{
  "best_metric": 2.9055657386779785,
  "best_model_checkpoint": "bert_12_layer_model_v2_complete_training_new_wt_init/checkpoint-110000",
  "epoch": 0.9012190925625523,
  "global_step": 110001,
  "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": 9.3837,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 7.5515,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5e-06,
      "loss": 7.3375,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 7.2338,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.5e-06,
      "loss": 7.1692,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "learning_rate": 3e-06,
      "loss": 7.0812,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.5e-06,
      "loss": 6.9798,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.000000000000001e-06,
      "loss": 6.906,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.5e-06,
      "loss": 6.8266,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-06,
      "loss": 6.7704,
      "step": 5000
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.500000000000001e-06,
      "loss": 6.7049,
      "step": 5500
    },
    {
      "epoch": 0.05,
      "learning_rate": 6e-06,
      "loss": 6.6487,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 6.6085,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "learning_rate": 7e-06,
      "loss": 6.5664,
      "step": 7000
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.500000000000001e-06,
      "loss": 6.5347,
      "step": 7500
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.000000000000001e-06,
      "loss": 6.4952,
      "step": 8000
    },
    {
      "epoch": 0.07,
      "learning_rate": 8.5e-06,
      "loss": 6.4644,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "learning_rate": 9e-06,
      "loss": 6.443,
      "step": 9000
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.5e-06,
      "loss": 6.4184,
      "step": 9500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1e-05,
      "loss": 6.4002,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.131173745590539,
      "eval_loss": 6.357146739959717,
      "eval_runtime": 1773.2677,
      "eval_samples_per_second": 173.891,
      "eval_steps_per_second": 3.623,
      "step": 10000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.991670692498626e-06,
      "loss": 6.3796,
      "step": 10500
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.983341384997253e-06,
      "loss": 6.3659,
      "step": 11000
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.975012077495878e-06,
      "loss": 6.3464,
      "step": 11500
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.966682769994504e-06,
      "loss": 6.3316,
      "step": 12000
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.95835346249313e-06,
      "loss": 6.3095,
      "step": 12500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.950024154991756e-06,
      "loss": 6.3008,
      "step": 13000
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.941694847490381e-06,
      "loss": 6.2896,
      "step": 13500
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.933365539989006e-06,
      "loss": 6.2792,
      "step": 14000
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.925036232487631e-06,
      "loss": 6.2623,
      "step": 14500
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.916706924986258e-06,
      "loss": 6.2537,
      "step": 15000
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.908377617484883e-06,
      "loss": 6.2372,
      "step": 15500
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.90004830998351e-06,
      "loss": 6.2354,
      "step": 16000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.891719002482134e-06,
      "loss": 6.2328,
      "step": 16500
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.883389694980761e-06,
      "loss": 6.2136,
      "step": 17000
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.875060387479386e-06,
      "loss": 6.2083,
      "step": 17500
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.866731079978011e-06,
      "loss": 6.1813,
      "step": 18000
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.858401772476636e-06,
      "loss": 6.1418,
      "step": 18500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.850072464975263e-06,
      "loss": 6.055,
      "step": 19000
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.84174315747389e-06,
      "loss": 5.8363,
      "step": 19500
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.833413849972514e-06,
      "loss": 5.5302,
      "step": 20000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.24268464050725605,
      "eval_loss": 5.188525676727295,
      "eval_runtime": 1774.1484,
      "eval_samples_per_second": 173.805,
      "eval_steps_per_second": 3.621,
      "step": 20000
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.82508454247114e-06,
      "loss": 5.1932,
      "step": 20500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.816755234969766e-06,
      "loss": 4.9775,
      "step": 21000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.808425927468391e-06,
      "loss": 4.8531,
      "step": 21500
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.800096619967016e-06,
      "loss": 4.7375,
      "step": 22000
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.791767312465643e-06,
      "loss": 4.6615,
      "step": 22500
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.783438004964268e-06,
      "loss": 4.58,
      "step": 23000
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.775108697462895e-06,
      "loss": 4.5249,
      "step": 23500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.76677938996152e-06,
      "loss": 4.4636,
      "step": 24000
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.758450082460145e-06,
      "loss": 4.4196,
      "step": 24500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.750120774958771e-06,
      "loss": 4.3688,
      "step": 25000
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.741791467457396e-06,
      "loss": 4.3229,
      "step": 25500
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.733462159956021e-06,
      "loss": 4.288,
      "step": 26000
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.725132852454648e-06,
      "loss": 4.259,
      "step": 26500
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.716803544953275e-06,
      "loss": 4.2195,
      "step": 27000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.7084742374519e-06,
      "loss": 4.1809,
      "step": 27500
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.700144929950525e-06,
      "loss": 4.1484,
      "step": 28000
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.69181562244915e-06,
      "loss": 4.1208,
      "step": 28500
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.683486314947776e-06,
      "loss": 4.0918,
      "step": 29000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.675157007446401e-06,
      "loss": 4.0612,
      "step": 29500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.666827699945028e-06,
      "loss": 4.04,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.38633637025458273,
      "eval_loss": 3.807072401046753,
      "eval_runtime": 1776.8144,
      "eval_samples_per_second": 173.544,
      "eval_steps_per_second": 3.616,
      "step": 30000
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.658498392443653e-06,
      "loss": 4.0116,
      "step": 30500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.65016908494228e-06,
      "loss": 3.9919,
      "step": 31000
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.641839777440905e-06,
      "loss": 3.975,
      "step": 31500
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.63351046993953e-06,
      "loss": 3.9507,
      "step": 32000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.625181162438155e-06,
      "loss": 3.9299,
      "step": 32500
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.616851854936781e-06,
      "loss": 3.9127,
      "step": 33000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.608522547435406e-06,
      "loss": 3.8914,
      "step": 33500
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.600193239934033e-06,
      "loss": 3.8767,
      "step": 34000
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.591863932432658e-06,
      "loss": 3.8565,
      "step": 34500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.583534624931285e-06,
      "loss": 3.8429,
      "step": 35000
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.57520531742991e-06,
      "loss": 3.8354,
      "step": 35500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.566876009928535e-06,
      "loss": 3.8166,
      "step": 36000
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.55854670242716e-06,
      "loss": 3.8012,
      "step": 36500
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.550217394925787e-06,
      "loss": 3.7845,
      "step": 37000
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.541888087424413e-06,
      "loss": 3.7705,
      "step": 37500
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.533558779923038e-06,
      "loss": 3.7533,
      "step": 38000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.525229472421663e-06,
      "loss": 3.7372,
      "step": 38500
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.51690016492029e-06,
      "loss": 3.7276,
      "step": 39000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.508570857418915e-06,
      "loss": 3.7208,
      "step": 39500
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.50024154991754e-06,
      "loss": 3.7185,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.4245870635744566,
      "eval_loss": 3.477015256881714,
      "eval_runtime": 1790.9757,
      "eval_samples_per_second": 172.172,
      "eval_steps_per_second": 3.587,
      "step": 40000
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.491912242416167e-06,
      "loss": 3.6981,
      "step": 40500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.483582934914792e-06,
      "loss": 3.6868,
      "step": 41000
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.475253627413418e-06,
      "loss": 3.6754,
      "step": 41500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.466924319912043e-06,
      "loss": 3.6727,
      "step": 42000
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.458595012410668e-06,
      "loss": 3.6546,
      "step": 42500
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.450265704909295e-06,
      "loss": 3.6451,
      "step": 43000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.44193639740792e-06,
      "loss": 3.6305,
      "step": 43500
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.433607089906545e-06,
      "loss": 3.6316,
      "step": 44000
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.425277782405172e-06,
      "loss": 3.6172,
      "step": 44500
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.416948474903798e-06,
      "loss": 3.6128,
      "step": 45000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.408619167402423e-06,
      "loss": 3.6058,
      "step": 45500
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.400289859901048e-06,
      "loss": 3.5869,
      "step": 46000
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.391960552399673e-06,
      "loss": 3.5816,
      "step": 46500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.3836312448983e-06,
      "loss": 3.57,
      "step": 47000
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.375301937396925e-06,
      "loss": 3.5669,
      "step": 47500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.366972629895552e-06,
      "loss": 3.5581,
      "step": 48000
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.358643322394177e-06,
      "loss": 3.5509,
      "step": 48500
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.350314014892804e-06,
      "loss": 3.54,
      "step": 49000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.341984707391429e-06,
      "loss": 3.5367,
      "step": 49500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.333655399890054e-06,
      "loss": 3.5317,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.4440765725047366,
      "eval_loss": 3.3048713207244873,
      "eval_runtime": 1778.9922,
      "eval_samples_per_second": 173.332,
      "eval_steps_per_second": 3.612,
      "step": 50000
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.325326092388679e-06,
      "loss": 3.5202,
      "step": 50500
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.316996784887305e-06,
      "loss": 3.5179,
      "step": 51000
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.30866747738593e-06,
      "loss": 3.5111,
      "step": 51500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.300338169884557e-06,
      "loss": 3.5022,
      "step": 52000
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.292008862383182e-06,
      "loss": 3.4971,
      "step": 52500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.283679554881809e-06,
      "loss": 3.4898,
      "step": 53000
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.275350247380434e-06,
      "loss": 3.4841,
      "step": 53500
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.267020939879059e-06,
      "loss": 3.472,
      "step": 54000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.258691632377685e-06,
      "loss": 3.4732,
      "step": 54500
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.25036232487631e-06,
      "loss": 3.462,
      "step": 55000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.242033017374937e-06,
      "loss": 3.4614,
      "step": 55500
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.233703709873562e-06,
      "loss": 3.4485,
      "step": 56000
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.225374402372189e-06,
      "loss": 3.4455,
      "step": 56500
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.217045094870814e-06,
      "loss": 3.434,
      "step": 57000
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.208715787369439e-06,
      "loss": 3.436,
      "step": 57500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.200386479868064e-06,
      "loss": 3.4277,
      "step": 58000
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.19205717236669e-06,
      "loss": 3.4214,
      "step": 58500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.183727864865315e-06,
      "loss": 3.4203,
      "step": 59000
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.175398557363942e-06,
      "loss": 3.4182,
      "step": 59500
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.167069249862567e-06,
      "loss": 3.4184,
      "step": 60000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.4558045032888814,
      "eval_loss": 3.1983070373535156,
      "eval_runtime": 1779.243,
      "eval_samples_per_second": 173.307,
      "eval_steps_per_second": 3.611,
      "step": 60000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.158739942361194e-06,
      "loss": 3.402,
      "step": 60500
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.150410634859819e-06,
      "loss": 3.3935,
      "step": 61000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.142081327358444e-06,
      "loss": 3.3958,
      "step": 61500
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.133752019857069e-06,
      "loss": 3.3797,
      "step": 62000
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.125422712355696e-06,
      "loss": 3.384,
      "step": 62500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.11709340485432e-06,
      "loss": 3.3799,
      "step": 63000
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.108764097352947e-06,
      "loss": 3.3649,
      "step": 63500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.100434789851572e-06,
      "loss": 3.3658,
      "step": 64000
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.092105482350199e-06,
      "loss": 3.3604,
      "step": 64500
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.083776174848824e-06,
      "loss": 3.3555,
      "step": 65000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.075446867347449e-06,
      "loss": 3.3608,
      "step": 65500
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.067117559846076e-06,
      "loss": 3.3513,
      "step": 66000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.0587882523447e-06,
      "loss": 3.3411,
      "step": 66500
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.050458944843327e-06,
      "loss": 3.3318,
      "step": 67000
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.042129637341952e-06,
      "loss": 3.3344,
      "step": 67500
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.033800329840577e-06,
      "loss": 3.3314,
      "step": 68000
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.025471022339204e-06,
      "loss": 3.3322,
      "step": 68500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.017141714837829e-06,
      "loss": 3.3217,
      "step": 69000
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.008812407336454e-06,
      "loss": 3.3274,
      "step": 69500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.00048309983508e-06,
      "loss": 3.3161,
      "step": 70000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.4649789254202149,
      "eval_loss": 3.121884822845459,
      "eval_runtime": 1775.2672,
      "eval_samples_per_second": 173.696,
      "eval_steps_per_second": 3.619,
      "step": 70000
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.992153792333706e-06,
      "loss": 3.3057,
      "step": 70500
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.983824484832332e-06,
      "loss": 3.3017,
      "step": 71000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.975495177330957e-06,
      "loss": 3.3071,
      "step": 71500
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.967165869829582e-06,
      "loss": 3.2954,
      "step": 72000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.958836562328209e-06,
      "loss": 3.2954,
      "step": 72500
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.950507254826834e-06,
      "loss": 3.285,
      "step": 73000
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.942177947325459e-06,
      "loss": 3.2896,
      "step": 73500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.933848639824086e-06,
      "loss": 3.2804,
      "step": 74000
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.925519332322713e-06,
      "loss": 3.2784,
      "step": 74500
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.917190024821338e-06,
      "loss": 3.2786,
      "step": 75000
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.908860717319963e-06,
      "loss": 3.2695,
      "step": 75500
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.900531409818588e-06,
      "loss": 3.265,
      "step": 76000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.892202102317214e-06,
      "loss": 3.2599,
      "step": 76500
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.88387279481584e-06,
      "loss": 3.2542,
      "step": 77000
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.875543487314466e-06,
      "loss": 3.2535,
      "step": 77500
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.867214179813091e-06,
      "loss": 3.2447,
      "step": 78000
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.858884872311718e-06,
      "loss": 3.2502,
      "step": 78500
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.850555564810343e-06,
      "loss": 3.2402,
      "step": 79000
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.842226257308968e-06,
      "loss": 3.2471,
      "step": 79500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.833896949807593e-06,
      "loss": 3.2417,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.47264960931951194,
      "eval_loss": 3.051072120666504,
      "eval_runtime": 1774.7293,
      "eval_samples_per_second": 173.748,
      "eval_steps_per_second": 3.62,
      "step": 80000
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.82556764230622e-06,
      "loss": 3.237,
      "step": 80500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.817238334804844e-06,
      "loss": 3.2342,
      "step": 81000
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.808909027303471e-06,
      "loss": 3.2275,
      "step": 81500
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.800579719802096e-06,
      "loss": 3.2233,
      "step": 82000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.792250412300723e-06,
      "loss": 3.2102,
      "step": 82500
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.783921104799348e-06,
      "loss": 3.2215,
      "step": 83000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.775591797297973e-06,
      "loss": 3.2177,
      "step": 83500
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.767262489796598e-06,
      "loss": 3.2137,
      "step": 84000
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.758933182295224e-06,
      "loss": 3.2057,
      "step": 84500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.750603874793851e-06,
      "loss": 3.2044,
      "step": 85000
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.742274567292476e-06,
      "loss": 3.2064,
      "step": 85500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.733945259791101e-06,
      "loss": 3.1964,
      "step": 86000
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.725615952289728e-06,
      "loss": 3.1985,
      "step": 86500
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.717286644788353e-06,
      "loss": 3.1973,
      "step": 87000
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.708957337286978e-06,
      "loss": 3.1906,
      "step": 87500
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.700628029785605e-06,
      "loss": 3.1869,
      "step": 88000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.69229872228423e-06,
      "loss": 3.1878,
      "step": 88500
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.683969414782856e-06,
      "loss": 3.1825,
      "step": 89000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.675640107281481e-06,
      "loss": 3.1757,
      "step": 89500
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.667310799780106e-06,
      "loss": 3.1771,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.478947317395192,
      "eval_loss": 2.99343204498291,
      "eval_runtime": 1777.6294,
      "eval_samples_per_second": 173.465,
      "eval_steps_per_second": 3.614,
      "step": 90000
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.658981492278733e-06,
      "loss": 3.1692,
      "step": 90500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.650652184777358e-06,
      "loss": 3.1725,
      "step": 91000
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.642322877275983e-06,
      "loss": 3.1707,
      "step": 91500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.63399356977461e-06,
      "loss": 3.1731,
      "step": 92000
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.625664262273236e-06,
      "loss": 3.1641,
      "step": 92500
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.617334954771861e-06,
      "loss": 3.1672,
      "step": 93000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.609005647270486e-06,
      "loss": 3.1546,
      "step": 93500
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.600676339769111e-06,
      "loss": 3.1527,
      "step": 94000
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.592347032267738e-06,
      "loss": 3.1567,
      "step": 94500
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.584017724766363e-06,
      "loss": 3.1501,
      "step": 95000
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.57568841726499e-06,
      "loss": 3.1509,
      "step": 95500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.567359109763615e-06,
      "loss": 3.1496,
      "step": 96000
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.559029802262241e-06,
      "loss": 3.1443,
      "step": 96500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.550700494760866e-06,
      "loss": 3.1406,
      "step": 97000
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.542371187259491e-06,
      "loss": 3.1413,
      "step": 97500
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.534041879758118e-06,
      "loss": 3.139,
      "step": 98000
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.525712572256743e-06,
      "loss": 3.1429,
      "step": 98500
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.517383264755368e-06,
      "loss": 3.1361,
      "step": 99000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.509053957253995e-06,
      "loss": 3.1263,
      "step": 99500
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.500724649752622e-06,
      "loss": 3.1276,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "eval_accuracy": 0.4849775759736167,
      "eval_loss": 2.945034980773926,
      "eval_runtime": 1774.8372,
      "eval_samples_per_second": 173.738,
      "eval_steps_per_second": 3.62,
      "step": 100000
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.492395342251247e-06,
      "loss": 3.1236,
      "step": 100500
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.484066034749872e-06,
      "loss": 3.1159,
      "step": 101000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.475736727248497e-06,
      "loss": 3.1237,
      "step": 101500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.467407419747123e-06,
      "loss": 3.1164,
      "step": 102000
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.459078112245748e-06,
      "loss": 3.114,
      "step": 102500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.450748804744375e-06,
      "loss": 3.1142,
      "step": 103000
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.442419497243e-06,
      "loss": 3.1087,
      "step": 103500
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.434090189741627e-06,
      "loss": 3.1037,
      "step": 104000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.425760882240252e-06,
      "loss": 3.1154,
      "step": 104500
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.417431574738877e-06,
      "loss": 3.1043,
      "step": 105000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.409102267237502e-06,
      "loss": 3.1092,
      "step": 105500
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.400772959736128e-06,
      "loss": 3.1004,
      "step": 106000
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.392443652234753e-06,
      "loss": 3.0983,
      "step": 106500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.38411434473338e-06,
      "loss": 3.0877,
      "step": 107000
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.375785037232005e-06,
      "loss": 3.0864,
      "step": 107500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.367455729730632e-06,
      "loss": 3.0938,
      "step": 108000
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.359126422229257e-06,
      "loss": 3.0869,
      "step": 108500
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.350797114727882e-06,
      "loss": 3.0846,
      "step": 109000
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.342467807226507e-06,
      "loss": 3.084,
      "step": 109500
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.334138499725133e-06,
      "loss": 3.0795,
      "step": 110000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.48951067327044,
      "eval_loss": 2.9055657386779785,
      "eval_runtime": 1776.964,
      "eval_samples_per_second": 173.53,
      "eval_steps_per_second": 3.616,
      "step": 110000
    },
    {
      "epoch": 0.9,
      "step": 110001,
      "total_flos": 1.545850768759718e+18,
      "train_loss": 4.050417150244161,
      "train_runtime": 109133.2709,
      "train_samples_per_second": 268.422,
      "train_steps_per_second": 5.592
    }
  ],
  "max_steps": 610290,
  "num_train_epochs": 5,
  "total_flos": 1.545850768759718e+18,
  "trial_name": null,
  "trial_params": null
}