{
  "best_metric": 5.959463119506836,
  "best_model_checkpoint": "bert_12_layer_model_v3_complete_training_new_emb_compress_48/checkpoint-220000",
  "epoch": 1.8715446754821479,
  "eval_steps": 10000,
  "global_step": 228437,
  "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": 10.1944,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 10.0821,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5e-06,
      "loss": 10.0056,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 9.8977,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.5e-06,
      "loss": 9.7731,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-06,
      "loss": 9.6289,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.5e-06,
      "loss": 9.4552,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.000000000000001e-06,
      "loss": 9.2593,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.5e-06,
      "loss": 9.0418,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-06,
      "loss": 8.8074,
      "step": 5000
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.500000000000001e-06,
      "loss": 8.5628,
      "step": 5500
    },
    {
      "epoch": 0.05,
      "learning_rate": 6e-06,
      "loss": 8.3118,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 8.0665,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "learning_rate": 7e-06,
      "loss": 7.8375,
      "step": 7000
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.500000000000001e-06,
      "loss": 7.6306,
      "step": 7500
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.000000000000001e-06,
      "loss": 7.4542,
      "step": 8000
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.5e-06,
      "loss": 7.3165,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "learning_rate": 9e-06,
      "loss": 7.2245,
      "step": 9000
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.5e-06,
      "loss": 7.1592,
      "step": 9500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1e-05,
      "loss": 7.1148,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.08279736975484835,
      "eval_loss": 7.092103958129883,
      "eval_runtime": 1608.3621,
      "eval_samples_per_second": 191.721,
      "eval_steps_per_second": 3.995,
      "step": 10000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.991670692498626e-06,
      "loss": 7.0718,
      "step": 10500
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.983341384997253e-06,
      "loss": 7.025,
      "step": 11000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.975012077495878e-06,
      "loss": 6.9898,
      "step": 11500
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.966682769994504e-06,
      "loss": 6.9628,
      "step": 12000
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.95835346249313e-06,
      "loss": 6.9377,
      "step": 12500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.950024154991756e-06,
      "loss": 6.9052,
      "step": 13000
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.941694847490381e-06,
      "loss": 6.8898,
      "step": 13500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.933365539989006e-06,
      "loss": 6.8726,
      "step": 14000
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.925036232487631e-06,
      "loss": 6.8448,
      "step": 14500
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.916706924986258e-06,
      "loss": 6.827,
      "step": 15000
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.908377617484883e-06,
      "loss": 6.8116,
      "step": 15500
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.90004830998351e-06,
      "loss": 6.7988,
      "step": 16000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.891719002482134e-06,
      "loss": 6.7828,
      "step": 16500
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.883389694980761e-06,
      "loss": 6.7658,
      "step": 17000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.875060387479386e-06,
      "loss": 6.7577,
      "step": 17500
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.866731079978011e-06,
      "loss": 6.741,
      "step": 18000
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.858401772476636e-06,
      "loss": 6.7266,
      "step": 18500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.850072464975263e-06,
      "loss": 6.7127,
      "step": 19000
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.84174315747389e-06,
      "loss": 6.7075,
      "step": 19500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.833413849972514e-06,
      "loss": 6.6864,
      "step": 20000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.10781515354671665,
      "eval_loss": 6.687861442565918,
      "eval_runtime": 1613.4134,
      "eval_samples_per_second": 191.12,
      "eval_steps_per_second": 3.982,
      "step": 20000
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.82508454247114e-06,
      "loss": 6.6822,
      "step": 20500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.816755234969766e-06,
      "loss": 6.6796,
      "step": 21000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.808425927468391e-06,
      "loss": 6.6565,
      "step": 21500
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.800096619967016e-06,
      "loss": 6.6529,
      "step": 22000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.791767312465643e-06,
      "loss": 6.646,
      "step": 22500
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.783438004964268e-06,
      "loss": 6.6419,
      "step": 23000
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.775108697462895e-06,
      "loss": 6.6263,
      "step": 23500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.76677938996152e-06,
      "loss": 6.6259,
      "step": 24000
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.758450082460145e-06,
      "loss": 6.607,
      "step": 24500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.750120774958771e-06,
      "loss": 6.6196,
      "step": 25000
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.741791467457396e-06,
      "loss": 6.6021,
      "step": 25500
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.733462159956021e-06,
      "loss": 6.5996,
      "step": 26000
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.725132852454648e-06,
      "loss": 6.5905,
      "step": 26500
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.716803544953275e-06,
      "loss": 6.5782,
      "step": 27000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.7084742374519e-06,
      "loss": 6.5708,
      "step": 27500
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.700144929950525e-06,
      "loss": 6.567,
      "step": 28000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.69181562244915e-06,
      "loss": 6.567,
      "step": 28500
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.683486314947776e-06,
      "loss": 6.5564,
      "step": 29000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.675157007446401e-06,
      "loss": 6.5535,
      "step": 29500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.666827699945028e-06,
      "loss": 6.5451,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.11838404385340955,
      "eval_loss": 6.543458461761475,
      "eval_runtime": 1629.5563,
      "eval_samples_per_second": 189.227,
      "eval_steps_per_second": 3.943,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.658498392443653e-06,
      "loss": 6.5368,
      "step": 30500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.65016908494228e-06,
      "loss": 6.5332,
      "step": 31000
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.641839777440905e-06,
      "loss": 6.5326,
      "step": 31500
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.63351046993953e-06,
      "loss": 6.5324,
      "step": 32000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.625181162438155e-06,
      "loss": 6.5224,
      "step": 32500
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.616851854936781e-06,
      "loss": 6.5124,
      "step": 33000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.608522547435406e-06,
      "loss": 6.5035,
      "step": 33500
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.600193239934033e-06,
      "loss": 6.5056,
      "step": 34000
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.591863932432658e-06,
      "loss": 6.5019,
      "step": 34500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.583534624931285e-06,
      "loss": 6.4976,
      "step": 35000
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.57520531742991e-06,
      "loss": 6.5032,
      "step": 35500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.566876009928535e-06,
      "loss": 6.4875,
      "step": 36000
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.55854670242716e-06,
      "loss": 6.4797,
      "step": 36500
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.550217394925787e-06,
      "loss": 6.477,
      "step": 37000
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.541888087424413e-06,
      "loss": 6.4819,
      "step": 37500
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.533558779923038e-06,
      "loss": 6.4632,
      "step": 38000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.525229472421663e-06,
      "loss": 6.4702,
      "step": 38500
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.51690016492029e-06,
      "loss": 6.4625,
      "step": 39000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.508570857418915e-06,
      "loss": 6.4569,
      "step": 39500
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.50024154991754e-06,
      "loss": 6.4606,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.12623867082041954,
      "eval_loss": 6.451531410217285,
      "eval_runtime": 1623.6423,
      "eval_samples_per_second": 189.916,
      "eval_steps_per_second": 3.957,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.491912242416167e-06,
      "loss": 6.4475,
      "step": 40500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.483582934914792e-06,
      "loss": 6.4493,
      "step": 41000
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.475253627413418e-06,
      "loss": 6.4458,
      "step": 41500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.466924319912043e-06,
      "loss": 6.4429,
      "step": 42000
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.458595012410668e-06,
      "loss": 6.4426,
      "step": 42500
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.450265704909295e-06,
      "loss": 6.4316,
      "step": 43000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.44193639740792e-06,
      "loss": 6.4323,
      "step": 43500
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.433607089906545e-06,
      "loss": 6.4293,
      "step": 44000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.425277782405172e-06,
      "loss": 6.4227,
      "step": 44500
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.416948474903798e-06,
      "loss": 6.4175,
      "step": 45000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.408619167402423e-06,
      "loss": 6.4096,
      "step": 45500
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.400289859901048e-06,
      "loss": 6.4115,
      "step": 46000
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.391960552399673e-06,
      "loss": 6.4044,
      "step": 46500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.3836312448983e-06,
      "loss": 6.4094,
      "step": 47000
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.375301937396925e-06,
      "loss": 6.4104,
      "step": 47500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.366972629895552e-06,
      "loss": 6.4036,
      "step": 48000
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.358643322394177e-06,
      "loss": 6.3931,
      "step": 48500
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.350314014892804e-06,
      "loss": 6.3938,
      "step": 49000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.341984707391429e-06,
      "loss": 6.3958,
      "step": 49500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.333655399890054e-06,
      "loss": 6.3851,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.13119015571432516,
      "eval_loss": 6.385060787200928,
      "eval_runtime": 1624.192,
      "eval_samples_per_second": 189.852,
      "eval_steps_per_second": 3.956,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.325326092388679e-06,
      "loss": 6.39,
      "step": 50500
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.316996784887305e-06,
      "loss": 6.3824,
      "step": 51000
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.30866747738593e-06,
      "loss": 6.3842,
      "step": 51500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.300338169884557e-06,
      "loss": 6.3739,
      "step": 52000
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.292008862383182e-06,
      "loss": 6.3724,
      "step": 52500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.283679554881809e-06,
      "loss": 6.3647,
      "step": 53000
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.275350247380434e-06,
      "loss": 6.3667,
      "step": 53500
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.267020939879059e-06,
      "loss": 6.3661,
      "step": 54000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.258691632377685e-06,
      "loss": 6.3543,
      "step": 54500
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.25036232487631e-06,
      "loss": 6.3613,
      "step": 55000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.242033017374937e-06,
      "loss": 6.3576,
      "step": 55500
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.233703709873562e-06,
      "loss": 6.3587,
      "step": 56000
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.225374402372189e-06,
      "loss": 6.3561,
      "step": 56500
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.217045094870814e-06,
      "loss": 6.3522,
      "step": 57000
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.208715787369439e-06,
      "loss": 6.3527,
      "step": 57500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.200386479868064e-06,
      "loss": 6.339,
      "step": 58000
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.19205717236669e-06,
      "loss": 6.3507,
      "step": 58500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.183727864865315e-06,
      "loss": 6.3377,
      "step": 59000
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.175398557363942e-06,
      "loss": 6.3422,
      "step": 59500
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.167069249862567e-06,
      "loss": 6.3371,
      "step": 60000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.13421311122916418,
      "eval_loss": 6.335691452026367,
      "eval_runtime": 1613.3881,
      "eval_samples_per_second": 191.123,
      "eval_steps_per_second": 3.982,
      "step": 60000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.158739942361194e-06,
      "loss": 6.3399,
      "step": 60500
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.150410634859819e-06,
      "loss": 6.3343,
      "step": 61000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.142081327358444e-06,
      "loss": 6.3314,
      "step": 61500
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.133752019857069e-06,
      "loss": 6.3309,
      "step": 62000
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.125422712355696e-06,
      "loss": 6.3234,
      "step": 62500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.11709340485432e-06,
      "loss": 6.329,
      "step": 63000
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.108764097352947e-06,
      "loss": 6.3251,
      "step": 63500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.100434789851572e-06,
      "loss": 6.3108,
      "step": 64000
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.092105482350199e-06,
      "loss": 6.3138,
      "step": 64500
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.083776174848824e-06,
      "loss": 6.3164,
      "step": 65000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.075446867347449e-06,
      "loss": 6.3148,
      "step": 65500
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.067117559846076e-06,
      "loss": 6.3193,
      "step": 66000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.0587882523447e-06,
      "loss": 6.3052,
      "step": 66500
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.050458944843327e-06,
      "loss": 6.3135,
      "step": 67000
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.042129637341952e-06,
      "loss": 6.3054,
      "step": 67500
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.033800329840577e-06,
      "loss": 6.297,
      "step": 68000
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.025471022339204e-06,
      "loss": 6.3014,
      "step": 68500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.017141714837829e-06,
      "loss": 6.2981,
      "step": 69000
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.008812407336454e-06,
      "loss": 6.2956,
      "step": 69500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.00048309983508e-06,
      "loss": 6.2971,
      "step": 70000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.13728091000158268,
      "eval_loss": 6.292341709136963,
      "eval_runtime": 1609.1726,
      "eval_samples_per_second": 191.624,
      "eval_steps_per_second": 3.993,
      "step": 70000
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.992153792333706e-06,
      "loss": 6.2955,
      "step": 70500
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.983824484832332e-06,
      "loss": 6.294,
      "step": 71000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.975495177330957e-06,
      "loss": 6.2903,
      "step": 71500
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.967165869829582e-06,
      "loss": 6.2966,
      "step": 72000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.958836562328209e-06,
      "loss": 6.2846,
      "step": 72500
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.950507254826834e-06,
      "loss": 6.2835,
      "step": 73000
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.942177947325459e-06,
      "loss": 6.2811,
      "step": 73500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.933848639824086e-06,
      "loss": 6.2799,
      "step": 74000
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.925519332322713e-06,
      "loss": 6.2785,
      "step": 74500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.917190024821338e-06,
      "loss": 6.2818,
      "step": 75000
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.908860717319963e-06,
      "loss": 6.2802,
      "step": 75500
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.900531409818588e-06,
      "loss": 6.2821,
      "step": 76000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.892202102317214e-06,
      "loss": 6.2762,
      "step": 76500
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.88387279481584e-06,
      "loss": 6.2745,
      "step": 77000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.875543487314466e-06,
      "loss": 6.2687,
      "step": 77500
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.867214179813091e-06,
      "loss": 6.2672,
      "step": 78000
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.858884872311718e-06,
      "loss": 6.2691,
      "step": 78500
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.850555564810343e-06,
      "loss": 6.263,
      "step": 79000
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.842226257308968e-06,
      "loss": 6.2648,
      "step": 79500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.833896949807593e-06,
      "loss": 6.2682,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.1399281619793832,
      "eval_loss": 6.260497093200684,
      "eval_runtime": 1612.178,
      "eval_samples_per_second": 191.267,
      "eval_steps_per_second": 3.985,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.82556764230622e-06,
      "loss": 6.2653,
      "step": 80500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.817238334804844e-06,
      "loss": 6.2615,
      "step": 81000
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.808909027303471e-06,
      "loss": 6.2563,
      "step": 81500
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.800579719802096e-06,
      "loss": 6.2596,
      "step": 82000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.792250412300723e-06,
      "loss": 6.2583,
      "step": 82500
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.783921104799348e-06,
      "loss": 6.2509,
      "step": 83000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.775591797297973e-06,
      "loss": 6.2523,
      "step": 83500
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.767262489796598e-06,
      "loss": 6.252,
      "step": 84000
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.758933182295224e-06,
      "loss": 6.2415,
      "step": 84500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.750603874793851e-06,
      "loss": 6.2488,
      "step": 85000
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.742274567292476e-06,
      "loss": 6.2427,
      "step": 85500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.733945259791101e-06,
      "loss": 6.2487,
      "step": 86000
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.725615952289728e-06,
      "loss": 6.2364,
      "step": 86500
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.717286644788353e-06,
      "loss": 6.2351,
      "step": 87000
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.708957337286978e-06,
      "loss": 6.2459,
      "step": 87500
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.700628029785605e-06,
      "loss": 6.2361,
      "step": 88000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.69229872228423e-06,
      "loss": 6.2371,
      "step": 88500
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.683969414782856e-06,
      "loss": 6.2384,
      "step": 89000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.675640107281481e-06,
      "loss": 6.2287,
      "step": 89500
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.667310799780106e-06,
      "loss": 6.2352,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.1411253882983398,
      "eval_loss": 6.230065822601318,
      "eval_runtime": 1606.8691,
      "eval_samples_per_second": 191.899,
      "eval_steps_per_second": 3.998,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.658981492278733e-06,
      "loss": 6.233,
      "step": 90500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.650652184777358e-06,
      "loss": 6.2279,
      "step": 91000
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.642322877275983e-06,
      "loss": 6.2259,
      "step": 91500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.63399356977461e-06,
      "loss": 6.2287,
      "step": 92000
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.625664262273236e-06,
      "loss": 6.2283,
      "step": 92500
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.617334954771861e-06,
      "loss": 6.2256,
      "step": 93000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.609005647270486e-06,
      "loss": 6.2266,
      "step": 93500
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.600676339769111e-06,
      "loss": 6.2232,
      "step": 94000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.592347032267738e-06,
      "loss": 6.2206,
      "step": 94500
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.584017724766363e-06,
      "loss": 6.2108,
      "step": 95000
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.57568841726499e-06,
      "loss": 6.2287,
      "step": 95500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.567359109763615e-06,
      "loss": 6.2203,
      "step": 96000
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.559029802262241e-06,
      "loss": 6.2146,
      "step": 96500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.550700494760866e-06,
      "loss": 6.2202,
      "step": 97000
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.542371187259491e-06,
      "loss": 6.2101,
      "step": 97500
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.534041879758118e-06,
      "loss": 6.2087,
      "step": 98000
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.525712572256743e-06,
      "loss": 6.2154,
      "step": 98500
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.517383264755368e-06,
      "loss": 6.213,
      "step": 99000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.509053957253995e-06,
      "loss": 6.2128,
      "step": 99500
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.500724649752622e-06,
      "loss": 6.214,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "eval_accuracy": 0.14302454097126002,
      "eval_loss": 6.2089948654174805,
      "eval_runtime": 1605.1836,
      "eval_samples_per_second": 192.1,
      "eval_steps_per_second": 4.003,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.492395342251247e-06,
      "loss": 6.2073,
      "step": 100500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.484066034749872e-06,
      "loss": 6.2018,
      "step": 101000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.475736727248497e-06,
      "loss": 6.2041,
      "step": 101500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.467407419747123e-06,
      "loss": 6.2018,
      "step": 102000
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.459078112245748e-06,
      "loss": 6.2067,
      "step": 102500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.450748804744375e-06,
      "loss": 6.2007,
      "step": 103000
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.442419497243e-06,
      "loss": 6.203,
      "step": 103500
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.434090189741627e-06,
      "loss": 6.2016,
      "step": 104000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.425760882240252e-06,
      "loss": 6.1997,
      "step": 104500
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.417431574738877e-06,
      "loss": 6.1931,
      "step": 105000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.409102267237502e-06,
      "loss": 6.1906,
      "step": 105500
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.400772959736128e-06,
      "loss": 6.1953,
      "step": 106000
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.392443652234753e-06,
      "loss": 6.1989,
      "step": 106500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.38411434473338e-06,
      "loss": 6.2008,
      "step": 107000
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.375785037232005e-06,
      "loss": 6.1923,
      "step": 107500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.367455729730632e-06,
      "loss": 6.1895,
      "step": 108000
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.359126422229257e-06,
      "loss": 6.1903,
      "step": 108500
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.350797114727882e-06,
      "loss": 6.1927,
      "step": 109000
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.342467807226507e-06,
      "loss": 6.1884,
      "step": 109500
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.334138499725133e-06,
      "loss": 6.1837,
      "step": 110000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.14433779437099495,
      "eval_loss": 6.186517715454102,
      "eval_runtime": 1607.4607,
      "eval_samples_per_second": 191.828,
      "eval_steps_per_second": 3.997,
      "step": 110000
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.32580919222376e-06,
      "loss": 6.1838,
      "step": 110500
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.317479884722385e-06,
      "loss": 6.1885,
      "step": 111000
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.30915057722101e-06,
      "loss": 6.1905,
      "step": 111500
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.300821269719637e-06,
      "loss": 6.1885,
      "step": 112000
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.292491962218262e-06,
      "loss": 6.1866,
      "step": 112500
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.284162654716887e-06,
      "loss": 6.1846,
      "step": 113000
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.275833347215514e-06,
      "loss": 6.1803,
      "step": 113500
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.267504039714139e-06,
      "loss": 6.1877,
      "step": 114000
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.259174732212765e-06,
      "loss": 6.1766,
      "step": 114500
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.25084542471139e-06,
      "loss": 6.183,
      "step": 115000
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.242516117210015e-06,
      "loss": 6.1856,
      "step": 115500
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.234186809708642e-06,
      "loss": 6.1832,
      "step": 116000
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.225857502207267e-06,
      "loss": 6.1825,
      "step": 116500
    },
    {
      "epoch": 0.96,
      "learning_rate": 8.217528194705892e-06,
      "loss": 6.1767,
      "step": 117000
    },
    {
      "epoch": 0.96,
      "learning_rate": 8.209198887204519e-06,
      "loss": 6.1739,
      "step": 117500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.200869579703145e-06,
      "loss": 6.1742,
      "step": 118000
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.19254027220177e-06,
      "loss": 6.1734,
      "step": 118500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.184210964700395e-06,
      "loss": 6.184,
      "step": 119000
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.17588165719902e-06,
      "loss": 6.1747,
      "step": 119500
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.167552349697647e-06,
      "loss": 6.1726,
      "step": 120000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.14511534445770935,
      "eval_loss": 6.168183326721191,
      "eval_runtime": 1610.8389,
      "eval_samples_per_second": 191.426,
      "eval_steps_per_second": 3.989,
      "step": 120000
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.159223042196272e-06,
      "loss": 6.1655,
      "step": 120500
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.150893734694899e-06,
      "loss": 6.1762,
      "step": 121000
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.142564427193524e-06,
      "loss": 6.1687,
      "step": 121500
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.13423511969215e-06,
      "loss": 6.1684,
      "step": 122000
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.125905812190775e-06,
      "loss": 6.1652,
      "step": 122500
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.1175765046894e-06,
      "loss": 6.1614,
      "step": 123000
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.109247197188025e-06,
      "loss": 6.156,
      "step": 123500
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.100917889686652e-06,
      "loss": 6.1706,
      "step": 124000
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.092588582185277e-06,
      "loss": 6.1648,
      "step": 124500
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.084259274683904e-06,
      "loss": 6.1651,
      "step": 125000
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.075929967182529e-06,
      "loss": 6.1644,
      "step": 125500
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.067600659681156e-06,
      "loss": 6.1595,
      "step": 126000
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.05927135217978e-06,
      "loss": 6.1612,
      "step": 126500
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.050942044678406e-06,
      "loss": 6.1484,
      "step": 127000
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.04261273717703e-06,
      "loss": 6.157,
      "step": 127500
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.034283429675657e-06,
      "loss": 6.1555,
      "step": 128000
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.025954122174284e-06,
      "loss": 6.1507,
      "step": 128500
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.017624814672909e-06,
      "loss": 6.1634,
      "step": 129000
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.009295507171534e-06,
      "loss": 6.1528,
      "step": 129500
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.00096619967016e-06,
      "loss": 6.1524,
      "step": 130000
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.14635244683283277,
      "eval_loss": 6.149813652038574,
      "eval_runtime": 1608.6347,
      "eval_samples_per_second": 191.688,
      "eval_steps_per_second": 3.994,
      "step": 130000
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.992636892168786e-06,
      "loss": 6.1552,
      "step": 130500
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.98430758466741e-06,
      "loss": 6.1459,
      "step": 131000
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.975978277166037e-06,
      "loss": 6.1498,
      "step": 131500
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.967648969664662e-06,
      "loss": 6.1442,
      "step": 132000
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.959319662163289e-06,
      "loss": 6.1533,
      "step": 132500
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.950990354661914e-06,
      "loss": 6.1517,
      "step": 133000
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.942661047160539e-06,
      "loss": 6.1437,
      "step": 133500
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.934331739659166e-06,
      "loss": 6.1518,
      "step": 134000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.92600243215779e-06,
      "loss": 6.1485,
      "step": 134500
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.917673124656416e-06,
      "loss": 6.1399,
      "step": 135000
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.909343817155042e-06,
      "loss": 6.1445,
      "step": 135500
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.901014509653669e-06,
      "loss": 6.1384,
      "step": 136000
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.892685202152294e-06,
      "loss": 6.136,
      "step": 136500
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.884355894650919e-06,
      "loss": 6.1411,
      "step": 137000
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.876026587149544e-06,
      "loss": 6.1387,
      "step": 137500
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.86769727964817e-06,
      "loss": 6.1391,
      "step": 138000
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.859367972146796e-06,
      "loss": 6.1363,
      "step": 138500
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.851038664645423e-06,
      "loss": 6.1321,
      "step": 139000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.842709357144048e-06,
      "loss": 6.1389,
      "step": 139500
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.834380049642674e-06,
      "loss": 6.1293,
      "step": 140000
    },
    {
      "epoch": 1.15,
      "eval_accuracy": 0.14675081733675413,
      "eval_loss": 6.130023956298828,
      "eval_runtime": 1606.9582,
      "eval_samples_per_second": 191.888,
      "eval_steps_per_second": 3.998,
      "step": 140000
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.8260507421413e-06,
      "loss": 6.1245,
      "step": 140500
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.817721434639924e-06,
      "loss": 6.1362,
      "step": 141000
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.809392127138551e-06,
      "loss": 6.1254,
      "step": 141500
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.801062819637176e-06,
      "loss": 6.1254,
      "step": 142000
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.792733512135801e-06,
      "loss": 6.131,
      "step": 142500
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.784404204634428e-06,
      "loss": 6.1205,
      "step": 143000
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.776074897133054e-06,
      "loss": 6.1233,
      "step": 143500
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.76774558963168e-06,
      "loss": 6.1229,
      "step": 144000
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.759416282130304e-06,
      "loss": 6.1191,
      "step": 144500
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.75108697462893e-06,
      "loss": 6.1237,
      "step": 145000
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.742757667127556e-06,
      "loss": 6.1212,
      "step": 145500
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.734428359626181e-06,
      "loss": 6.1127,
      "step": 146000
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.726099052124808e-06,
      "loss": 6.116,
      "step": 146500
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.717769744623433e-06,
      "loss": 6.1189,
      "step": 147000
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.70944043712206e-06,
      "loss": 6.1169,
      "step": 147500
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.701111129620684e-06,
      "loss": 6.1094,
      "step": 148000
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.69278182211931e-06,
      "loss": 6.1152,
      "step": 148500
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.684452514617934e-06,
      "loss": 6.1079,
      "step": 149000
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.676123207116561e-06,
      "loss": 6.1085,
      "step": 149500
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.667793899615186e-06,
      "loss": 6.1116,
      "step": 150000
    },
    {
      "epoch": 1.23,
      "eval_accuracy": 0.14789823607983668,
      "eval_loss": 6.1025519371032715,
      "eval_runtime": 1607.6105,
      "eval_samples_per_second": 191.81,
      "eval_steps_per_second": 3.997,
      "step": 150000
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.659464592113813e-06,
      "loss": 6.106,
      "step": 150500
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.651135284612438e-06,
      "loss": 6.1075,
      "step": 151000
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.642805977111065e-06,
      "loss": 6.1057,
      "step": 151500
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.63447666960969e-06,
      "loss": 6.1048,
      "step": 152000
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.6261473621083145e-06,
      "loss": 6.1066,
      "step": 152500
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.61781805460694e-06,
      "loss": 6.101,
      "step": 153000
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.609488747105567e-06,
      "loss": 6.1037,
      "step": 153500
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.601159439604192e-06,
      "loss": 6.0998,
      "step": 154000
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.592830132102817e-06,
      "loss": 6.1011,
      "step": 154500
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.584500824601443e-06,
      "loss": 6.0939,
      "step": 155000
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.57617151710007e-06,
      "loss": 6.09,
      "step": 155500
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.567842209598695e-06,
      "loss": 6.0978,
      "step": 156000
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.5595129020973205e-06,
      "loss": 6.0897,
      "step": 156500
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.5511835945959455e-06,
      "loss": 6.0852,
      "step": 157000
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.542854287094572e-06,
      "loss": 6.0903,
      "step": 157500
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.534524979593197e-06,
      "loss": 6.0912,
      "step": 158000
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.526195672091823e-06,
      "loss": 6.0882,
      "step": 158500
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.517866364590448e-06,
      "loss": 6.0811,
      "step": 159000
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.509537057089075e-06,
      "loss": 6.0862,
      "step": 159500
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.5012077495877e-06,
      "loss": 6.0839,
      "step": 160000
    },
    {
      "epoch": 1.31,
      "eval_accuracy": 0.14901064810187634,
      "eval_loss": 6.079655170440674,
      "eval_runtime": 1607.0901,
      "eval_samples_per_second": 191.872,
      "eval_steps_per_second": 3.998,
      "step": 160000
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.492878442086326e-06,
      "loss": 6.0803,
      "step": 160500
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.484549134584951e-06,
      "loss": 6.08,
      "step": 161000
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.476219827083577e-06,
      "loss": 6.0849,
      "step": 161500
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.467890519582202e-06,
      "loss": 6.0769,
      "step": 162000
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.459561212080828e-06,
      "loss": 6.0888,
      "step": 162500
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.451231904579453e-06,
      "loss": 6.0808,
      "step": 163000
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.44290259707808e-06,
      "loss": 6.082,
      "step": 163500
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.434573289576706e-06,
      "loss": 6.0744,
      "step": 164000
    },
    {
      "epoch": 1.35,
      "learning_rate": 7.426243982075331e-06,
      "loss": 6.0707,
      "step": 164500
    },
    {
      "epoch": 1.35,
      "learning_rate": 7.417914674573956e-06,
      "loss": 6.0774,
      "step": 165000
    },
    {
      "epoch": 1.36,
      "learning_rate": 7.409585367072582e-06,
      "loss": 6.0699,
      "step": 165500
    },
    {
      "epoch": 1.36,
      "learning_rate": 7.401256059571208e-06,
      "loss": 6.0706,
      "step": 166000
    },
    {
      "epoch": 1.36,
      "learning_rate": 7.392926752069833e-06,
      "loss": 6.0643,
      "step": 166500
    },
    {
      "epoch": 1.37,
      "learning_rate": 7.384597444568459e-06,
      "loss": 6.0753,
      "step": 167000
    },
    {
      "epoch": 1.37,
      "learning_rate": 7.376268137067085e-06,
      "loss": 6.0738,
      "step": 167500
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.367938829565711e-06,
      "loss": 6.0664,
      "step": 168000
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.359609522064336e-06,
      "loss": 6.0662,
      "step": 168500
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.351280214562962e-06,
      "loss": 6.0658,
      "step": 169000
    },
    {
      "epoch": 1.39,
      "learning_rate": 7.3429509070615875e-06,
      "loss": 6.0655,
      "step": 169500
    },
    {
      "epoch": 1.39,
      "learning_rate": 7.334621599560213e-06,
      "loss": 6.0616,
      "step": 170000
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.14990009359739007,
      "eval_loss": 6.059047222137451,
      "eval_runtime": 1611.7462,
      "eval_samples_per_second": 191.318,
      "eval_steps_per_second": 3.986,
      "step": 170000
    },
    {
      "epoch": 1.4,
      "learning_rate": 7.326292292058838e-06,
      "loss": 6.0584,
      "step": 170500
    },
    {
      "epoch": 1.4,
      "learning_rate": 7.317962984557464e-06,
      "loss": 6.0612,
      "step": 171000
    },
    {
      "epoch": 1.41,
      "learning_rate": 7.309633677056091e-06,
      "loss": 6.0609,
      "step": 171500
    },
    {
      "epoch": 1.41,
      "learning_rate": 7.301304369554716e-06,
      "loss": 6.0555,
      "step": 172000
    },
    {
      "epoch": 1.41,
      "learning_rate": 7.292975062053341e-06,
      "loss": 6.0591,
      "step": 172500
    },
    {
      "epoch": 1.42,
      "learning_rate": 7.284645754551967e-06,
      "loss": 6.0575,
      "step": 173000
    },
    {
      "epoch": 1.42,
      "learning_rate": 7.2763164470505934e-06,
      "loss": 6.0613,
      "step": 173500
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.2679871395492184e-06,
      "loss": 6.055,
      "step": 174000
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.259657832047844e-06,
      "loss": 6.0562,
      "step": 174500
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.251328524546469e-06,
      "loss": 6.0579,
      "step": 175000
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.242999217045096e-06,
      "loss": 6.0557,
      "step": 175500
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.234669909543721e-06,
      "loss": 6.0571,
      "step": 176000
    },
    {
      "epoch": 1.45,
      "learning_rate": 7.226340602042347e-06,
      "loss": 6.0529,
      "step": 176500
    },
    {
      "epoch": 1.45,
      "learning_rate": 7.218011294540972e-06,
      "loss": 6.0521,
      "step": 177000
    },
    {
      "epoch": 1.45,
      "learning_rate": 7.2096819870395985e-06,
      "loss": 6.0502,
      "step": 177500
    },
    {
      "epoch": 1.46,
      "learning_rate": 7.2013526795382235e-06,
      "loss": 6.0525,
      "step": 178000
    },
    {
      "epoch": 1.46,
      "learning_rate": 7.193023372036849e-06,
      "loss": 6.0488,
      "step": 178500
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.184694064535474e-06,
      "loss": 6.0494,
      "step": 179000
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.176364757034101e-06,
      "loss": 6.0532,
      "step": 179500
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.168035449532726e-06,
      "loss": 6.0508,
      "step": 180000
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.15093336523337356,
      "eval_loss": 6.039904594421387,
      "eval_runtime": 1605.7246,
      "eval_samples_per_second": 192.035,
      "eval_steps_per_second": 4.001,
      "step": 180000
    },
    {
      "epoch": 1.48,
      "learning_rate": 7.159706142031352e-06,
      "loss": 6.0446,
      "step": 180500
    },
    {
      "epoch": 1.48,
      "learning_rate": 7.151376834529977e-06,
      "loss": 6.0378,
      "step": 181000
    },
    {
      "epoch": 1.49,
      "learning_rate": 7.143047527028604e-06,
      "loss": 6.0419,
      "step": 181500
    },
    {
      "epoch": 1.49,
      "learning_rate": 7.1347182195272295e-06,
      "loss": 6.043,
      "step": 182000
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.1263889120258545e-06,
      "loss": 6.0403,
      "step": 182500
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.118059604524481e-06,
      "loss": 6.0423,
      "step": 183000
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.109730297023106e-06,
      "loss": 6.0387,
      "step": 183500
    },
    {
      "epoch": 1.51,
      "learning_rate": 7.101400989521732e-06,
      "loss": 6.0434,
      "step": 184000
    },
    {
      "epoch": 1.51,
      "learning_rate": 7.093071682020357e-06,
      "loss": 6.041,
      "step": 184500
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.084742374518984e-06,
      "loss": 6.0311,
      "step": 185000
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.076413067017609e-06,
      "loss": 6.0378,
      "step": 185500
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.068083759516235e-06,
      "loss": 6.0342,
      "step": 186000
    },
    {
      "epoch": 1.53,
      "learning_rate": 7.05975445201486e-06,
      "loss": 6.0264,
      "step": 186500
    },
    {
      "epoch": 1.53,
      "learning_rate": 7.051425144513486e-06,
      "loss": 6.0371,
      "step": 187000
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.043095837012111e-06,
      "loss": 6.0353,
      "step": 187500
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.034766529510737e-06,
      "loss": 6.0325,
      "step": 188000
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.026437222009362e-06,
      "loss": 6.0289,
      "step": 188500
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.018107914507989e-06,
      "loss": 6.0332,
      "step": 189000
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.009778607006615e-06,
      "loss": 6.0328,
      "step": 189500
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.00144929950524e-06,
      "loss": 6.0311,
      "step": 190000
    },
    {
      "epoch": 1.56,
      "eval_accuracy": 0.15172702709984232,
      "eval_loss": 6.023284435272217,
      "eval_runtime": 1606.8679,
      "eval_samples_per_second": 191.899,
      "eval_steps_per_second": 3.998,
      "step": 190000
    },
    {
      "epoch": 1.56,
      "learning_rate": 6.993119992003865e-06,
      "loss": 6.0273,
      "step": 190500
    },
    {
      "epoch": 1.56,
      "learning_rate": 6.984790684502491e-06,
      "loss": 6.0271,
      "step": 191000
    },
    {
      "epoch": 1.57,
      "learning_rate": 6.976461377001117e-06,
      "loss": 6.024,
      "step": 191500
    },
    {
      "epoch": 1.57,
      "learning_rate": 6.968132069499742e-06,
      "loss": 6.0182,
      "step": 192000
    },
    {
      "epoch": 1.58,
      "learning_rate": 6.959802761998368e-06,
      "loss": 6.0259,
      "step": 192500
    },
    {
      "epoch": 1.58,
      "learning_rate": 6.951473454496994e-06,
      "loss": 6.0253,
      "step": 193000
    },
    {
      "epoch": 1.59,
      "learning_rate": 6.94314414699562e-06,
      "loss": 6.0209,
      "step": 193500
    },
    {
      "epoch": 1.59,
      "learning_rate": 6.934814839494245e-06,
      "loss": 6.0227,
      "step": 194000
    },
    {
      "epoch": 1.59,
      "learning_rate": 6.926485531992871e-06,
      "loss": 6.0271,
      "step": 194500
    },
    {
      "epoch": 1.6,
      "learning_rate": 6.9181562244914965e-06,
      "loss": 6.0201,
      "step": 195000
    },
    {
      "epoch": 1.6,
      "learning_rate": 6.909826916990122e-06,
      "loss": 6.0119,
      "step": 195500
    },
    {
      "epoch": 1.61,
      "learning_rate": 6.901497609488747e-06,
      "loss": 6.0168,
      "step": 196000
    },
    {
      "epoch": 1.61,
      "learning_rate": 6.893168301987373e-06,
      "loss": 6.0167,
      "step": 196500
    },
    {
      "epoch": 1.61,
      "learning_rate": 6.884838994485999e-06,
      "loss": 6.0213,
      "step": 197000
    },
    {
      "epoch": 1.62,
      "learning_rate": 6.876509686984625e-06,
      "loss": 6.0122,
      "step": 197500
    },
    {
      "epoch": 1.62,
      "learning_rate": 6.86818037948325e-06,
      "loss": 6.0152,
      "step": 198000
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.859851071981876e-06,
      "loss": 6.0136,
      "step": 198500
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.8515217644805024e-06,
      "loss": 6.0163,
      "step": 199000
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.8431924569791274e-06,
      "loss": 6.0119,
      "step": 199500
    },
    {
      "epoch": 1.64,
      "learning_rate": 6.8348631494777525e-06,
      "loss": 6.015,
      "step": 200000
    },
    {
      "epoch": 1.64,
      "eval_accuracy": 0.15330433087040413,
      "eval_loss": 6.004796981811523,
      "eval_runtime": 1605.7958,
      "eval_samples_per_second": 192.027,
      "eval_steps_per_second": 4.001,
      "step": 200000
    },
    {
      "epoch": 1.64,
      "learning_rate": 6.826533841976378e-06,
      "loss": 6.0116,
      "step": 200500
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.818204534475005e-06,
      "loss": 6.0154,
      "step": 201000
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.80987522697363e-06,
      "loss": 6.0078,
      "step": 201500
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.801545919472256e-06,
      "loss": 6.0092,
      "step": 202000
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.793216611970881e-06,
      "loss": 6.006,
      "step": 202500
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.7848873044695075e-06,
      "loss": 6.0057,
      "step": 203000
    },
    {
      "epoch": 1.67,
      "learning_rate": 6.7765579969681325e-06,
      "loss": 6.0042,
      "step": 203500
    },
    {
      "epoch": 1.67,
      "learning_rate": 6.768228689466758e-06,
      "loss": 6.0061,
      "step": 204000
    },
    {
      "epoch": 1.68,
      "learning_rate": 6.759899381965383e-06,
      "loss": 5.9964,
      "step": 204500
    },
    {
      "epoch": 1.68,
      "learning_rate": 6.75157007446401e-06,
      "loss": 6.0046,
      "step": 205000
    },
    {
      "epoch": 1.68,
      "learning_rate": 6.743240766962635e-06,
      "loss": 5.9985,
      "step": 205500
    },
    {
      "epoch": 1.69,
      "learning_rate": 6.734911459461261e-06,
      "loss": 6.007,
      "step": 206000
    },
    {
      "epoch": 1.69,
      "learning_rate": 6.726582151959886e-06,
      "loss": 6.0006,
      "step": 206500
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.718252844458513e-06,
      "loss": 6.0019,
      "step": 207000
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.709923536957138e-06,
      "loss": 5.9919,
      "step": 207500
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.7015942294557635e-06,
      "loss": 5.9976,
      "step": 208000
    },
    {
      "epoch": 1.71,
      "learning_rate": 6.6932649219543885e-06,
      "loss": 5.992,
      "step": 208500
    },
    {
      "epoch": 1.71,
      "learning_rate": 6.684935614453015e-06,
      "loss": 5.9965,
      "step": 209000
    },
    {
      "epoch": 1.72,
      "learning_rate": 6.676606306951641e-06,
      "loss": 5.9967,
      "step": 209500
    },
    {
      "epoch": 1.72,
      "learning_rate": 6.668276999450266e-06,
      "loss": 5.985,
      "step": 210000
    },
    {
      "epoch": 1.72,
      "eval_accuracy": 0.15468790550565392,
      "eval_loss": 5.986328125,
      "eval_runtime": 1607.6064,
      "eval_samples_per_second": 191.811,
      "eval_steps_per_second": 3.997,
      "step": 210000
    },
    {
      "epoch": 1.72,
      "learning_rate": 6.659947691948891e-06,
      "loss": 5.9893,
      "step": 210500
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.651618384447518e-06,
      "loss": 5.9891,
      "step": 211000
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.643289076946144e-06,
      "loss": 5.992,
      "step": 211500
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.634959769444769e-06,
      "loss": 5.9876,
      "step": 212000
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.6266304619433945e-06,
      "loss": 5.9899,
      "step": 212500
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.61830115444202e-06,
      "loss": 5.9888,
      "step": 213000
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.609971846940646e-06,
      "loss": 5.9844,
      "step": 213500
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.601642539439271e-06,
      "loss": 5.9861,
      "step": 214000
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.593313231937897e-06,
      "loss": 5.9838,
      "step": 214500
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.584983924436523e-06,
      "loss": 5.9799,
      "step": 215000
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.576654616935149e-06,
      "loss": 5.9758,
      "step": 215500
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.568325309433774e-06,
      "loss": 5.9844,
      "step": 216000
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.5599960019323996e-06,
      "loss": 5.9741,
      "step": 216500
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.551666694431026e-06,
      "loss": 5.9756,
      "step": 217000
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.543337386929651e-06,
      "loss": 5.975,
      "step": 217500
    },
    {
      "epoch": 1.79,
      "learning_rate": 6.535008079428276e-06,
      "loss": 5.9748,
      "step": 218000
    },
    {
      "epoch": 1.79,
      "learning_rate": 6.526678771926902e-06,
      "loss": 5.9685,
      "step": 218500
    },
    {
      "epoch": 1.79,
      "learning_rate": 6.518349464425529e-06,
      "loss": 5.9644,
      "step": 219000
    },
    {
      "epoch": 1.8,
      "learning_rate": 6.510020156924154e-06,
      "loss": 5.9698,
      "step": 219500
    },
    {
      "epoch": 1.8,
      "learning_rate": 6.50169084942278e-06,
      "loss": 5.9661,
      "step": 220000
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.1573209769958548,
      "eval_loss": 5.959463119506836,
      "eval_runtime": 1609.5369,
      "eval_samples_per_second": 191.581,
      "eval_steps_per_second": 3.992,
      "step": 220000
    },
    {
      "epoch": 1.81,
      "learning_rate": 6.493361541921405e-06,
      "loss": 5.9707,
      "step": 220500
    },
    {
      "epoch": 1.81,
      "learning_rate": 6.485032234420031e-06,
      "loss": 5.9604,
      "step": 221000
    },
    {
      "epoch": 1.81,
      "learning_rate": 6.476702926918656e-06,
      "loss": 5.9645,
      "step": 221500
    },
    {
      "epoch": 1.82,
      "learning_rate": 6.468373619417282e-06,
      "loss": 5.9629,
      "step": 222000
    },
    {
      "epoch": 1.82,
      "learning_rate": 6.460044311915907e-06,
      "loss": 5.9583,
      "step": 222500
    },
    {
      "epoch": 1.83,
      "learning_rate": 6.451715004414534e-06,
      "loss": 5.9663,
      "step": 223000
    },
    {
      "epoch": 1.83,
      "learning_rate": 6.443385696913159e-06,
      "loss": 5.9534,
      "step": 223500
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.435056389411785e-06,
      "loss": 5.957,
      "step": 224000
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.42672708191041e-06,
      "loss": 5.9581,
      "step": 224500
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.4183977744090365e-06,
      "loss": 5.9619,
      "step": 225000
    },
    {
      "epoch": 1.85,
      "learning_rate": 6.4100684669076615e-06,
      "loss": 5.9522,
      "step": 225500
    },
    {
      "epoch": 1.85,
      "learning_rate": 6.401739159406287e-06,
      "loss": 5.9561,
      "step": 226000
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.393409851904914e-06,
      "loss": 5.9523,
      "step": 226500
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.385080544403539e-06,
      "loss": 5.9429,
      "step": 227000
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.376751236902165e-06,
      "loss": 5.9474,
      "step": 227500
    },
    {
      "epoch": 1.87,
      "learning_rate": 6.36842192940079e-06,
      "loss": 5.9505,
      "step": 228000
    },
    {
      "epoch": 1.87,
      "step": 228437,
      "total_flos": 3.158360285333422e+18,
      "train_loss": 6.326340861561735,
      "train_runtime": 198000.3394,
      "train_samples_per_second": 147.948,
      "train_steps_per_second": 3.082
    }
  ],
  "logging_steps": 500,
  "max_steps": 610290,
  "num_train_epochs": 5,
  "save_steps": 10000,
  "total_flos": 3.158360285333422e+18,
  "trial_name": null,
  "trial_params": null
}