{
  "best_metric": 2.6333401203155518,
  "best_model_checkpoint": "bert_base_96/checkpoint-160000",
  "epoch": 1.327606547706828,
  "global_step": 162045,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5.000000000000001e-07,
      "loss": 5.6747,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 5.6704,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5e-06,
      "loss": 5.6728,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 5.6664,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.5e-06,
      "loss": 5.6627,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-06,
      "loss": 5.6618,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.5e-06,
      "loss": 5.6592,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.000000000000001e-06,
      "loss": 5.662,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.5e-06,
      "loss": 5.6541,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-06,
      "loss": 5.6555,
      "step": 5000
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.500000000000001e-06,
      "loss": 5.6451,
      "step": 5500
    },
    {
      "epoch": 0.05,
      "learning_rate": 6e-06,
      "loss": 5.6436,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 5.6427,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "learning_rate": 7e-06,
      "loss": 5.6325,
      "step": 7000
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.500000000000001e-06,
      "loss": 5.6317,
      "step": 7500
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.000000000000001e-06,
      "loss": 5.623,
      "step": 8000
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.5e-06,
      "loss": 5.6137,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "learning_rate": 9e-06,
      "loss": 5.6142,
      "step": 9000
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.5e-06,
      "loss": 5.6072,
      "step": 9500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1e-05,
      "loss": 5.6041,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.1751239486362649,
      "eval_loss": 5.556705951690674,
      "eval_runtime": 2022.3509,
      "eval_samples_per_second": 152.474,
      "eval_steps_per_second": 3.177,
      "step": 10000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.991670692498626e-06,
      "loss": 5.5977,
      "step": 10500
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.983341384997253e-06,
      "loss": 5.5972,
      "step": 11000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.975012077495878e-06,
      "loss": 5.5857,
      "step": 11500
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.966682769994504e-06,
      "loss": 5.579,
      "step": 12000
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.95835346249313e-06,
      "loss": 5.5696,
      "step": 12500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.950024154991756e-06,
      "loss": 5.5638,
      "step": 13000
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.941694847490381e-06,
      "loss": 5.5597,
      "step": 13500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.933365539989006e-06,
      "loss": 5.5533,
      "step": 14000
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.925036232487631e-06,
      "loss": 5.5401,
      "step": 14500
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.916706924986258e-06,
      "loss": 5.537,
      "step": 15000
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.908377617484883e-06,
      "loss": 5.5251,
      "step": 15500
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.90004830998351e-06,
      "loss": 5.5249,
      "step": 16000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.891719002482134e-06,
      "loss": 5.5218,
      "step": 16500
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.883389694980761e-06,
      "loss": 5.5084,
      "step": 17000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.875060387479386e-06,
      "loss": 5.5067,
      "step": 17500
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.866731079978011e-06,
      "loss": 5.4997,
      "step": 18000
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.858401772476636e-06,
      "loss": 5.4967,
      "step": 18500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.850072464975263e-06,
      "loss": 5.4855,
      "step": 19000
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.84174315747389e-06,
      "loss": 5.4769,
      "step": 19500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.833413849972514e-06,
      "loss": 5.4727,
      "step": 20000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.19529749517644787,
      "eval_loss": 5.394957542419434,
      "eval_runtime": 2021.976,
      "eval_samples_per_second": 152.502,
      "eval_steps_per_second": 3.178,
      "step": 20000
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.82508454247114e-06,
      "loss": 5.4639,
      "step": 20500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.816755234969766e-06,
      "loss": 5.4583,
      "step": 21000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.808425927468391e-06,
      "loss": 5.4551,
      "step": 21500
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.800096619967016e-06,
      "loss": 5.4439,
      "step": 22000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.791767312465643e-06,
      "loss": 5.4381,
      "step": 22500
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.783438004964268e-06,
      "loss": 5.4265,
      "step": 23000
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.775108697462895e-06,
      "loss": 5.4225,
      "step": 23500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.76677938996152e-06,
      "loss": 5.4116,
      "step": 24000
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.758450082460145e-06,
      "loss": 5.4075,
      "step": 24500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.750120774958771e-06,
      "loss": 5.4026,
      "step": 25000
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.741791467457396e-06,
      "loss": 5.3926,
      "step": 25500
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.733462159956021e-06,
      "loss": 5.3871,
      "step": 26000
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.725132852454648e-06,
      "loss": 5.3869,
      "step": 26500
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.716803544953275e-06,
      "loss": 5.379,
      "step": 27000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.7084742374519e-06,
      "loss": 5.3693,
      "step": 27500
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.700144929950525e-06,
      "loss": 5.3635,
      "step": 28000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.69181562244915e-06,
      "loss": 5.3595,
      "step": 28500
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.683486314947776e-06,
      "loss": 5.3503,
      "step": 29000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.675157007446401e-06,
      "loss": 5.3427,
      "step": 29500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.666827699945028e-06,
      "loss": 5.3385,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.21506515987864405,
      "eval_loss": 5.227707386016846,
      "eval_runtime": 2020.7062,
      "eval_samples_per_second": 152.598,
      "eval_steps_per_second": 3.18,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.658498392443653e-06,
      "loss": 5.3281,
      "step": 30500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.65016908494228e-06,
      "loss": 5.3242,
      "step": 31000
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.641839777440905e-06,
      "loss": 5.3208,
      "step": 31500
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.63351046993953e-06,
      "loss": 5.3088,
      "step": 32000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.625181162438155e-06,
      "loss": 5.3013,
      "step": 32500
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.616851854936781e-06,
      "loss": 5.297,
      "step": 33000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.608522547435406e-06,
      "loss": 5.2886,
      "step": 33500
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.600193239934033e-06,
      "loss": 5.2822,
      "step": 34000
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.591863932432658e-06,
      "loss": 5.2704,
      "step": 34500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.583534624931285e-06,
      "loss": 5.2675,
      "step": 35000
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.57520531742991e-06,
      "loss": 5.2689,
      "step": 35500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.566876009928535e-06,
      "loss": 5.2584,
      "step": 36000
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.55854670242716e-06,
      "loss": 5.2501,
      "step": 36500
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.550217394925787e-06,
      "loss": 5.2425,
      "step": 37000
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.541888087424413e-06,
      "loss": 5.2309,
      "step": 37500
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.533558779923038e-06,
      "loss": 5.2195,
      "step": 38000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.525229472421663e-06,
      "loss": 5.213,
      "step": 38500
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.51690016492029e-06,
      "loss": 5.2071,
      "step": 39000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.508570857418915e-06,
      "loss": 5.2018,
      "step": 39500
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.50024154991754e-06,
      "loss": 5.2033,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.23352877002453412,
      "eval_loss": 5.060659885406494,
      "eval_runtime": 2019.9755,
      "eval_samples_per_second": 152.653,
      "eval_steps_per_second": 3.181,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.491912242416167e-06,
      "loss": 5.1896,
      "step": 40500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.483582934914792e-06,
      "loss": 5.1826,
      "step": 41000
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.475253627413418e-06,
      "loss": 5.1746,
      "step": 41500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.466924319912043e-06,
      "loss": 5.1703,
      "step": 42000
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.458595012410668e-06,
      "loss": 5.1556,
      "step": 42500
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.450265704909295e-06,
      "loss": 5.147,
      "step": 43000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.44193639740792e-06,
      "loss": 5.1313,
      "step": 43500
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.433607089906545e-06,
      "loss": 5.1213,
      "step": 44000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.425277782405172e-06,
      "loss": 5.0971,
      "step": 44500
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.416948474903798e-06,
      "loss": 5.0726,
      "step": 45000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.408619167402423e-06,
      "loss": 5.0427,
      "step": 45500
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.400289859901048e-06,
      "loss": 5.005,
      "step": 46000
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.391960552399673e-06,
      "loss": 4.9731,
      "step": 46500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.3836312448983e-06,
      "loss": 4.9338,
      "step": 47000
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.375301937396925e-06,
      "loss": 4.9047,
      "step": 47500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.366972629895552e-06,
      "loss": 4.8757,
      "step": 48000
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.358643322394177e-06,
      "loss": 4.8494,
      "step": 48500
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.350314014892804e-06,
      "loss": 4.8235,
      "step": 49000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.341984707391429e-06,
      "loss": 4.8007,
      "step": 49500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.333655399890054e-06,
      "loss": 4.7807,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.29099806447893206,
      "eval_loss": 4.561127662658691,
      "eval_runtime": 2022.9707,
      "eval_samples_per_second": 152.427,
      "eval_steps_per_second": 3.176,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.325326092388679e-06,
      "loss": 4.7544,
      "step": 50500
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.316996784887305e-06,
      "loss": 4.7364,
      "step": 51000
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.30866747738593e-06,
      "loss": 4.7173,
      "step": 51500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.300338169884557e-06,
      "loss": 4.6945,
      "step": 52000
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.292008862383182e-06,
      "loss": 4.6756,
      "step": 52500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.283679554881809e-06,
      "loss": 4.659,
      "step": 53000
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.275350247380434e-06,
      "loss": 4.6358,
      "step": 53500
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.267020939879059e-06,
      "loss": 4.6186,
      "step": 54000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.258691632377685e-06,
      "loss": 4.6,
      "step": 54500
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.25036232487631e-06,
      "loss": 4.5756,
      "step": 55000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.242033017374937e-06,
      "loss": 4.5598,
      "step": 55500
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.233703709873562e-06,
      "loss": 4.5248,
      "step": 56000
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.225374402372189e-06,
      "loss": 4.4891,
      "step": 56500
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.217045094870814e-06,
      "loss": 4.4484,
      "step": 57000
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.208715787369439e-06,
      "loss": 4.4173,
      "step": 57500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.200386479868064e-06,
      "loss": 4.3734,
      "step": 58000
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.19205717236669e-06,
      "loss": 4.3178,
      "step": 58500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.183727864865315e-06,
      "loss": 4.2702,
      "step": 59000
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.175398557363942e-06,
      "loss": 4.2339,
      "step": 59500
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.167069249862567e-06,
      "loss": 4.1994,
      "step": 60000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.3519629475668502,
      "eval_loss": 4.003937244415283,
      "eval_runtime": 2023.0275,
      "eval_samples_per_second": 152.423,
      "eval_steps_per_second": 3.176,
      "step": 60000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.158739942361194e-06,
      "loss": 4.1583,
      "step": 60500
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.150410634859819e-06,
      "loss": 4.1263,
      "step": 61000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.142081327358444e-06,
      "loss": 4.1041,
      "step": 61500
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.133752019857069e-06,
      "loss": 4.0725,
      "step": 62000
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.125422712355696e-06,
      "loss": 4.0584,
      "step": 62500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.11709340485432e-06,
      "loss": 4.0382,
      "step": 63000
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.108764097352947e-06,
      "loss": 4.004,
      "step": 63500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.100434789851572e-06,
      "loss": 3.9919,
      "step": 64000
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.092105482350199e-06,
      "loss": 3.9729,
      "step": 64500
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.083776174848824e-06,
      "loss": 3.9569,
      "step": 65000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.075446867347449e-06,
      "loss": 3.9429,
      "step": 65500
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.067117559846076e-06,
      "loss": 3.9256,
      "step": 66000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.0587882523447e-06,
      "loss": 3.9038,
      "step": 66500
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.050458944843327e-06,
      "loss": 3.8824,
      "step": 67000
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.042129637341952e-06,
      "loss": 3.874,
      "step": 67500
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.033800329840577e-06,
      "loss": 3.8635,
      "step": 68000
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.025471022339204e-06,
      "loss": 3.8512,
      "step": 68500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.017141714837829e-06,
      "loss": 3.831,
      "step": 69000
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.008812407336454e-06,
      "loss": 3.8224,
      "step": 69500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.00048309983508e-06,
      "loss": 3.8039,
      "step": 70000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.39063701352494984,
      "eval_loss": 3.650850772857666,
      "eval_runtime": 2019.3536,
      "eval_samples_per_second": 152.7,
      "eval_steps_per_second": 3.182,
      "step": 70000
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.992153792333706e-06,
      "loss": 3.793,
      "step": 70500
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.983824484832332e-06,
      "loss": 3.7767,
      "step": 71000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.975495177330957e-06,
      "loss": 3.7713,
      "step": 71500
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.967165869829582e-06,
      "loss": 3.7544,
      "step": 72000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.958836562328209e-06,
      "loss": 3.7474,
      "step": 72500
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.950507254826834e-06,
      "loss": 3.7238,
      "step": 73000
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.942177947325459e-06,
      "loss": 3.7185,
      "step": 73500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.933848639824086e-06,
      "loss": 3.6989,
      "step": 74000
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.925519332322713e-06,
      "loss": 3.6917,
      "step": 74500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.917190024821338e-06,
      "loss": 3.6831,
      "step": 75000
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.908860717319963e-06,
      "loss": 3.6593,
      "step": 75500
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.900531409818588e-06,
      "loss": 3.6493,
      "step": 76000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.892202102317214e-06,
      "loss": 3.6347,
      "step": 76500
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.88387279481584e-06,
      "loss": 3.6199,
      "step": 77000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.875543487314466e-06,
      "loss": 3.6104,
      "step": 77500
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.867214179813091e-06,
      "loss": 3.5947,
      "step": 78000
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.858884872311718e-06,
      "loss": 3.5889,
      "step": 78500
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.850555564810343e-06,
      "loss": 3.5742,
      "step": 79000
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.842226257308968e-06,
      "loss": 3.5641,
      "step": 79500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.833896949807593e-06,
      "loss": 3.5516,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.4262595441568208,
      "eval_loss": 3.379427433013916,
      "eval_runtime": 2024.6698,
      "eval_samples_per_second": 152.299,
      "eval_steps_per_second": 3.173,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.82556764230622e-06,
      "loss": 3.539,
      "step": 80500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.817238334804844e-06,
      "loss": 3.5291,
      "step": 81000
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.808909027303471e-06,
      "loss": 3.5092,
      "step": 81500
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.800579719802096e-06,
      "loss": 3.4993,
      "step": 82000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.792250412300723e-06,
      "loss": 3.4753,
      "step": 82500
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.783921104799348e-06,
      "loss": 3.4757,
      "step": 83000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.775591797297973e-06,
      "loss": 3.4625,
      "step": 83500
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.767262489796598e-06,
      "loss": 3.4475,
      "step": 84000
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.758933182295224e-06,
      "loss": 3.4323,
      "step": 84500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.750603874793851e-06,
      "loss": 3.4254,
      "step": 85000
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.742274567292476e-06,
      "loss": 3.4152,
      "step": 85500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.733945259791101e-06,
      "loss": 3.3977,
      "step": 86000
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.725615952289728e-06,
      "loss": 3.3938,
      "step": 86500
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.717286644788353e-06,
      "loss": 3.3823,
      "step": 87000
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.708957337286978e-06,
      "loss": 3.3697,
      "step": 87500
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.700628029785605e-06,
      "loss": 3.3572,
      "step": 88000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.69229872228423e-06,
      "loss": 3.3522,
      "step": 88500
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.683969414782856e-06,
      "loss": 3.3394,
      "step": 89000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.675640107281481e-06,
      "loss": 3.3283,
      "step": 89500
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.667310799780106e-06,
      "loss": 3.3199,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.46065701806340276,
      "eval_loss": 3.144608736038208,
      "eval_runtime": 2018.4911,
      "eval_samples_per_second": 152.766,
      "eval_steps_per_second": 3.183,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.658981492278733e-06,
      "loss": 3.3048,
      "step": 90500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.650652184777358e-06,
      "loss": 3.3035,
      "step": 91000
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.642322877275983e-06,
      "loss": 3.2933,
      "step": 91500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.63399356977461e-06,
      "loss": 3.2903,
      "step": 92000
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.625664262273236e-06,
      "loss": 3.2737,
      "step": 92500
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.617334954771861e-06,
      "loss": 3.2725,
      "step": 93000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.609005647270486e-06,
      "loss": 3.2533,
      "step": 93500
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.600676339769111e-06,
      "loss": 3.2468,
      "step": 94000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.592347032267738e-06,
      "loss": 3.2471,
      "step": 94500
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.584017724766363e-06,
      "loss": 3.2345,
      "step": 95000
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.57568841726499e-06,
      "loss": 3.2312,
      "step": 95500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.567359109763615e-06,
      "loss": 3.2234,
      "step": 96000
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.559029802262241e-06,
      "loss": 3.213,
      "step": 96500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.550700494760866e-06,
      "loss": 3.2066,
      "step": 97000
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.542371187259491e-06,
      "loss": 3.202,
      "step": 97500
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.534041879758118e-06,
      "loss": 3.1947,
      "step": 98000
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.525712572256743e-06,
      "loss": 3.1917,
      "step": 98500
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.517383264755368e-06,
      "loss": 3.1849,
      "step": 99000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.509053957253995e-06,
      "loss": 3.1712,
      "step": 99500
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.500724649752622e-06,
      "loss": 3.1682,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "eval_accuracy": 0.4795398790446386,
      "eval_loss": 3.005267858505249,
      "eval_runtime": 2019.0302,
      "eval_samples_per_second": 152.725,
      "eval_steps_per_second": 3.182,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.492395342251247e-06,
      "loss": 3.16,
      "step": 100500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.484066034749872e-06,
      "loss": 3.1461,
      "step": 101000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.475736727248497e-06,
      "loss": 3.1539,
      "step": 101500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.467407419747123e-06,
      "loss": 3.1408,
      "step": 102000
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.459078112245748e-06,
      "loss": 3.1372,
      "step": 102500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.450748804744375e-06,
      "loss": 3.1337,
      "step": 103000
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.442419497243e-06,
      "loss": 3.1249,
      "step": 103500
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.434090189741627e-06,
      "loss": 3.1144,
      "step": 104000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.425760882240252e-06,
      "loss": 3.1242,
      "step": 104500
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.417431574738877e-06,
      "loss": 3.1116,
      "step": 105000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.409102267237502e-06,
      "loss": 3.1128,
      "step": 105500
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.400772959736128e-06,
      "loss": 3.1007,
      "step": 106000
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.392443652234753e-06,
      "loss": 3.0957,
      "step": 106500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.38411434473338e-06,
      "loss": 3.0817,
      "step": 107000
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.375785037232005e-06,
      "loss": 3.0774,
      "step": 107500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.367455729730632e-06,
      "loss": 3.0822,
      "step": 108000
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.359126422229257e-06,
      "loss": 3.0736,
      "step": 108500
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.350797114727882e-06,
      "loss": 3.0668,
      "step": 109000
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.342467807226507e-06,
      "loss": 3.065,
      "step": 109500
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.334138499725133e-06,
      "loss": 3.0597,
      "step": 110000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.49193333243221515,
      "eval_loss": 2.913475275039673,
      "eval_runtime": 2020.005,
      "eval_samples_per_second": 152.651,
      "eval_steps_per_second": 3.181,
      "step": 110000
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.32580919222376e-06,
      "loss": 3.058,
      "step": 110500
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.317479884722385e-06,
      "loss": 3.0514,
      "step": 111000
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.30915057722101e-06,
      "loss": 3.0463,
      "step": 111500
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.300821269719637e-06,
      "loss": 3.0376,
      "step": 112000
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.292491962218262e-06,
      "loss": 3.0426,
      "step": 112500
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.284162654716887e-06,
      "loss": 3.0338,
      "step": 113000
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.275833347215514e-06,
      "loss": 3.0263,
      "step": 113500
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.267504039714139e-06,
      "loss": 3.0303,
      "step": 114000
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.259174732212765e-06,
      "loss": 3.0215,
      "step": 114500
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.25084542471139e-06,
      "loss": 3.0199,
      "step": 115000
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.242516117210015e-06,
      "loss": 3.0135,
      "step": 115500
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.234186809708642e-06,
      "loss": 3.006,
      "step": 116000
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.225857502207267e-06,
      "loss": 3.0012,
      "step": 116500
    },
    {
      "epoch": 0.96,
      "learning_rate": 8.217528194705892e-06,
      "loss": 2.998,
      "step": 117000
    },
    {
      "epoch": 0.96,
      "learning_rate": 8.209198887204519e-06,
      "loss": 2.9991,
      "step": 117500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.200869579703145e-06,
      "loss": 2.9927,
      "step": 118000
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.19254027220177e-06,
      "loss": 2.9942,
      "step": 118500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.184210964700395e-06,
      "loss": 2.9894,
      "step": 119000
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.17588165719902e-06,
      "loss": 2.9799,
      "step": 119500
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.167552349697647e-06,
      "loss": 2.9814,
      "step": 120000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.501779410535526,
      "eval_loss": 2.8330531120300293,
      "eval_runtime": 2018.8078,
      "eval_samples_per_second": 152.742,
      "eval_steps_per_second": 3.183,
      "step": 120000
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.159223042196272e-06,
      "loss": 2.9748,
      "step": 120500
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.150893734694899e-06,
      "loss": 2.9687,
      "step": 121000
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.142564427193524e-06,
      "loss": 2.9705,
      "step": 121500
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.13423511969215e-06,
      "loss": 2.9591,
      "step": 122000
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.125905812190775e-06,
      "loss": 2.9563,
      "step": 122500
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.1175765046894e-06,
      "loss": 2.954,
      "step": 123000
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.109247197188025e-06,
      "loss": 2.948,
      "step": 123500
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.100917889686652e-06,
      "loss": 2.9432,
      "step": 124000
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.092588582185277e-06,
      "loss": 2.9438,
      "step": 124500
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.084259274683904e-06,
      "loss": 2.9395,
      "step": 125000
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.075929967182529e-06,
      "loss": 2.9411,
      "step": 125500
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.067600659681156e-06,
      "loss": 2.9319,
      "step": 126000
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.05927135217978e-06,
      "loss": 2.9297,
      "step": 126500
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.050942044678406e-06,
      "loss": 2.9275,
      "step": 127000
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.04261273717703e-06,
      "loss": 2.9254,
      "step": 127500
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.034283429675657e-06,
      "loss": 2.9216,
      "step": 128000
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.025954122174284e-06,
      "loss": 2.9151,
      "step": 128500
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.017624814672909e-06,
      "loss": 2.9206,
      "step": 129000
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.009295507171534e-06,
      "loss": 2.9115,
      "step": 129500
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.00096619967016e-06,
      "loss": 2.907,
      "step": 130000
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.5099985482252666,
      "eval_loss": 2.7723934650421143,
      "eval_runtime": 2022.4256,
      "eval_samples_per_second": 152.468,
      "eval_steps_per_second": 3.177,
      "step": 130000
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.992636892168786e-06,
      "loss": 2.9019,
      "step": 130500
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.98430758466741e-06,
      "loss": 2.9107,
      "step": 131000
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.975978277166037e-06,
      "loss": 2.9036,
      "step": 131500
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.967648969664662e-06,
      "loss": 2.8998,
      "step": 132000
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.959319662163289e-06,
      "loss": 2.8984,
      "step": 132500
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.950990354661914e-06,
      "loss": 2.8916,
      "step": 133000
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.942661047160539e-06,
      "loss": 2.8888,
      "step": 133500
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.934331739659166e-06,
      "loss": 2.884,
      "step": 134000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.92600243215779e-06,
      "loss": 2.8865,
      "step": 134500
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.917673124656416e-06,
      "loss": 2.8757,
      "step": 135000
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.909343817155042e-06,
      "loss": 2.8808,
      "step": 135500
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.901014509653669e-06,
      "loss": 2.8763,
      "step": 136000
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.892685202152294e-06,
      "loss": 2.8675,
      "step": 136500
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.884355894650919e-06,
      "loss": 2.8757,
      "step": 137000
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.876026587149544e-06,
      "loss": 2.8635,
      "step": 137500
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.86769727964817e-06,
      "loss": 2.8643,
      "step": 138000
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.859367972146796e-06,
      "loss": 2.8588,
      "step": 138500
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.851038664645423e-06,
      "loss": 2.8591,
      "step": 139000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.842709357144048e-06,
      "loss": 2.8554,
      "step": 139500
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.834380049642674e-06,
      "loss": 2.8532,
      "step": 140000
    },
    {
      "epoch": 1.15,
      "eval_accuracy": 0.5169768747303877,
      "eval_loss": 2.7199792861938477,
      "eval_runtime": 2018.7949,
      "eval_samples_per_second": 152.743,
      "eval_steps_per_second": 3.183,
      "step": 140000
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.8260507421413e-06,
      "loss": 2.8517,
      "step": 140500
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.817721434639924e-06,
      "loss": 2.8474,
      "step": 141000
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.809392127138551e-06,
      "loss": 2.8368,
      "step": 141500
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.801062819637176e-06,
      "loss": 2.8396,
      "step": 142000
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.792733512135801e-06,
      "loss": 2.8453,
      "step": 142500
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.784404204634428e-06,
      "loss": 2.8366,
      "step": 143000
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.776074897133054e-06,
      "loss": 2.8355,
      "step": 143500
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.76774558963168e-06,
      "loss": 2.8371,
      "step": 144000
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.759416282130304e-06,
      "loss": 2.8325,
      "step": 144500
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.75108697462893e-06,
      "loss": 2.8254,
      "step": 145000
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.742757667127556e-06,
      "loss": 2.8204,
      "step": 145500
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.734428359626181e-06,
      "loss": 2.8194,
      "step": 146000
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.726099052124808e-06,
      "loss": 2.8221,
      "step": 146500
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.717769744623433e-06,
      "loss": 2.8181,
      "step": 147000
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.70944043712206e-06,
      "loss": 2.8078,
      "step": 147500
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.701111129620684e-06,
      "loss": 2.8101,
      "step": 148000
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.69278182211931e-06,
      "loss": 2.8158,
      "step": 148500
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.684452514617934e-06,
      "loss": 2.8091,
      "step": 149000
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.676123207116561e-06,
      "loss": 2.8139,
      "step": 149500
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.667793899615186e-06,
      "loss": 2.8044,
      "step": 150000
    },
    {
      "epoch": 1.23,
      "eval_accuracy": 0.5226956511260885,
      "eval_loss": 2.6759424209594727,
      "eval_runtime": 2018.5063,
      "eval_samples_per_second": 152.764,
      "eval_steps_per_second": 3.183,
      "step": 150000
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.659464592113813e-06,
      "loss": 2.8052,
      "step": 150500
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.651135284612438e-06,
      "loss": 2.8044,
      "step": 151000
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.642805977111065e-06,
      "loss": 2.7948,
      "step": 151500
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.63447666960969e-06,
      "loss": 2.7968,
      "step": 152000
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.6261473621083145e-06,
      "loss": 2.7953,
      "step": 152500
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.61781805460694e-06,
      "loss": 2.7902,
      "step": 153000
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.609488747105567e-06,
      "loss": 2.7947,
      "step": 153500
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.601159439604192e-06,
      "loss": 2.7851,
      "step": 154000
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.592830132102817e-06,
      "loss": 2.784,
      "step": 154500
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.584500824601443e-06,
      "loss": 2.7839,
      "step": 155000
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.57617151710007e-06,
      "loss": 2.7773,
      "step": 155500
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.567842209598695e-06,
      "loss": 2.779,
      "step": 156000
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.5595129020973205e-06,
      "loss": 2.7752,
      "step": 156500
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.5511835945959455e-06,
      "loss": 2.7779,
      "step": 157000
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.542854287094572e-06,
      "loss": 2.7672,
      "step": 157500
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.534524979593197e-06,
      "loss": 2.7718,
      "step": 158000
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.526195672091823e-06,
      "loss": 2.7622,
      "step": 158500
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.517866364590448e-06,
      "loss": 2.7701,
      "step": 159000
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.509537057089075e-06,
      "loss": 2.7645,
      "step": 159500
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.5012077495877e-06,
      "loss": 2.7694,
      "step": 160000
    },
    {
      "epoch": 1.31,
      "eval_accuracy": 0.5281042082913918,
      "eval_loss": 2.6333401203155518,
      "eval_runtime": 2021.1996,
      "eval_samples_per_second": 152.561,
      "eval_steps_per_second": 3.179,
      "step": 160000
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.492878442086326e-06,
      "loss": 2.7694,
      "step": 160500
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.484549134584951e-06,
      "loss": 2.7578,
      "step": 161000
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.476219827083577e-06,
      "loss": 2.7541,
      "step": 161500
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.467890519582202e-06,
      "loss": 2.7553,
      "step": 162000
    },
    {
      "epoch": 1.33,
      "step": 162045,
      "total_flos": 2.047242298445529e+18,
      "train_loss": 3.9368587984137284,
      "train_runtime": 179983.0001,
      "train_samples_per_second": 162.759,
      "train_steps_per_second": 3.391
    }
  ],
  "max_steps": 610290,
  "num_train_epochs": 5,
  "total_flos": 2.047242298445529e+18,
  "trial_name": null,
  "trial_params": null
}