)": 128.51,
      "step": 67000,
      "train_speed(iter/s)": 1.637652
    },
    {
      "epoch": 1.6996448503297819,
      "eval_acc": 0.6463044290884336,
      "eval_loss": 1.5730608701705933,
      "eval_runtime": 58.275,
      "eval_samples_per_second": 109.309,
      "eval_steps_per_second": 27.336,
      "step": 67000
    },
    {
      "acc": 0.67243071,
      "epoch": 1.6997716894977168,
      "grad_norm": 5.1875,
      "learning_rate": 6.03529918751668e-07,
      "loss": 1.56185894,
      "memory(GiB)": 128.51,
      "step": 67005,
      "train_speed(iter/s)": 1.63516
    },
    {
      "acc": 0.65515356,
      "epoch": 1.6998985286656518,
      "grad_norm": 5.8125,
      "learning_rate": 6.030305776239637e-07,
      "loss": 1.61485138,
      "memory(GiB)": 128.51,
      "step": 67010,
      "train_speed(iter/s)": 1.635173
    },
    {
      "acc": 0.66119003,
      "epoch": 1.700025367833587,
      "grad_norm": 5.625,
      "learning_rate": 6.02531429894605e-07,
      "loss": 1.62632561,
      "memory(GiB)": 128.51,
      "step": 67015,
      "train_speed(iter/s)": 1.635186
    },
    {
      "acc": 0.65236435,
      "epoch": 1.700152207001522,
      "grad_norm": 5.71875,
      "learning_rate": 6.020324755855478e-07,
      "loss": 1.6575634,
      "memory(GiB)": 128.51,
      "step": 67020,
      "train_speed(iter/s)": 1.635199
    },
    {
      "acc": 0.66233978,
      "epoch": 1.7002790461694572,
      "grad_norm": 7.53125,
      "learning_rate": 6.015337147187378e-07,
      "loss": 1.5628746,
      "memory(GiB)": 128.51,
      "step": 67025,
      "train_speed(iter/s)": 1.635212
    },
    {
      "acc": 0.64645081,
      "epoch": 1.7004058853373922,
      "grad_norm": 5.625,
      "learning_rate": 6.010351473161124e-07,
      "loss": 1.62819366,
      "memory(GiB)": 128.51,
      "step": 67030,
      "train_speed(iter/s)": 1.635224
    },
    {
      "acc": 0.65663452,
      "epoch": 1.7005327245053272,
      "grad_norm": 4.78125,
      "learning_rate": 6.005367733996014e-07,
      "loss": 1.59264259,
      "memory(GiB)": 128.51,
      "step": 67035,
      "train_speed(iter/s)": 1.635237
    },
    {
      "acc": 0.67016044,
      "epoch": 1.7006595636732622,
      "grad_norm": 5.625,
      "learning_rate": 6.000385929911251e-07,
      "loss": 1.55682907,
      "memory(GiB)": 128.51,
      "step": 67040,
      "train_speed(iter/s)": 1.63525
    },
    {
      "acc": 0.67255087,
      "epoch": 1.7007864028411972,
      "grad_norm": 5.375,
      "learning_rate": 5.995406061125952e-07,
      "loss": 1.5465332,
      "memory(GiB)": 128.51,
      "step": 67045,
      "train_speed(iter/s)": 1.635262
    },
    {
      "acc": 0.64551592,
      "epoch": 1.7009132420091324,
      "grad_norm": 7.21875,
      "learning_rate": 5.990428127859182e-07,
      "loss": 1.63492146,
      "memory(GiB)": 128.51,
      "step": 67050,
      "train_speed(iter/s)": 1.635276
    },
    {
      "acc": 0.65595264,
      "epoch": 1.7010400811770676,
      "grad_norm": 5.6875,
      "learning_rate": 5.985452130329855e-07,
      "loss": 1.62923775,
      "memory(GiB)": 128.51,
      "step": 67055,
      "train_speed(iter/s)": 1.635289
    },
    {
      "acc": 0.6516037,
      "epoch": 1.7011669203450026,
      "grad_norm": 6.5625,
      "learning_rate": 5.980478068756851e-07,
      "loss": 1.65989189,
      "memory(GiB)": 128.51,
      "step": 67060,
      "train_speed(iter/s)": 1.635301
    },
    {
      "acc": 0.64937429,
      "epoch": 1.7012937595129376,
      "grad_norm": 6.71875,
      "learning_rate": 5.975505943358961e-07,
      "loss": 1.65548401,
      "memory(GiB)": 128.51,
      "step": 67065,
      "train_speed(iter/s)": 1.635313
    },
    {
      "acc": 0.6575778,
      "epoch": 1.7014205986808726,
      "grad_norm": 7.1875,
      "learning_rate": 5.970535754354889e-07,
      "loss": 1.57781773,
      "memory(GiB)": 128.51,
      "step": 67070,
      "train_speed(iter/s)": 1.635324
    },
    {
      "acc": 0.64836845,
      "epoch": 1.7015474378488076,
      "grad_norm": 4.875,
      "learning_rate": 5.965567501963215e-07,
      "loss": 1.57942142,
      "memory(GiB)": 128.51,
      "step": 67075,
      "train_speed(iter/s)": 1.635337
    },
    {
      "acc": 0.66571617,
      "epoch": 1.7016742770167428,
      "grad_norm": 7.40625,
      "learning_rate": 5.960601186402476e-07,
      "loss": 1.54892979,
      "memory(GiB)": 128.51,
      "step": 67080,
      "train_speed(iter/s)": 1.635349
    },
    {
      "acc": 0.65545506,
      "epoch": 1.7018011161846778,
      "grad_norm": 5.46875,
      "learning_rate": 5.95563680789113e-07,
      "loss": 1.57882843,
      "memory(GiB)": 128.51,
      "step": 67085,
      "train_speed(iter/s)": 1.635362
    },
    {
      "acc": 0.65290594,
      "epoch": 1.701927955352613,
      "grad_norm": 6.0625,
      "learning_rate": 5.95067436664753e-07,
      "loss": 1.61973362,
      "memory(GiB)": 128.51,
      "step": 67090,
      "train_speed(iter/s)": 1.635374
    },
    {
      "acc": 0.67909002,
      "epoch": 1.702054794520548,
      "grad_norm": 6.96875,
      "learning_rate": 5.945713862889918e-07,
      "loss": 1.53332214,
      "memory(GiB)": 128.51,
      "step": 67095,
      "train_speed(iter/s)": 1.635387
    },
    {
      "acc": 0.65608969,
      "epoch": 1.702181633688483,
      "grad_norm": 5.625,
      "learning_rate": 5.940755296836504e-07,
      "loss": 1.63927383,
      "memory(GiB)": 128.51,
      "step": 67100,
      "train_speed(iter/s)": 1.635401
    },
    {
      "acc": 0.6650589,
      "epoch": 1.702308472856418,
      "grad_norm": 5.4375,
      "learning_rate": 5.935798668705389e-07,
      "loss": 1.61723461,
      "memory(GiB)": 128.51,
      "step": 67105,
      "train_speed(iter/s)": 1.635414
    },
    {
      "acc": 0.66094608,
      "epoch": 1.7024353120243532,
      "grad_norm": 5.34375,
      "learning_rate": 5.930843978714584e-07,
      "loss": 1.58662138,
      "memory(GiB)": 128.51,
      "step": 67110,
      "train_speed(iter/s)": 1.635427
    },
    {
      "acc": 0.6680903,
      "epoch": 1.7025621511922882,
      "grad_norm": 5.34375,
      "learning_rate": 5.925891227081998e-07,
      "loss": 1.5670907,
      "memory(GiB)": 128.51,
      "step": 67115,
      "train_speed(iter/s)": 1.635438
    },
    {
      "acc": 0.66787357,
      "epoch": 1.7026889903602234,
      "grad_norm": 8.0,
      "learning_rate": 5.920940414025506e-07,
      "loss": 1.56318722,
      "memory(GiB)": 128.51,
      "step": 67120,
      "train_speed(iter/s)": 1.635452
    },
    {
      "acc": 0.66724529,
      "epoch": 1.7028158295281584,
      "grad_norm": 6.375,
      "learning_rate": 5.91599153976285e-07,
      "loss": 1.51742411,
      "memory(GiB)": 128.51,
      "step": 67125,
      "train_speed(iter/s)": 1.635464
    },
    {
      "acc": 0.65388689,
      "epoch": 1.7029426686960933,
      "grad_norm": 4.9375,
      "learning_rate": 5.911044604511712e-07,
      "loss": 1.62871475,
      "memory(GiB)": 128.51,
      "step": 67130,
      "train_speed(iter/s)": 1.635478
    },
    {
      "acc": 0.65295963,
      "epoch": 1.7030695078640283,
      "grad_norm": 5.28125,
      "learning_rate": 5.906099608489669e-07,
      "loss": 1.64161377,
      "memory(GiB)": 128.51,
      "step": 67135,
      "train_speed(iter/s)": 1.635489
    },
    {
      "acc": 0.65288839,
      "epoch": 1.7031963470319633,
      "grad_norm": 4.9375,
      "learning_rate": 5.901156551914233e-07,
      "loss": 1.59084301,
      "memory(GiB)": 128.51,
      "step": 67140,
      "train_speed(iter/s)": 1.635501
    },
    {
      "acc": 0.64179258,
      "epoch": 1.7033231861998985,
      "grad_norm": 6.0625,
      "learning_rate": 5.896215435002816e-07,
      "loss": 1.63305626,
      "memory(GiB)": 128.51,
      "step": 67145,
      "train_speed(iter/s)": 1.635509
    },
    {
      "acc": 0.65226526,
      "epoch": 1.7034500253678337,
      "grad_norm": 5.1875,
      "learning_rate": 5.891276257972762e-07,
      "loss": 1.57875299,
      "memory(GiB)": 128.51,
      "step": 67150,
      "train_speed(iter/s)": 1.635521
    },
    {
      "acc": 0.65882301,
      "epoch": 1.7035768645357687,
      "grad_norm": 5.15625,
      "learning_rate": 5.886339021041304e-07,
      "loss": 1.57733374,
      "memory(GiB)": 128.51,
      "step": 67155,
      "train_speed(iter/s)": 1.635533
    },
    {
      "acc": 0.65831118,
      "epoch": 1.7037037037037037,
      "grad_norm": 5.46875,
      "learning_rate": 5.881403724425605e-07,
      "loss": 1.64175663,
      "memory(GiB)": 128.51,
      "step": 67160,
      "train_speed(iter/s)": 1.635547
    },
    {
      "acc": 0.65446987,
      "epoch": 1.7038305428716387,
      "grad_norm": 6.21875,
      "learning_rate": 5.876470368342752e-07,
      "loss": 1.57208385,
      "memory(GiB)": 128.51,
      "step": 67165,
      "train_speed(iter/s)": 1.635559
    },
    {
      "acc": 0.66269779,
      "epoch": 1.7039573820395737,
      "grad_norm": 4.96875,
      "learning_rate": 5.871538953009725e-07,
      "loss": 1.51582699,
      "memory(GiB)": 128.51,
      "step": 67170,
      "train_speed(iter/s)": 1.635572
    },
    {
      "acc": 0.66614838,
      "epoch": 1.704084221207509,
      "grad_norm": 5.53125,
      "learning_rate": 5.866609478643437e-07,
      "loss": 1.53599463,
      "memory(GiB)": 128.51,
      "step": 67175,
      "train_speed(iter/s)": 1.635583
    },
    {
      "acc": 0.64267693,
      "epoch": 1.704211060375444,
      "grad_norm": 5.78125,
      "learning_rate": 5.861681945460706e-07,
      "loss": 1.64508476,
      "memory(GiB)": 128.51,
      "step": 67180,
      "train_speed(iter/s)": 1.635596
    },
    {
      "acc": 0.63538084,
      "epoch": 1.704337899543379,
      "grad_norm": 6.46875,
      "learning_rate": 5.856756353678255e-07,
      "loss": 1.65825348,
      "memory(GiB)": 128.51,
      "step": 67185,
      "train_speed(iter/s)": 1.635608
    },
    {
      "acc": 0.65387354,
      "epoch": 1.704464738711314,
      "grad_norm": 5.34375,
      "learning_rate": 5.851832703512766e-07,
      "loss": 1.57906122,
      "memory(GiB)": 128.51,
      "step": 67190,
      "train_speed(iter/s)": 1.635621
    },
    {
      "acc": 0.65083556,
      "epoch": 1.704591577879249,
      "grad_norm": 5.125,
      "learning_rate": 5.846910995180771e-07,
      "loss": 1.56133184,
      "memory(GiB)": 128.51,
      "step": 67195,
      "train_speed(iter/s)": 1.635634
    },
    {
      "acc": 0.64118786,
      "epoch": 1.704718417047184,
      "grad_norm": 4.71875,
      "learning_rate": 5.841991228898758e-07,
      "loss": 1.60160198,
      "memory(GiB)": 128.51,
      "step": 67200,
      "train_speed(iter/s)": 1.635648
    },
    {
      "acc": 0.67618542,
      "epoch": 1.704845256215119,
      "grad_norm": 6.40625,
      "learning_rate": 5.837073404883115e-07,
      "loss": 1.54821815,
      "memory(GiB)": 128.51,
      "step": 67205,
      "train_speed(iter/s)": 1.635661
    },
    {
      "acc": 0.67320957,
      "epoch": 1.7049720953830543,
      "grad_norm": 5.875,
      "learning_rate": 5.832157523350179e-07,
      "loss": 1.5800415,
      "memory(GiB)": 128.51,
      "step": 67210,
      "train_speed(iter/s)": 1.635674
    },
    {
      "acc": 0.64668779,
      "epoch": 1.7050989345509895,
      "grad_norm": 5.9375,
      "learning_rate": 5.827243584516135e-07,
      "loss": 1.61795712,
      "memory(GiB)": 128.51,
      "step": 67215,
      "train_speed(iter/s)": 1.635688
    },
    {
      "acc": 0.65144224,
      "epoch": 1.7052257737189245,
      "grad_norm": 5.40625,
      "learning_rate": 5.822331588597129e-07,
      "loss": 1.62538376,
      "memory(GiB)": 128.51,
      "step": 67220,
      "train_speed(iter/s)": 1.635701
    },
    {
      "acc": 0.66135716,
      "epoch": 1.7053526128868595,
      "grad_norm": 6.59375,
      "learning_rate": 5.817421535809226e-07,
      "loss": 1.58652143,
      "memory(GiB)": 128.51,
      "step": 67225,
      "train_speed(iter/s)": 1.635714
    },
    {
      "acc": 0.6567132,
      "epoch": 1.7054794520547945,
      "grad_norm": 5.71875,
      "learning_rate": 5.812513426368399e-07,
      "loss": 1.59401484,
      "memory(GiB)": 128.51,
      "step": 67230,
      "train_speed(iter/s)": 1.635727
    },
    {
      "acc": 0.66141739,
      "epoch": 1.7056062912227294,
      "grad_norm": 5.3125,
      "learning_rate": 5.807607260490489e-07,
      "loss": 1.60627651,
      "memory(GiB)": 128.51,
      "step": 67235,
      "train_speed(iter/s)": 1.635739
    },
    {
      "acc": 0.66541567,
      "epoch": 1.7057331303906647,
      "grad_norm": 5.375,
      "learning_rate": 5.802703038391333e-07,
      "loss": 1.54779749,
      "memory(GiB)": 128.51,
      "step": 67240,
      "train_speed(iter/s)": 1.635753
    },
    {
      "acc": 0.64876361,
      "epoch": 1.7058599695585996,
      "grad_norm": 5.40625,
      "learning_rate": 5.797800760286621e-07,
      "loss": 1.61090012,
      "memory(GiB)": 128.51,
      "step": 67245,
      "train_speed(iter/s)": 1.635766
    },
    {
      "acc": 0.65214796,
      "epoch": 1.7059868087265349,
      "grad_norm": 6.34375,
      "learning_rate": 5.79290042639199e-07,
      "loss": 1.57541056,
      "memory(GiB)": 128.51,
      "step": 67250,
      "train_speed(iter/s)": 1.635778
    },
    {
      "acc": 0.67495232,
      "epoch": 1.7061136478944698,
      "grad_norm": 5.46875,
      "learning_rate": 5.788002036922946e-07,
      "loss": 1.56865396,
      "memory(GiB)": 128.51,
      "step": 67255,
      "train_speed(iter/s)": 1.635792
    },
    {
      "acc": 0.65226698,
      "epoch": 1.7062404870624048,
      "grad_norm": 6.34375,
      "learning_rate": 5.783105592094978e-07,
      "loss": 1.60335197,
      "memory(GiB)": 128.51,
      "step": 67260,
      "train_speed(iter/s)": 1.635806
    },
    {
      "acc": 0.66577034,
      "epoch": 1.7063673262303398,
      "grad_norm": 5.84375,
      "learning_rate": 5.778211092123437e-07,
      "loss": 1.6350111,
      "memory(GiB)": 128.51,
      "step": 67265,
      "train_speed(iter/s)": 1.635819
    },
    {
      "acc": 0.66672201,
      "epoch": 1.706494165398275,
      "grad_norm": 6.59375,
      "learning_rate": 5.77331853722361e-07,
      "loss": 1.61463795,
      "memory(GiB)": 128.51,
      "step": 67270,
      "train_speed(iter/s)": 1.635833
    },
    {
      "acc": 0.65614357,
      "epoch": 1.70662100456621,
      "grad_norm": 5.28125,
      "learning_rate": 5.768427927610698e-07,
      "loss": 1.68490486,
      "memory(GiB)": 128.51,
      "step": 67275,
      "train_speed(iter/s)": 1.635845
    },
    {
      "acc": 0.66344318,
      "epoch": 1.7067478437341452,
      "grad_norm": 8.8125,
      "learning_rate": 5.763539263499796e-07,
      "loss": 1.53171568,
      "memory(GiB)": 128.51,
      "step": 67280,
      "train_speed(iter/s)": 1.635858
    },
    {
      "acc": 0.66403871,
      "epoch": 1.7068746829020802,
      "grad_norm": 5.28125,
      "learning_rate": 5.758652545105941e-07,
      "loss": 1.55713673,
      "memory(GiB)": 128.51,
      "step": 67285,
      "train_speed(iter/s)": 1.63587
    },
    {
      "acc": 0.64168429,
      "epoch": 1.7070015220700152,
      "grad_norm": 6.0,
      "learning_rate": 5.753767772644075e-07,
      "loss": 1.62596092,
      "memory(GiB)": 128.51,
      "step": 67290,
      "train_speed(iter/s)": 1.635882
    },
    {
      "acc": 0.66364594,
      "epoch": 1.7071283612379502,
      "grad_norm": 6.34375,
      "learning_rate": 5.748884946329048e-07,
      "loss": 1.54136763,
      "memory(GiB)": 128.51,
      "step": 67295,
      "train_speed(iter/s)": 1.635895
    },
    {
      "acc": 0.66366282,
      "epoch": 1.7072552004058852,
      "grad_norm": 6.4375,
      "learning_rate": 5.744004066375625e-07,
      "loss": 1.59002447,
      "memory(GiB)": 128.51,
      "step": 67300,
      "train_speed(iter/s)": 1.635907
    },
    {
      "acc": 0.68015742,
      "epoch": 1.7073820395738204,
      "grad_norm": 5.15625,
      "learning_rate": 5.739125132998491e-07,
      "loss": 1.54480391,
      "memory(GiB)": 128.51,
      "step": 67305,
      "train_speed(iter/s)": 1.63592
    },
    {
      "acc": 0.65216713,
      "epoch": 1.7075088787417556,
      "grad_norm": 5.78125,
      "learning_rate": 5.734248146412242e-07,
      "loss": 1.61866341,
      "memory(GiB)": 128.51,
      "step": 67310,
      "train_speed(iter/s)": 1.635931
    },
    {
      "acc": 0.64957247,
      "epoch": 1.7076357179096906,
      "grad_norm": 7.0625,
      "learning_rate": 5.729373106831399e-07,
      "loss": 1.63667259,
      "memory(GiB)": 128.51,
      "step": 67315,
      "train_speed(iter/s)": 1.635943
    },
    {
      "acc": 0.6638835,
      "epoch": 1.7077625570776256,
      "grad_norm": 11.25,
      "learning_rate": 5.724500014470374e-07,
      "loss": 1.6611311,
      "memory(GiB)": 128.51,
      "step": 67320,
      "train_speed(iter/s)": 1.635955
    },
    {
      "acc": 0.64782643,
      "epoch": 1.7078893962455606,
      "grad_norm": 5.6875,
      "learning_rate": 5.719628869543509e-07,
      "loss": 1.61336021,
      "memory(GiB)": 128.51,
      "step": 67325,
      "train_speed(iter/s)": 1.635968
    },
    {
      "acc": 0.6541482,
      "epoch": 1.7080162354134956,
      "grad_norm": 6.3125,
      "learning_rate": 5.71475967226508e-07,
      "loss": 1.5971468,
      "memory(GiB)": 128.51,
      "step": 67330,
      "train_speed(iter/s)": 1.635979
    },
    {
      "acc": 0.67253056,
      "epoch": 1.7081430745814308,
      "grad_norm": 7.0,
      "learning_rate": 5.709892422849233e-07,
      "loss": 1.54063168,
      "memory(GiB)": 128.51,
      "step": 67335,
      "train_speed(iter/s)": 1.635992
    },
    {
      "acc": 0.64050703,
      "epoch": 1.7082699137493658,
      "grad_norm": 6.96875,
      "learning_rate": 5.705027121510059e-07,
      "loss": 1.65965195,
      "memory(GiB)": 128.51,
      "step": 67340,
      "train_speed(iter/s)": 1.636003
    },
    {
      "acc": 0.65276108,
      "epoch": 1.708396752917301,
      "grad_norm": 6.25,
      "learning_rate": 5.700163768461542e-07,
      "loss": 1.57592535,
      "memory(GiB)": 128.51,
      "step": 67345,
      "train_speed(iter/s)": 1.636016
    },
    {
      "acc": 0.63521709,
      "epoch": 1.708523592085236,
      "grad_norm": 5.65625,
      "learning_rate": 5.695302363917626e-07,
      "loss": 1.6635664,
      "memory(GiB)": 128.51,
      "step": 67350,
      "train_speed(iter/s)": 1.636029
    },
    {
      "acc": 0.66569362,
      "epoch": 1.708650431253171,
      "grad_norm": 7.875,
      "learning_rate": 5.690442908092115e-07,
      "loss": 1.55869045,
      "memory(GiB)": 128.51,
      "step": 67355,
      "train_speed(iter/s)": 1.636041
    },
    {
      "acc": 0.64959393,
      "epoch": 1.708777270421106,
      "grad_norm": 8.3125,
      "learning_rate": 5.685585401198735e-07,
      "loss": 1.62649384,
      "memory(GiB)": 128.51,
      "step": 67360,
      "train_speed(iter/s)": 1.636052
    },
    {
      "acc": 0.65645819,
      "epoch": 1.708904109589041,
      "grad_norm": 7.90625,
      "learning_rate": 5.680729843451172e-07,
      "loss": 1.59811745,
      "memory(GiB)": 128.51,
      "step": 67365,
      "train_speed(iter/s)": 1.636064
    },
    {
      "acc": 0.63494329,
      "epoch": 1.7090309487569761,
      "grad_norm": 4.65625,
      "learning_rate": 5.675876235062994e-07,
      "loss": 1.69742622,
      "memory(GiB)": 128.51,
      "step": 67370,
      "train_speed(iter/s)": 1.636076
    },
    {
      "acc": 0.65067978,
      "epoch": 1.7091577879249114,
      "grad_norm": 5.84375,
      "learning_rate": 5.671024576247657e-07,
      "loss": 1.56292629,
      "memory(GiB)": 128.51,
      "step": 67375,
      "train_speed(iter/s)": 1.636088
    },
    {
      "acc": 0.66673517,
      "epoch": 1.7092846270928463,
      "grad_norm": 4.5625,
      "learning_rate": 5.666174867218572e-07,
      "loss": 1.54882193,
      "memory(GiB)": 128.51,
      "step": 67380,
      "train_speed(iter/s)": 1.636099
    },
    {
      "acc": 0.66848035,
      "epoch": 1.7094114662607813,
      "grad_norm": 6.21875,
      "learning_rate": 5.661327108189058e-07,
      "loss": 1.52031937,
      "memory(GiB)": 128.51,
      "step": 67385,
      "train_speed(iter/s)": 1.636111
    },
    {
      "acc": 0.66235709,
      "epoch": 1.7095383054287163,
      "grad_norm": 5.65625,
      "learning_rate": 5.65648129937234e-07,
      "loss": 1.54351664,
      "memory(GiB)": 128.51,
      "step": 67390,
      "train_speed(iter/s)": 1.636124
    },
    {
      "acc": 0.65491085,
      "epoch": 1.7096651445966513,
      "grad_norm": 6.40625,
      "learning_rate": 5.651637440981545e-07,
      "loss": 1.61414585,
      "memory(GiB)": 128.51,
      "step": 67395,
      "train_speed(iter/s)": 1.636137
    },
    {
      "acc": 0.65579281,
      "epoch": 1.7097919837645865,
      "grad_norm": 6.1875,
      "learning_rate": 5.646795533229738e-07,
      "loss": 1.65341644,
      "memory(GiB)": 128.51,
      "step": 67400,
      "train_speed(iter/s)": 1.636149
    },
    {
      "acc": 0.65887375,
      "epoch": 1.7099188229325215,
      "grad_norm": 6.625,
      "learning_rate": 5.641955576329888e-07,
      "loss": 1.597859,
      "memory(GiB)": 128.51,
      "step": 67405,
      "train_speed(iter/s)": 1.636163
    },
    {
      "acc": 0.67939582,
      "epoch": 1.7100456621004567,
      "grad_norm": 5.1875,
      "learning_rate": 5.637117570494877e-07,
      "loss": 1.52714405,
      "memory(GiB)": 128.51,
      "step": 67410,
      "train_speed(iter/s)": 1.636175
    },
    {
      "acc": 0.64111986,
      "epoch": 1.7101725012683917,
      "grad_norm": 4.96875,
      "learning_rate": 5.632281515937504e-07,
      "loss": 1.63542213,
      "memory(GiB)": 128.51,
      "step": 67415,
      "train_speed(iter/s)": 1.636188
    },
    {
      "acc": 0.65833836,
      "epoch": 1.7102993404363267,
      "grad_norm": 5.125,
      "learning_rate": 5.627447412870474e-07,
      "loss": 1.57399855,
      "memory(GiB)": 128.51,
      "step": 67420,
      "train_speed(iter/s)": 1.6362
    },
    {
      "acc": 0.65467939,
      "epoch": 1.7104261796042617,
      "grad_norm": 6.09375,
      "learning_rate": 5.62261526150642e-07,
      "loss": 1.62727623,
      "memory(GiB)": 128.51,
      "step": 67425,
      "train_speed(iter/s)": 1.636211
    },
    {
      "acc": 0.6677947,
      "epoch": 1.710553018772197,
      "grad_norm": 6.46875,
      "learning_rate": 5.617785062057873e-07,
      "loss": 1.59308891,
      "memory(GiB)": 128.51,
      "step": 67430,
      "train_speed(iter/s)": 1.636223
    },
    {
      "acc": 0.64166551,
      "epoch": 1.7106798579401319,
      "grad_norm": 5.28125,
      "learning_rate": 5.612956814737291e-07,
      "loss": 1.59966021,
      "memory(GiB)": 128.51,
      "step": 67435,
      "train_speed(iter/s)": 1.636236
    },
    {
      "acc": 0.6683826,
      "epoch": 1.710806697108067,
      "grad_norm": 6.65625,
      "learning_rate": 5.608130519757044e-07,
      "loss": 1.50156555,
      "memory(GiB)": 128.51,
      "step": 67440,
      "train_speed(iter/s)": 1.636249
    },
    {
      "acc": 0.66201143,
      "epoch": 1.710933536276002,
      "grad_norm": 5.96875,
      "learning_rate": 5.603306177329415e-07,
      "loss": 1.56475639,
      "memory(GiB)": 128.51,
      "step": 67445,
      "train_speed(iter/s)": 1.636264
    },
    {
      "acc": 0.65424442,
      "epoch": 1.711060375443937,
      "grad_norm": 5.125,
      "learning_rate": 5.598483787666592e-07,
      "loss": 1.55552483,
      "memory(GiB)": 128.51,
      "step": 67450,
      "train_speed(iter/s)": 1.636276
    },
    {
      "acc": 0.65675545,
      "epoch": 1.711187214611872,
      "grad_norm": 6.5,
      "learning_rate": 5.593663350980693e-07,
      "loss": 1.65335522,
      "memory(GiB)": 128.51,
      "step": 67455,
      "train_speed(iter/s)": 1.636289
    },
    {
      "acc": 0.67183008,
      "epoch": 1.711314053779807,
      "grad_norm": 6.28125,
      "learning_rate": 5.588844867483739e-07,
      "loss": 1.56383057,
      "memory(GiB)": 128.51,
      "step": 67460,
      "train_speed(iter/s)": 1.636301
    },
    {
      "acc": 0.66191173,
      "epoch": 1.7114408929477423,
      "grad_norm": 5.6875,
      "learning_rate": 5.58402833738767e-07,
      "loss": 1.51929502,
      "memory(GiB)": 128.51,
      "step": 67465,
      "train_speed(iter/s)": 1.636315
    },
    {
      "acc": 0.63522797,
      "epoch": 1.7115677321156775,
      "grad_norm": 4.59375,
      "learning_rate": 5.579213760904339e-07,
      "loss": 1.6719841,
      "memory(GiB)": 128.51,
      "step": 67470,
      "train_speed(iter/s)": 1.636328
    },
    {
      "acc": 0.65678344,
      "epoch": 1.7116945712836125,
      "grad_norm": 6.125,
      "learning_rate": 5.574401138245511e-07,
      "loss": 1.57955208,
      "memory(GiB)": 128.51,
      "step": 67475,
      "train_speed(iter/s)": 1.636341
    },
    {
      "acc": 0.65569644,
      "epoch": 1.7118214104515475,
      "grad_norm": 5.46875,
      "learning_rate": 5.569590469622865e-07,
      "loss": 1.55289574,
      "memory(GiB)": 128.51,
      "step": 67480,
      "train_speed(iter/s)": 1.636354
    },
    {
      "acc": 0.68059773,
      "epoch": 1.7119482496194824,
      "grad_norm": 5.625,
      "learning_rate": 5.564781755247989e-07,
      "loss": 1.52871294,
      "memory(GiB)": 128.51,
      "step": 67485,
      "train_speed(iter/s)": 1.636365
    },
    {
      "acc": 0.65363941,
      "epoch": 1.7120750887874174,
      "grad_norm": 6.9375,
      "learning_rate": 5.559974995332424e-07,
      "loss": 1.60149441,
      "memory(GiB)": 128.51,
      "step": 67490,
      "train_speed(iter/s)": 1.636378
    },
    {
      "acc": 0.6431273,
      "epoch": 1.7122019279553526,
      "grad_norm": 5.25,
      "learning_rate": 5.555170190087556e-07,
      "loss": 1.58752108,
      "memory(GiB)": 128.51,
      "step": 67495,
      "train_speed(iter/s)": 1.636391
    },
    {
      "acc": 0.64971752,
      "epoch": 1.7123287671232876,
      "grad_norm": 6.34375,
      "learning_rate": 5.550367339724721e-07,
      "loss": 1.62518997,
      "memory(GiB)": 128.51,
      "step": 67500,
      "train_speed(iter/s)": 1.636404
    },
    {
      "acc": 0.66027694,
      "epoch": 1.7124556062912228,
      "grad_norm": 5.875,
      "learning_rate": 5.545566444455198e-07,
      "loss": 1.58222637,
      "memory(GiB)": 128.51,
      "step": 67505,
      "train_speed(iter/s)": 1.636417
    },
    {
      "acc": 0.66492968,
      "epoch": 1.7125824454591578,
      "grad_norm": 5.96875,
      "learning_rate": 5.540767504490147e-07,
      "loss": 1.64748383,
      "memory(GiB)": 128.51,
      "step": 67510,
      "train_speed(iter/s)": 1.63643
    },
    {
      "acc": 0.66160536,
      "epoch": 1.7127092846270928,
      "grad_norm": 5.21875,
      "learning_rate": 5.535970520040629e-07,
      "loss": 1.52424345,
      "memory(GiB)": 128.51,
      "step": 67515,
      "train_speed(iter/s)": 1.636443
    },
    {
      "acc": 0.66720095,
      "epoch": 1.7128361237950278,
      "grad_norm": 4.65625,
      "learning_rate": 5.53117549131763e-07,
      "loss": 1.54493313,
      "memory(GiB)": 128.51,
      "step": 67520,
      "train_speed(iter/s)": 1.636456
    },
    {
      "acc": 0.66251659,
      "epoch": 1.7129629629629628,
      "grad_norm": 6.1875,
      "learning_rate": 5.526382418532089e-07,
      "loss": 1.56596413,
      "memory(GiB)": 128.51,
      "step": 67525,
      "train_speed(iter/s)": 1.636469
    },
    {
      "acc": 0.66081324,
      "epoch": 1.713089802130898,
      "grad_norm": 7.59375,
      "learning_rate": 5.521591301894813e-07,
      "loss": 1.58382988,
      "memory(GiB)": 128.51,
      "step": 67530,
      "train_speed(iter/s)": 1.636482
    },
    {
      "acc": 0.66171446,
      "epoch": 1.7132166412988332,
      "grad_norm": 5.15625,
      "learning_rate": 5.516802141616517e-07,
      "loss": 1.59539871,
      "memory(GiB)": 128.51,
      "step": 67535,
      "train_speed(iter/s)": 1.636496
    },
    {
      "acc": 0.65765657,
      "epoch": 1.7133434804667682,
      "grad_norm": 4.875,
      "learning_rate": 5.512014937907872e-07,
      "loss": 1.594205,
      "memory(GiB)": 128.51,
      "step": 67540,
      "train_speed(iter/s)": 1.636508
    },
    {
      "acc": 0.65852518,
      "epoch": 1.7134703196347032,
      "grad_norm": 5.53125,
      "learning_rate": 5.507229690979438e-07,
      "loss": 1.53196783,
      "memory(GiB)": 128.51,
      "step": 67545,
      "train_speed(iter/s)": 1.63652
    },
    {
      "acc": 0.64753685,
      "epoch": 1.7135971588026382,
      "grad_norm": 6.3125,
      "learning_rate": 5.5024464010417e-07,
      "loss": 1.63069267,
      "memory(GiB)": 128.51,
      "step": 67550,
      "train_speed(iter/s)": 1.636534
    },
    {
      "acc": 0.67058296,
      "epoch": 1.7137239979705732,
      "grad_norm": 5.9375,
      "learning_rate": 5.497665068305014e-07,
      "loss": 1.55970478,
      "memory(GiB)": 128.51,
      "step": 67555,
      "train_speed(iter/s)": 1.636547
    },
    {
      "acc": 0.64686761,
      "epoch": 1.7138508371385084,
      "grad_norm": 6.1875,
      "learning_rate": 5.492885692979716e-07,
      "loss": 1.63346825,
      "memory(GiB)": 128.51,
      "step": 67560,
      "train_speed(iter/s)": 1.636559
    },
    {
      "acc": 0.65042729,
      "epoch": 1.7139776763064434,
      "grad_norm": 6.09375,
      "learning_rate": 5.488108275276016e-07,
      "loss": 1.55949078,
      "memory(GiB)": 128.51,
      "step": 67565,
      "train_speed(iter/s)": 1.636572
    },
    {
      "acc": 0.66530113,
      "epoch": 1.7141045154743786,
      "grad_norm": 6.125,
      "learning_rate": 5.483332815404046e-07,
      "loss": 1.57052898,
      "memory(GiB)": 128.51,
      "step": 67570,
      "train_speed(iter/s)": 1.636585
    },
    {
      "acc": 0.65633655,
      "epoch": 1.7142313546423136,
      "grad_norm": 5.5625,
      "learning_rate": 5.478559313573855e-07,
      "loss": 1.61563721,
      "memory(GiB)": 128.51,
      "step": 67575,
      "train_speed(iter/s)": 1.636597
    },
    {
      "acc": 0.66037369,
      "epoch": 1.7143581938102486,
      "grad_norm": 6.0625,
      "learning_rate": 5.473787769995392e-07,
      "loss": 1.59547806,
      "memory(GiB)": 128.51,
      "step": 67580,
      "train_speed(iter/s)": 1.63661
    },
    {
      "acc": 0.63931317,
      "epoch": 1.7144850329781836,
      "grad_norm": 5.96875,
      "learning_rate": 5.469018184878544e-07,
      "loss": 1.63119316,
      "memory(GiB)": 128.51,
      "step": 67585,
      "train_speed(iter/s)": 1.636622
    },
    {
      "acc": 0.6508801,
      "epoch": 1.7146118721461188,
      "grad_norm": 5.1875,
      "learning_rate": 5.464250558433088e-07,
      "loss": 1.58278198,
      "memory(GiB)": 128.51,
      "step": 67590,
      "train_speed(iter/s)": 1.636635
    },
    {
      "acc": 0.64154372,
      "epoch": 1.7147387113140538,
      "grad_norm": 5.21875,
      "learning_rate": 5.459484890868732e-07,
      "loss": 1.66356468,
      "memory(GiB)": 128.51,
      "step": 67595,
      "train_speed(iter/s)": 1.636648
    },
    {
      "acc": 0.65137215,
      "epoch": 1.714865550481989,
      "grad_norm": 6.65625,
      "learning_rate": 5.454721182395096e-07,
      "loss": 1.64805431,
      "memory(GiB)": 128.51,
      "step": 67600,
      "train_speed(iter/s)": 1.63666
    },
    {
      "acc": 0.66730328,
      "epoch": 1.714992389649924,
      "grad_norm": 5.625,
      "learning_rate": 5.449959433221697e-07,
      "loss": 1.5982955,
      "memory(GiB)": 128.51,
      "step": 67605,
      "train_speed(iter/s)": 1.636672
    },
    {
      "acc": 0.65276961,
      "epoch": 1.715119228817859,
      "grad_norm": 5.28125,
      "learning_rate": 5.44519964355798e-07,
      "loss": 1.64188347,
      "memory(GiB)": 128.51,
      "step": 67610,
      "train_speed(iter/s)": 1.636684
    },
    {
      "acc": 0.64667115,
      "epoch": 1.715246067985794,
      "grad_norm": 6.3125,
      "learning_rate": 5.440441813613312e-07,
      "loss": 1.6179575,
      "memory(GiB)": 128.51,
      "step": 67615,
      "train_speed(iter/s)": 1.636698
    },
    {
      "acc": 0.65816946,
      "epoch": 1.715372907153729,
      "grad_norm": 5.4375,
      "learning_rate": 5.435685943596953e-07,
      "loss": 1.62623596,
      "memory(GiB)": 128.51,
      "step": 67620,
      "train_speed(iter/s)": 1.636709
    },
    {
      "acc": 0.66269526,
      "epoch": 1.7154997463216641,
      "grad_norm": 5.5,
      "learning_rate": 5.430932033718083e-07,
      "loss": 1.54525394,
      "memory(GiB)": 128.51,
      "step": 67625,
      "train_speed(iter/s)": 1.636721
    },
    {
      "acc": 0.66235828,
      "epoch": 1.7156265854895993,
      "grad_norm": 5.21875,
      "learning_rate": 5.426180084185828e-07,
      "loss": 1.60276299,
      "memory(GiB)": 128.51,
      "step": 67630,
      "train_speed(iter/s)": 1.636733
    },
    {
      "acc": 0.66971216,
      "epoch": 1.7157534246575343,
      "grad_norm": 7.125,
      "learning_rate": 5.421430095209173e-07,
      "loss": 1.60329056,
      "memory(GiB)": 128.51,
      "step": 67635,
      "train_speed(iter/s)": 1.636746
    },
    {
      "acc": 0.64574385,
      "epoch": 1.7158802638254693,
      "grad_norm": 5.46875,
      "learning_rate": 5.416682066997048e-07,
      "loss": 1.62317638,
      "memory(GiB)": 128.51,
      "step": 67640,
      "train_speed(iter/s)": 1.636758
    },
    {
      "acc": 0.66928387,
      "epoch": 1.7160071029934043,
      "grad_norm": 5.375,
      "learning_rate": 5.411935999758288e-07,
      "loss": 1.60135002,
      "memory(GiB)": 128.51,
      "step": 67645,
      "train_speed(iter/s)": 1.63677
    },
    {
      "acc": 0.65422292,
      "epoch": 1.7161339421613393,
      "grad_norm": 6.9375,
      "learning_rate": 5.407191893701674e-07,
      "loss": 1.57047758,
      "memory(GiB)": 128.51,
      "step": 67650,
      "train_speed(iter/s)": 1.636783
    },
    {
      "acc": 0.65626907,
      "epoch": 1.7162607813292745,
      "grad_norm": 5.90625,
      "learning_rate": 5.402449749035843e-07,
      "loss": 1.59983549,
      "memory(GiB)": 128.51,
      "step": 67655,
      "train_speed(iter/s)": 1.636794
    },
    {
      "acc": 0.65422726,
      "epoch": 1.7163876204972095,
      "grad_norm": 6.40625,
      "learning_rate": 5.397709565969378e-07,
      "loss": 1.56725245,
      "memory(GiB)": 128.51,
      "step": 67660,
      "train_speed(iter/s)": 1.636806
    },
    {
      "acc": 0.65677891,
      "epoch": 1.7165144596651447,
      "grad_norm": 6.8125,
      "learning_rate": 5.392971344710785e-07,
      "loss": 1.64886169,
      "memory(GiB)": 128.51,
      "step": 67665,
      "train_speed(iter/s)": 1.636818
    },
    {
      "acc": 0.66282663,
      "epoch": 1.7166412988330797,
      "grad_norm": 6.1875,
      "learning_rate": 5.388235085468485e-07,
      "loss": 1.62454166,
      "memory(GiB)": 128.51,
      "step": 67670,
      "train_speed(iter/s)": 1.63683
    },
    {
      "acc": 0.66598711,
      "epoch": 1.7167681380010147,
      "grad_norm": 5.96875,
      "learning_rate": 5.383500788450757e-07,
      "loss": 1.5794363,
      "memory(GiB)": 128.51,
      "step": 67675,
      "train_speed(iter/s)": 1.636843
    },
    {
      "acc": 0.65941963,
      "epoch": 1.7168949771689497,
      "grad_norm": 7.0,
      "learning_rate": 5.378768453865879e-07,
      "loss": 1.64889221,
      "memory(GiB)": 128.51,
      "step": 67680,
      "train_speed(iter/s)": 1.636855
    },
    {
      "acc": 0.65301218,
      "epoch": 1.7170218163368847,
      "grad_norm": 6.8125,
      "learning_rate": 5.374038081921978e-07,
      "loss": 1.61743774,
      "memory(GiB)": 128.51,
      "step": 67685,
      "train_speed(iter/s)": 1.636866
    },
    {
      "acc": 0.65661535,
      "epoch": 1.7171486555048199,
      "grad_norm": 6.59375,
      "learning_rate": 5.369309672827139e-07,
      "loss": 1.62625866,
      "memory(GiB)": 128.51,
      "step": 67690,
      "train_speed(iter/s)": 1.636878
    },
    {
      "acc": 0.66227503,
      "epoch": 1.717275494672755,
      "grad_norm": 6.9375,
      "learning_rate": 5.364583226789299e-07,
      "loss": 1.57511816,
      "memory(GiB)": 128.51,
      "step": 67695,
      "train_speed(iter/s)": 1.636889
    },
    {
      "acc": 0.66967993,
      "epoch": 1.71740233384069,
      "grad_norm": 5.15625,
      "learning_rate": 5.359858744016378e-07,
      "loss": 1.59977551,
      "memory(GiB)": 128.51,
      "step": 67700,
      "train_speed(iter/s)": 1.636901
    },
    {
      "acc": 0.64543953,
      "epoch": 1.717529173008625,
      "grad_norm": 4.96875,
      "learning_rate": 5.355136224716179e-07,
      "loss": 1.58033867,
      "memory(GiB)": 128.51,
      "step": 67705,
      "train_speed(iter/s)": 1.636913
    },
    {
      "acc": 0.6615799,
      "epoch": 1.71765601217656,
      "grad_norm": 5.25,
      "learning_rate": 5.350415669096409e-07,
      "loss": 1.54568233,
      "memory(GiB)": 128.51,
      "step": 67710,
      "train_speed(iter/s)": 1.636925
    },
    {
      "acc": 0.66160154,
      "epoch": 1.717782851344495,
      "grad_norm": 7.1875,
      "learning_rate": 5.345697077364708e-07,
      "loss": 1.61753101,
      "memory(GiB)": 128.51,
      "step": 67715,
      "train_speed(iter/s)": 1.636938
    },
    {
      "acc": 0.63484812,
      "epoch": 1.7179096905124303,
      "grad_norm": 5.0,
      "learning_rate": 5.340980449728617e-07,
      "loss": 1.67677536,
      "memory(GiB)": 128.51,
      "step": 67720,
      "train_speed(iter/s)": 1.63695
    },
    {
      "acc": 0.66475959,
      "epoch": 1.7180365296803652,
      "grad_norm": 5.625,
      "learning_rate": 5.336265786395589e-07,
      "loss": 1.50078735,
      "memory(GiB)": 128.51,
      "step": 67725,
      "train_speed(iter/s)": 1.636962
    },
    {
      "acc": 0.66264863,
      "epoch": 1.7181633688483005,
      "grad_norm": 6.78125,
      "learning_rate": 5.331553087573005e-07,
      "loss": 1.58849535,
      "memory(GiB)": 128.51,
      "step": 67730,
      "train_speed(iter/s)": 1.636974
    },
    {
      "acc": 0.6571393,
      "epoch": 1.7182902080162354,
      "grad_norm": 5.03125,
      "learning_rate": 5.326842353468148e-07,
      "loss": 1.64206944,
      "memory(GiB)": 128.51,
      "step": 67735,
      "train_speed(iter/s)": 1.636986
    },
    {
      "acc": 0.66942153,
      "epoch": 1.7184170471841704,
      "grad_norm": 7.3125,
      "learning_rate": 5.322133584288214e-07,
      "loss": 1.55967379,
      "memory(GiB)": 128.51,
      "step": 67740,
      "train_speed(iter/s)": 1.636999
    },
    {
      "acc": 0.65423899,
      "epoch": 1.7185438863521054,
      "grad_norm": 5.9375,
      "learning_rate": 5.317426780240314e-07,
      "loss": 1.59145966,
      "memory(GiB)": 128.51,
      "step": 67745,
      "train_speed(iter/s)": 1.637011
    },
    {
      "acc": 0.66266003,
      "epoch": 1.7186707255200406,
      "grad_norm": 5.21875,
      "learning_rate": 5.312721941531479e-07,
      "loss": 1.57306519,
      "memory(GiB)": 128.51,
      "step": 67750,
      "train_speed(iter/s)": 1.637023
    },
    {
      "acc": 0.65121617,
      "epoch": 1.7187975646879756,
      "grad_norm": 5.96875,
      "learning_rate": 5.308019068368647e-07,
      "loss": 1.62663727,
      "memory(GiB)": 128.51,
      "step": 67755,
      "train_speed(iter/s)": 1.637035
    },
    {
      "acc": 0.64435482,
      "epoch": 1.7189244038559108,
      "grad_norm": 5.75,
      "learning_rate": 5.303318160958675e-07,
      "loss": 1.65798759,
      "memory(GiB)": 128.51,
      "step": 67760,
      "train_speed(iter/s)": 1.637047
    },
    {
      "acc": 0.66324158,
      "epoch": 1.7190512430238458,
      "grad_norm": 6.3125,
      "learning_rate": 5.298619219508317e-07,
      "loss": 1.56744471,
      "memory(GiB)": 128.51,
      "step": 67765,
      "train_speed(iter/s)": 1.637058
    },
    {
      "acc": 0.64399962,
      "epoch": 1.7191780821917808,
      "grad_norm": 5.15625,
      "learning_rate": 5.293922244224275e-07,
      "loss": 1.64972839,
      "memory(GiB)": 128.51,
      "step": 67770,
      "train_speed(iter/s)": 1.63707
    },
    {
      "acc": 0.65744452,
      "epoch": 1.7193049213597158,
      "grad_norm": 5.8125,
      "learning_rate": 5.289227235313133e-07,
      "loss": 1.66066113,
      "memory(GiB)": 128.51,
      "step": 67775,
      "train_speed(iter/s)": 1.637082
    },
    {
      "acc": 0.65435376,
      "epoch": 1.7194317605276508,
      "grad_norm": 7.375,
      "learning_rate": 5.284534192981389e-07,
      "loss": 1.64509163,
      "memory(GiB)": 128.51,
      "step": 67780,
      "train_speed(iter/s)": 1.637094
    },
    {
      "acc": 0.65661602,
      "epoch": 1.719558599695586,
      "grad_norm": 6.125,
      "learning_rate": 5.279843117435463e-07,
      "loss": 1.5995266,
      "memory(GiB)": 128.51,
      "step": 67785,
      "train_speed(iter/s)": 1.637106
    },
    {
      "acc": 0.66487389,
      "epoch": 1.7196854388635212,
      "grad_norm": 7.375,
      "learning_rate": 5.275154008881716e-07,
      "loss": 1.56443815,
      "memory(GiB)": 128.51,
      "step": 67790,
      "train_speed(iter/s)": 1.637119
    },
    {
      "acc": 0.65768166,
      "epoch": 1.7198122780314562,
      "grad_norm": 6.25,
      "learning_rate": 5.270466867526369e-07,
      "loss": 1.59378281,
      "memory(GiB)": 128.51,
      "step": 67795,
      "train_speed(iter/s)": 1.637132
    },
    {
      "acc": 0.66073885,
      "epoch": 1.7199391171993912,
      "grad_norm": 5.71875,
      "learning_rate": 5.265781693575578e-07,
      "loss": 1.60156002,
      "memory(GiB)": 128.51,
      "step": 67800,
      "train_speed(iter/s)": 1.637143
    },
    {
      "acc": 0.64835563,
      "epoch": 1.7200659563673262,
      "grad_norm": 5.3125,
      "learning_rate": 5.261098487235449e-07,
      "loss": 1.59911194,
      "memory(GiB)": 128.51,
      "step": 67805,
      "train_speed(iter/s)": 1.637155
    },
    {
      "acc": 0.65387764,
      "epoch": 1.7201927955352612,
      "grad_norm": 5.75,
      "learning_rate": 5.256417248711959e-07,
      "loss": 1.60365086,
      "memory(GiB)": 128.51,
      "step": 67810,
      "train_speed(iter/s)": 1.637168
    },
    {
      "acc": 0.66301355,
      "epoch": 1.7203196347031964,
      "grad_norm": 5.28125,
      "learning_rate": 5.251737978210997e-07,
      "loss": 1.57626514,
      "memory(GiB)": 128.51,
      "step": 67815,
      "train_speed(iter/s)": 1.637179
    },
    {
      "acc": 0.63855886,
      "epoch": 1.7204464738711314,
      "grad_norm": 5.0,
      "learning_rate": 5.247060675938376e-07,
      "loss": 1.66888084,
      "memory(GiB)": 128.51,
      "step": 67820,
      "train_speed(iter/s)": 1.63719
    },
    {
      "acc": 0.6596756,
      "epoch": 1.7205733130390666,
      "grad_norm": 5.84375,
      "learning_rate": 5.242385342099842e-07,
      "loss": 1.63618908,
      "memory(GiB)": 128.51,
      "step": 67825,
      "train_speed(iter/s)": 1.637203
    },
    {
      "acc": 0.6773294,
      "epoch": 1.7207001522070016,
      "grad_norm": 6.21875,
      "learning_rate": 5.237711976901039e-07,
      "loss": 1.53798943,
      "memory(GiB)": 128.51,
      "step": 67830,
      "train_speed(iter/s)": 1.637215
    },
    {
      "acc": 0.65338316,
      "epoch": 1.7208269913749366,
      "grad_norm": 5.59375,
      "learning_rate": 5.233040580547489e-07,
      "loss": 1.60261517,
      "memory(GiB)": 128.51,
      "step": 67835,
      "train_speed(iter/s)": 1.637226
    },
    {
      "acc": 0.64087706,
      "epoch": 1.7209538305428715,
      "grad_norm": 5.46875,
      "learning_rate": 5.228371153244699e-07,
      "loss": 1.61807747,
      "memory(GiB)": 128.51,
      "step": 67840,
      "train_speed(iter/s)": 1.637238
    },
    {
      "acc": 0.64424658,
      "epoch": 1.7210806697108065,
      "grad_norm": 7.28125,
      "learning_rate": 5.223703695198034e-07,
      "loss": 1.6849762,
      "memory(GiB)": 128.51,
      "step": 67845,
      "train_speed(iter/s)": 1.63725
    },
    {
      "acc": 0.65990434,
      "epoch": 1.7212075088787417,
      "grad_norm": 6.625,
      "learning_rate": 5.219038206612798e-07,
      "loss": 1.57745113,
      "memory(GiB)": 128.51,
      "step": 67850,
      "train_speed(iter/s)": 1.637262
    },
    {
      "acc": 0.65871782,
      "epoch": 1.721334348046677,
      "grad_norm": 8.3125,
      "learning_rate": 5.214374687694191e-07,
      "loss": 1.61559029,
      "memory(GiB)": 128.51,
      "step": 67855,
      "train_speed(iter/s)": 1.637273
    },
    {
      "acc": 0.65611963,
      "epoch": 1.721461187214612,
      "grad_norm": 7.625,
      "learning_rate": 5.209713138647332e-07,
      "loss": 1.54990177,
      "memory(GiB)": 128.51,
      "step": 67860,
      "train_speed(iter/s)": 1.637285
    },
    {
      "acc": 0.64143906,
      "epoch": 1.721588026382547,
      "grad_norm": 6.15625,
      "learning_rate": 5.20505355967727e-07,
      "loss": 1.70312443,
      "memory(GiB)": 128.51,
      "step": 67865,
      "train_speed(iter/s)": 1.637297
    },
    {
      "acc": 0.68299685,
      "epoch": 1.721714865550482,
      "grad_norm": 5.96875,
      "learning_rate": 5.200395950988945e-07,
      "loss": 1.52948208,
      "memory(GiB)": 128.51,
      "step": 67870,
      "train_speed(iter/s)": 1.637309
    },
    {
      "acc": 0.65109282,
      "epoch": 1.721841704718417,
      "grad_norm": 7.375,
      "learning_rate": 5.195740312787229e-07,
      "loss": 1.68921089,
      "memory(GiB)": 128.51,
      "step": 67875,
      "train_speed(iter/s)": 1.637322
    },
    {
      "acc": 0.6647438,
      "epoch": 1.7219685438863521,
      "grad_norm": 6.8125,
      "learning_rate": 5.191086645276883e-07,
      "loss": 1.56310205,
      "memory(GiB)": 128.51,
      "step": 67880,
      "train_speed(iter/s)": 1.637334
    },
    {
      "acc": 0.6477344,
      "epoch": 1.722095383054287,
      "grad_norm": 6.625,
      "learning_rate": 5.186434948662605e-07,
      "loss": 1.58649569,
      "memory(GiB)": 128.51,
      "step": 67885,
      "train_speed(iter/s)": 1.637346
    },
    {
      "acc": 0.64870615,
      "epoch": 1.7222222222222223,
      "grad_norm": 7.15625,
      "learning_rate": 5.181785223148999e-07,
      "loss": 1.61224403,
      "memory(GiB)": 128.51,
      "step": 67890,
      "train_speed(iter/s)": 1.637358
    },
    {
      "acc": 0.65779119,
      "epoch": 1.7223490613901573,
      "grad_norm": 5.4375,
      "learning_rate": 5.177137468940574e-07,
      "loss": 1.54895439,
      "memory(GiB)": 128.51,
      "step": 67895,
      "train_speed(iter/s)": 1.63737
    },
    {
      "acc": 0.63718061,
      "epoch": 1.7224759005580923,
      "grad_norm": 5.625,
      "learning_rate": 5.172491686241765e-07,
      "loss": 1.74033813,
      "memory(GiB)": 128.51,
      "step": 67900,
      "train_speed(iter/s)": 1.637383
    },
    {
      "acc": 0.65718231,
      "epoch": 1.7226027397260273,
      "grad_norm": 5.78125,
      "learning_rate": 5.167847875256904e-07,
      "loss": 1.60256824,
      "memory(GiB)": 128.51,
      "step": 67905,
      "train_speed(iter/s)": 1.637395
    },
    {
      "acc": 0.65998745,
      "epoch": 1.7227295788939625,
      "grad_norm": 5.5625,
      "learning_rate": 5.163206036190272e-07,
      "loss": 1.61081467,
      "memory(GiB)": 128.51,
      "step": 67910,
      "train_speed(iter/s)": 1.637407
    },
    {
      "acc": 0.67601619,
      "epoch": 1.7228564180618975,
      "grad_norm": 6.40625,
      "learning_rate": 5.15856616924601e-07,
      "loss": 1.51871977,
      "memory(GiB)": 128.51,
      "step": 67915,
      "train_speed(iter/s)": 1.637419
    },
    {
      "acc": 0.65791874,
      "epoch": 1.7229832572298327,
      "grad_norm": 5.9375,
      "learning_rate": 5.153928274628211e-07,
      "loss": 1.61719761,
      "memory(GiB)": 128.51,
      "step": 67920,
      "train_speed(iter/s)": 1.637431
    },
    {
      "acc": 0.65185332,
      "epoch": 1.7231100963977677,
      "grad_norm": 7.03125,
      "learning_rate": 5.149292352540857e-07,
      "loss": 1.65299034,
      "memory(GiB)": 128.51,
      "step": 67925,
      "train_speed(iter/s)": 1.637443
    },
    {
      "acc": 0.66740294,
      "epoch": 1.7232369355657027,
      "grad_norm": 6.4375,
      "learning_rate": 5.144658403187896e-07,
      "loss": 1.60723743,
      "memory(GiB)": 128.51,
      "step": 67930,
      "train_speed(iter/s)": 1.637456
    },
    {
      "acc": 0.64974055,
      "epoch": 1.7233637747336377,
      "grad_norm": 5.40625,
      "learning_rate": 5.14002642677311e-07,
      "loss": 1.59239941,
      "memory(GiB)": 128.51,
      "step": 67935,
      "train_speed(iter/s)": 1.637469
    },
    {
      "acc": 0.64810677,
      "epoch": 1.7234906139015727,
      "grad_norm": 5.53125,
      "learning_rate": 5.135396423500233e-07,
      "loss": 1.60865936,
      "memory(GiB)": 128.51,
      "step": 67940,
      "train_speed(iter/s)": 1.637481
    },
    {
      "acc": 0.63797407,
      "epoch": 1.7236174530695079,
      "grad_norm": 5.03125,
      "learning_rate": 5.130768393572943e-07,
      "loss": 1.69640751,
      "memory(GiB)": 128.51,
      "step": 67945,
      "train_speed(iter/s)": 1.637493
    },
    {
      "acc": 0.64058113,
      "epoch": 1.723744292237443,
      "grad_norm": 5.25,
      "learning_rate": 5.126142337194795e-07,
      "loss": 1.64647846,
      "memory(GiB)": 128.51,
      "step": 67950,
      "train_speed(iter/s)": 1.637505
    },
    {
      "acc": 0.67002277,
      "epoch": 1.723871131405378,
      "grad_norm": 6.625,
      "learning_rate": 5.121518254569241e-07,
      "loss": 1.50479069,
      "memory(GiB)": 128.51,
      "step": 67955,
      "train_speed(iter/s)": 1.637517
    },
    {
      "acc": 0.66783686,
      "epoch": 1.723997970573313,
      "grad_norm": 5.0625,
      "learning_rate": 5.116896145899675e-07,
      "loss": 1.60261726,
      "memory(GiB)": 128.51,
      "step": 67960,
      "train_speed(iter/s)": 1.63753
    },
    {
      "acc": 0.64873695,
      "epoch": 1.724124809741248,
      "grad_norm": 5.65625,
      "learning_rate": 5.112276011389416e-07,
      "loss": 1.66556511,
      "memory(GiB)": 128.51,
      "step": 67965,
      "train_speed(iter/s)": 1.63754
    },
    {
      "acc": 0.64816608,
      "epoch": 1.724251648909183,
      "grad_norm": 5.875,
      "learning_rate": 5.107657851241671e-07,
      "loss": 1.62191849,
      "memory(GiB)": 128.51,
      "step": 67970,
      "train_speed(iter/s)": 1.637552
    },
    {
      "acc": 0.67163992,
      "epoch": 1.7243784880771182,
      "grad_norm": 6.84375,
      "learning_rate": 5.103041665659553e-07,
      "loss": 1.53661757,
      "memory(GiB)": 128.51,
      "step": 67975,
      "train_speed(iter/s)": 1.637564
    },
    {
      "acc": 0.65030098,
      "epoch": 1.7245053272450532,
      "grad_norm": 5.0625,
      "learning_rate": 5.098427454846116e-07,
      "loss": 1.63800507,
      "memory(GiB)": 128.51,
      "step": 67980,
      "train_speed(iter/s)": 1.637576
    },
    {
      "acc": 0.67146072,
      "epoch": 1.7246321664129884,
      "grad_norm": 6.09375,
      "learning_rate": 5.093815219004311e-07,
      "loss": 1.53376112,
      "memory(GiB)": 128.51,
      "step": 67985,
      "train_speed(iter/s)": 1.637589
    },
    {
      "acc": 0.65836477,
      "epoch": 1.7247590055809234,
      "grad_norm": 5.65625,
      "learning_rate": 5.089204958337018e-07,
      "loss": 1.5691987,
      "memory(GiB)": 128.51,
      "step": 67990,
      "train_speed(iter/s)": 1.637602
    },
    {
      "acc": 0.63850951,
      "epoch": 1.7248858447488584,
      "grad_norm": 5.5,
      "learning_rate": 5.084596673046982e-07,
      "loss": 1.68628883,
      "memory(GiB)": 128.51,
      "step": 67995,
      "train_speed(iter/s)": 1.637614
    },
    {
      "acc": 0.66782904,
      "epoch": 1.7250126839167934,
      "grad_norm": 6.25,
      "learning_rate": 5.079990363336928e-07,
      "loss": 1.56277695,
      "memory(GiB)": 128.51,
      "step": 68000,
      "train_speed(iter/s)": 1.637626
    },
    {
      "epoch": 1.7250126839167934,
      "eval_acc": 0.646264333611078,
      "eval_loss": 1.5730524063110352,
      "eval_runtime": 58.6761,
      "eval_samples_per_second": 108.562,
      "eval_steps_per_second": 27.149,
      "step": 68000
    }
  ],
  "logging_steps": 5,
  "max_steps": 78840,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3396563922916475e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}