{
  "best_metric": 1.0465439558029175,
  "best_model_checkpoint": "output/bertlawbr/checkpoint-227500",
  "epoch": 20.0,
  "global_step": 228880,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 5e-06,
      "loss": 9.0906,
      "step": 500
    },
    {
      "epoch": 0.09,
      "learning_rate": 1e-05,
      "loss": 7.3178,
      "step": 1000
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.5e-05,
      "loss": 6.5353,
      "step": 1500
    },
    {
      "epoch": 0.17,
      "learning_rate": 2e-05,
      "loss": 6.3111,
      "step": 2000
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.5e-05,
      "loss": 6.1291,
      "step": 2500
    },
    {
      "epoch": 0.22,
      "eval_loss": 5.988766670227051,
      "eval_runtime": 683.0879,
      "eval_samples_per_second": 112.864,
      "eval_steps_per_second": 7.055,
      "step": 2500
    },
    {
      "epoch": 0.26,
      "learning_rate": 3e-05,
      "loss": 5.9734,
      "step": 3000
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.5e-05,
      "loss": 5.797,
      "step": 3500
    },
    {
      "epoch": 0.35,
      "learning_rate": 4e-05,
      "loss": 5.5955,
      "step": 4000
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.5e-05,
      "loss": 5.2733,
      "step": 4500
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 4.8604,
      "step": 5000
    },
    {
      "epoch": 0.44,
      "eval_loss": 4.484074115753174,
      "eval_runtime": 681.6714,
      "eval_samples_per_second": 113.098,
      "eval_steps_per_second": 7.069,
      "step": 5000
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.500000000000001e-05,
      "loss": 4.4273,
      "step": 5500
    },
    {
      "epoch": 0.52,
      "learning_rate": 6e-05,
      "loss": 4.049,
      "step": 6000
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.500000000000001e-05,
      "loss": 3.7574,
      "step": 6500
    },
    {
      "epoch": 0.61,
      "learning_rate": 7e-05,
      "loss": 3.5134,
      "step": 7000
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.500000000000001e-05,
      "loss": 3.3321,
      "step": 7500
    },
    {
      "epoch": 0.66,
      "eval_loss": 3.118971586227417,
      "eval_runtime": 682.3171,
      "eval_samples_per_second": 112.991,
      "eval_steps_per_second": 7.063,
      "step": 7500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8e-05,
      "loss": 3.1651,
      "step": 8000
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.5e-05,
      "loss": 3.0254,
      "step": 8500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.999000000000001e-05,
      "loss": 2.9285,
      "step": 9000
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.499e-05,
      "loss": 2.8345,
      "step": 9500
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.999000000000001e-05,
      "loss": 2.7579,
      "step": 10000
    },
    {
      "epoch": 0.87,
      "eval_loss": 2.608943223953247,
      "eval_runtime": 683.6111,
      "eval_samples_per_second": 112.778,
      "eval_steps_per_second": 7.049,
      "step": 10000
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.89432443879712e-05,
      "loss": 2.6686,
      "step": 10500
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.788437102922491e-05,
      "loss": 2.597,
      "step": 11000
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.682549767047862e-05,
      "loss": 2.5214,
      "step": 11500
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.576662431173232e-05,
      "loss": 2.4561,
      "step": 12000
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.470775095298603e-05,
      "loss": 2.4135,
      "step": 12500
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.3029026985168457,
      "eval_runtime": 682.8117,
      "eval_samples_per_second": 112.91,
      "eval_steps_per_second": 7.058,
      "step": 12500
    },
    {
      "epoch": 1.14,
      "learning_rate": 9.365099534095722e-05,
      "loss": 2.373,
      "step": 13000
    },
    {
      "epoch": 1.18,
      "learning_rate": 9.259212198221094e-05,
      "loss": 2.3231,
      "step": 13500
    },
    {
      "epoch": 1.22,
      "learning_rate": 9.153324862346464e-05,
      "loss": 2.2866,
      "step": 14000
    },
    {
      "epoch": 1.27,
      "learning_rate": 9.047437526471835e-05,
      "loss": 2.2534,
      "step": 14500
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.941550190597205e-05,
      "loss": 2.2136,
      "step": 15000
    },
    {
      "epoch": 1.31,
      "eval_loss": 2.124361753463745,
      "eval_runtime": 681.8811,
      "eval_samples_per_second": 113.064,
      "eval_steps_per_second": 7.067,
      "step": 15000
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.835874629394325e-05,
      "loss": 2.1843,
      "step": 15500
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.729987293519695e-05,
      "loss": 2.1551,
      "step": 16000
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.624099957645066e-05,
      "loss": 2.1253,
      "step": 16500
    },
    {
      "epoch": 1.49,
      "learning_rate": 8.518212621770438e-05,
      "loss": 2.1124,
      "step": 17000
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.412537060567557e-05,
      "loss": 2.0735,
      "step": 17500
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.9931037425994873,
      "eval_runtime": 682.7879,
      "eval_samples_per_second": 112.914,
      "eval_steps_per_second": 7.058,
      "step": 17500
    },
    {
      "epoch": 1.57,
      "learning_rate": 8.306649724692928e-05,
      "loss": 2.0535,
      "step": 18000
    },
    {
      "epoch": 1.62,
      "learning_rate": 8.200762388818297e-05,
      "loss": 2.0385,
      "step": 18500
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.095086827615418e-05,
      "loss": 2.0165,
      "step": 19000
    },
    {
      "epoch": 1.7,
      "learning_rate": 7.989199491740788e-05,
      "loss": 1.9965,
      "step": 19500
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.883312155866158e-05,
      "loss": 1.9684,
      "step": 20000
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.887757420539856,
      "eval_runtime": 683.0468,
      "eval_samples_per_second": 112.871,
      "eval_steps_per_second": 7.055,
      "step": 20000
    },
    {
      "epoch": 1.79,
      "learning_rate": 7.77742481999153e-05,
      "loss": 1.9452,
      "step": 20500
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.6715374841169e-05,
      "loss": 1.9475,
      "step": 21000
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.565650148242271e-05,
      "loss": 1.9304,
      "step": 21500
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.459762812367641e-05,
      "loss": 1.9201,
      "step": 22000
    },
    {
      "epoch": 1.97,
      "learning_rate": 7.353875476493012e-05,
      "loss": 1.891,
      "step": 22500
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.8076796531677246,
      "eval_runtime": 683.1394,
      "eval_samples_per_second": 112.855,
      "eval_steps_per_second": 7.054,
      "step": 22500
    },
    {
      "epoch": 2.01,
      "learning_rate": 7.248199915290131e-05,
      "loss": 1.8747,
      "step": 23000
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.142312579415502e-05,
      "loss": 1.8568,
      "step": 23500
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.036425243540872e-05,
      "loss": 1.832,
      "step": 24000
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.930749682337993e-05,
      "loss": 1.8274,
      "step": 24500
    },
    {
      "epoch": 2.18,
      "learning_rate": 6.824862346463363e-05,
      "loss": 1.8215,
      "step": 25000
    },
    {
      "epoch": 2.18,
      "eval_loss": 1.748734951019287,
      "eval_runtime": 675.7524,
      "eval_samples_per_second": 114.089,
      "eval_steps_per_second": 7.131,
      "step": 25000
    },
    {
      "epoch": 2.23,
      "learning_rate": 6.718975010588734e-05,
      "loss": 1.8079,
      "step": 25500
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.613087674714105e-05,
      "loss": 1.7946,
      "step": 26000
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.507200338839475e-05,
      "loss": 1.7862,
      "step": 26500
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.401313002964846e-05,
      "loss": 1.7719,
      "step": 27000
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.295425667090216e-05,
      "loss": 1.7577,
      "step": 27500
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.6874935626983643,
      "eval_runtime": 682.5902,
      "eval_samples_per_second": 112.946,
      "eval_steps_per_second": 7.06,
      "step": 27500
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.189538331215588e-05,
      "loss": 1.7455,
      "step": 28000
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.0838627700127074e-05,
      "loss": 1.7409,
      "step": 28500
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.977975434138078e-05,
      "loss": 1.7286,
      "step": 29000
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.8720880982634486e-05,
      "loss": 1.7205,
      "step": 29500
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.766200762388818e-05,
      "loss": 1.7113,
      "step": 30000
    },
    {
      "epoch": 2.62,
      "eval_loss": 1.6443524360656738,
      "eval_runtime": 681.9031,
      "eval_samples_per_second": 113.06,
      "eval_steps_per_second": 7.067,
      "step": 30000
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.660525201185939e-05,
      "loss": 1.7067,
      "step": 30500
    },
    {
      "epoch": 2.71,
      "learning_rate": 5.5546378653113096e-05,
      "loss": 1.6999,
      "step": 31000
    },
    {
      "epoch": 2.75,
      "learning_rate": 5.448750529436679e-05,
      "loss": 1.6787,
      "step": 31500
    },
    {
      "epoch": 2.8,
      "learning_rate": 5.34286319356205e-05,
      "loss": 1.6862,
      "step": 32000
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.2371876323591706e-05,
      "loss": 1.6776,
      "step": 32500
    },
    {
      "epoch": 2.84,
      "eval_loss": 1.6035559177398682,
      "eval_runtime": 677.3156,
      "eval_samples_per_second": 113.826,
      "eval_steps_per_second": 7.115,
      "step": 32500
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.1313002964845405e-05,
      "loss": 1.6667,
      "step": 33000
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.025412960609911e-05,
      "loss": 1.6636,
      "step": 33500
    },
    {
      "epoch": 2.97,
      "learning_rate": 4.919737399407031e-05,
      "loss": 1.6574,
      "step": 34000
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.813850063532402e-05,
      "loss": 1.6315,
      "step": 34500
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.707962727657773e-05,
      "loss": 1.6203,
      "step": 35000
    },
    {
      "epoch": 3.06,
      "eval_loss": 1.5608155727386475,
      "eval_runtime": 683.2194,
      "eval_samples_per_second": 112.842,
      "eval_steps_per_second": 7.053,
      "step": 35000
    },
    {
      "epoch": 3.1,
      "learning_rate": 4.602075391783143e-05,
      "loss": 1.6021,
      "step": 35500
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.496188055908513e-05,
      "loss": 1.609,
      "step": 36000
    },
    {
      "epoch": 3.19,
      "learning_rate": 4.390300720033884e-05,
      "loss": 1.6007,
      "step": 36500
    },
    {
      "epoch": 3.23,
      "learning_rate": 4.284413384159255e-05,
      "loss": 1.6025,
      "step": 37000
    },
    {
      "epoch": 3.28,
      "learning_rate": 4.178526048284626e-05,
      "loss": 1.6018,
      "step": 37500
    },
    {
      "epoch": 3.28,
      "eval_loss": 1.529254674911499,
      "eval_runtime": 682.7102,
      "eval_samples_per_second": 112.926,
      "eval_steps_per_second": 7.059,
      "step": 37500
    },
    {
      "epoch": 3.32,
      "learning_rate": 4.072638712409996e-05,
      "loss": 1.5893,
      "step": 38000
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.966963151207116e-05,
      "loss": 1.5829,
      "step": 38500
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.861075815332487e-05,
      "loss": 1.5757,
      "step": 39000
    },
    {
      "epoch": 3.45,
      "learning_rate": 3.755188479457857e-05,
      "loss": 1.5741,
      "step": 39500
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.649301143583227e-05,
      "loss": 1.5602,
      "step": 40000
    },
    {
      "epoch": 3.5,
      "eval_loss": 1.5043796300888062,
      "eval_runtime": 682.2924,
      "eval_samples_per_second": 112.996,
      "eval_steps_per_second": 7.063,
      "step": 40000
    },
    {
      "epoch": 3.54,
      "learning_rate": 3.543625582380348e-05,
      "loss": 1.5609,
      "step": 40500
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.4377382465057176e-05,
      "loss": 1.5567,
      "step": 41000
    },
    {
      "epoch": 3.63,
      "learning_rate": 3.331850910631088e-05,
      "loss": 1.5423,
      "step": 41500
    },
    {
      "epoch": 3.67,
      "learning_rate": 3.2259635747564595e-05,
      "loss": 1.5451,
      "step": 42000
    },
    {
      "epoch": 3.71,
      "learning_rate": 3.1202880135535786e-05,
      "loss": 1.5429,
      "step": 42500
    },
    {
      "epoch": 3.71,
      "eval_loss": 1.475334644317627,
      "eval_runtime": 682.6287,
      "eval_samples_per_second": 112.94,
      "eval_steps_per_second": 7.059,
      "step": 42500
    },
    {
      "epoch": 3.76,
      "learning_rate": 3.0144006776789495e-05,
      "loss": 1.5331,
      "step": 43000
    },
    {
      "epoch": 3.8,
      "learning_rate": 2.9085133418043205e-05,
      "loss": 1.5283,
      "step": 43500
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.802626005929691e-05,
      "loss": 1.528,
      "step": 44000
    },
    {
      "epoch": 3.89,
      "learning_rate": 2.696950444726811e-05,
      "loss": 1.5216,
      "step": 44500
    },
    {
      "epoch": 3.93,
      "learning_rate": 2.5910631088521814e-05,
      "loss": 1.5148,
      "step": 45000
    },
    {
      "epoch": 3.93,
      "eval_loss": 1.4471635818481445,
      "eval_runtime": 682.5313,
      "eval_samples_per_second": 112.956,
      "eval_steps_per_second": 7.06,
      "step": 45000
    },
    {
      "epoch": 3.98,
      "learning_rate": 2.485175772977552e-05,
      "loss": 1.5166,
      "step": 45500
    },
    {
      "epoch": 4.02,
      "learning_rate": 2.3795002117746718e-05,
      "loss": 1.49,
      "step": 46000
    },
    {
      "epoch": 4.06,
      "learning_rate": 2.2736128759000424e-05,
      "loss": 1.4958,
      "step": 46500
    },
    {
      "epoch": 4.11,
      "learning_rate": 2.167725540025413e-05,
      "loss": 1.4896,
      "step": 47000
    },
    {
      "epoch": 4.15,
      "learning_rate": 2.0618382041507836e-05,
      "loss": 1.4786,
      "step": 47500
    },
    {
      "epoch": 4.15,
      "eval_loss": 1.430184006690979,
      "eval_runtime": 682.3288,
      "eval_samples_per_second": 112.99,
      "eval_steps_per_second": 7.063,
      "step": 47500
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.9559508682761542e-05,
      "loss": 1.4665,
      "step": 48000
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.850063532401525e-05,
      "loss": 1.4743,
      "step": 48500
    },
    {
      "epoch": 4.28,
      "learning_rate": 1.7441761965268955e-05,
      "loss": 1.4731,
      "step": 49000
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.638288860652266e-05,
      "loss": 1.4792,
      "step": 49500
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.5326132994493858e-05,
      "loss": 1.4653,
      "step": 50000
    },
    {
      "epoch": 4.37,
      "eval_loss": 1.4128450155258179,
      "eval_runtime": 683.1234,
      "eval_samples_per_second": 112.858,
      "eval_steps_per_second": 7.054,
      "step": 50000
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.4267259635747566e-05,
      "loss": 1.4658,
      "step": 50500
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.3208386277001272e-05,
      "loss": 1.4619,
      "step": 51000
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.215163066497247e-05,
      "loss": 1.456,
      "step": 51500
    },
    {
      "epoch": 4.54,
      "learning_rate": 1.1092757306226175e-05,
      "loss": 1.4599,
      "step": 52000
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.0033883947479882e-05,
      "loss": 1.4496,
      "step": 52500
    },
    {
      "epoch": 4.59,
      "eval_loss": 1.3991329669952393,
      "eval_runtime": 682.7978,
      "eval_samples_per_second": 112.912,
      "eval_steps_per_second": 7.058,
      "step": 52500
    },
    {
      "epoch": 4.63,
      "learning_rate": 8.975010588733588e-06,
      "loss": 1.4476,
      "step": 53000
    },
    {
      "epoch": 4.67,
      "learning_rate": 7.916137229987294e-06,
      "loss": 1.4422,
      "step": 53500
    },
    {
      "epoch": 4.72,
      "learning_rate": 6.859381617958492e-06,
      "loss": 1.4399,
      "step": 54000
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.800508259212199e-06,
      "loss": 1.4356,
      "step": 54500
    },
    {
      "epoch": 4.81,
      "learning_rate": 4.741634900465904e-06,
      "loss": 1.4445,
      "step": 55000
    },
    {
      "epoch": 4.81,
      "eval_loss": 1.3943445682525635,
      "eval_runtime": 679.9865,
      "eval_samples_per_second": 113.379,
      "eval_steps_per_second": 7.087,
      "step": 55000
    },
    {
      "epoch": 4.85,
      "learning_rate": 5.645250861738798e-05,
      "loss": 1.4914,
      "step": 55500
    },
    {
      "epoch": 4.89,
      "learning_rate": 5.5973764841057066e-05,
      "loss": 1.5131,
      "step": 56000
    },
    {
      "epoch": 4.94,
      "learning_rate": 5.549502106472616e-05,
      "loss": 1.5147,
      "step": 56500
    },
    {
      "epoch": 4.98,
      "learning_rate": 5.5017234775947915e-05,
      "loss": 1.5136,
      "step": 57000
    },
    {
      "epoch": 5.02,
      "learning_rate": 5.4539448487169664e-05,
      "loss": 1.5114,
      "step": 57500
    },
    {
      "epoch": 5.02,
      "eval_loss": 1.455129861831665,
      "eval_runtime": 799.4542,
      "eval_samples_per_second": 96.436,
      "eval_steps_per_second": 6.028,
      "step": 57500
    },
    {
      "epoch": 5.07,
      "learning_rate": 5.4060704710838764e-05,
      "loss": 1.5048,
      "step": 58000
    },
    {
      "epoch": 5.11,
      "learning_rate": 5.3581960934507856e-05,
      "loss": 1.496,
      "step": 58500
    },
    {
      "epoch": 5.16,
      "learning_rate": 5.310321715817694e-05,
      "loss": 1.4956,
      "step": 59000
    },
    {
      "epoch": 5.2,
      "learning_rate": 5.262447338184604e-05,
      "loss": 1.4965,
      "step": 59500
    },
    {
      "epoch": 5.24,
      "learning_rate": 5.214668709306779e-05,
      "loss": 1.5054,
      "step": 60000
    },
    {
      "epoch": 5.24,
      "eval_loss": 1.4524710178375244,
      "eval_runtime": 799.9515,
      "eval_samples_per_second": 96.376,
      "eval_steps_per_second": 6.024,
      "step": 60000
    },
    {
      "epoch": 5.29,
      "learning_rate": 5.166794331673689e-05,
      "loss": 1.4931,
      "step": 60500
    },
    {
      "epoch": 5.33,
      "learning_rate": 5.1189199540405975e-05,
      "loss": 1.4855,
      "step": 61000
    },
    {
      "epoch": 5.37,
      "learning_rate": 5.0710455764075074e-05,
      "loss": 1.4925,
      "step": 61500
    },
    {
      "epoch": 5.42,
      "learning_rate": 5.023171198774416e-05,
      "loss": 1.4789,
      "step": 62000
    },
    {
      "epoch": 5.46,
      "learning_rate": 4.975296821141325e-05,
      "loss": 1.4817,
      "step": 62500
    },
    {
      "epoch": 5.46,
      "eval_loss": 1.4259248971939087,
      "eval_runtime": 800.5391,
      "eval_samples_per_second": 96.305,
      "eval_steps_per_second": 6.02,
      "step": 62500
    },
    {
      "epoch": 5.51,
      "learning_rate": 4.9274224435082345e-05,
      "loss": 1.4705,
      "step": 63000
    },
    {
      "epoch": 5.55,
      "learning_rate": 4.8795480658751444e-05,
      "loss": 1.4764,
      "step": 63500
    },
    {
      "epoch": 5.59,
      "learning_rate": 4.8317694369973194e-05,
      "loss": 1.4703,
      "step": 64000
    },
    {
      "epoch": 5.64,
      "learning_rate": 4.7838950593642286e-05,
      "loss": 1.4647,
      "step": 64500
    },
    {
      "epoch": 5.68,
      "learning_rate": 4.736020681731138e-05,
      "loss": 1.48,
      "step": 65000
    },
    {
      "epoch": 5.68,
      "eval_loss": 1.4077460765838623,
      "eval_runtime": 800.4986,
      "eval_samples_per_second": 96.31,
      "eval_steps_per_second": 6.02,
      "step": 65000
    },
    {
      "epoch": 5.72,
      "learning_rate": 4.688146304098047e-05,
      "loss": 1.4646,
      "step": 65500
    },
    {
      "epoch": 5.77,
      "learning_rate": 4.640271926464956e-05,
      "loss": 1.4588,
      "step": 66000
    },
    {
      "epoch": 5.81,
      "learning_rate": 4.5923975488318656e-05,
      "loss": 1.4652,
      "step": 66500
    },
    {
      "epoch": 5.85,
      "learning_rate": 4.5446189199540405e-05,
      "loss": 1.4505,
      "step": 67000
    },
    {
      "epoch": 5.9,
      "learning_rate": 4.49674454232095e-05,
      "loss": 1.4526,
      "step": 67500
    },
    {
      "epoch": 5.9,
      "eval_loss": 1.3912079334259033,
      "eval_runtime": 800.7475,
      "eval_samples_per_second": 96.28,
      "eval_steps_per_second": 6.018,
      "step": 67500
    },
    {
      "epoch": 5.94,
      "learning_rate": 4.44887016468786e-05,
      "loss": 1.4444,
      "step": 68000
    },
    {
      "epoch": 5.99,
      "learning_rate": 4.400995787054768e-05,
      "loss": 1.4416,
      "step": 68500
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.353217158176944e-05,
      "loss": 1.4435,
      "step": 69000
    },
    {
      "epoch": 6.07,
      "learning_rate": 4.305342780543853e-05,
      "loss": 1.4261,
      "step": 69500
    },
    {
      "epoch": 6.12,
      "learning_rate": 4.2574684029107624e-05,
      "loss": 1.4272,
      "step": 70000
    },
    {
      "epoch": 6.12,
      "eval_loss": 1.3726316690444946,
      "eval_runtime": 801.0967,
      "eval_samples_per_second": 96.238,
      "eval_steps_per_second": 6.016,
      "step": 70000
    },
    {
      "epoch": 6.16,
      "learning_rate": 4.2095940252776716e-05,
      "loss": 1.4291,
      "step": 70500
    },
    {
      "epoch": 6.2,
      "learning_rate": 4.161719647644581e-05,
      "loss": 1.4204,
      "step": 71000
    },
    {
      "epoch": 6.25,
      "learning_rate": 4.11384527001149e-05,
      "loss": 1.4143,
      "step": 71500
    },
    {
      "epoch": 6.29,
      "learning_rate": 4.065970892378399e-05,
      "loss": 1.4219,
      "step": 72000
    },
    {
      "epoch": 6.34,
      "learning_rate": 4.0180965147453086e-05,
      "loss": 1.4078,
      "step": 72500
    },
    {
      "epoch": 6.34,
      "eval_loss": 1.3595777750015259,
      "eval_runtime": 801.4171,
      "eval_samples_per_second": 96.2,
      "eval_steps_per_second": 6.013,
      "step": 72500
    },
    {
      "epoch": 6.38,
      "learning_rate": 3.970317885867484e-05,
      "loss": 1.3986,
      "step": 73000
    },
    {
      "epoch": 6.42,
      "learning_rate": 3.922443508234393e-05,
      "loss": 1.3945,
      "step": 73500
    },
    {
      "epoch": 6.47,
      "learning_rate": 3.874569130601302e-05,
      "loss": 1.404,
      "step": 74000
    },
    {
      "epoch": 6.51,
      "learning_rate": 3.826694752968212e-05,
      "loss": 1.4029,
      "step": 74500
    },
    {
      "epoch": 6.55,
      "learning_rate": 3.778916124090387e-05,
      "loss": 1.399,
      "step": 75000
    },
    {
      "epoch": 6.55,
      "eval_loss": 1.3450202941894531,
      "eval_runtime": 801.3866,
      "eval_samples_per_second": 96.203,
      "eval_steps_per_second": 6.013,
      "step": 75000
    },
    {
      "epoch": 6.6,
      "learning_rate": 3.731041746457296e-05,
      "loss": 1.3884,
      "step": 75500
    },
    {
      "epoch": 6.64,
      "learning_rate": 3.683263117579472e-05,
      "loss": 1.382,
      "step": 76000
    },
    {
      "epoch": 6.68,
      "learning_rate": 3.635388739946381e-05,
      "loss": 1.3888,
      "step": 76500
    },
    {
      "epoch": 6.73,
      "learning_rate": 3.58751436231329e-05,
      "loss": 1.3847,
      "step": 77000
    },
    {
      "epoch": 6.77,
      "learning_rate": 3.5396399846801995e-05,
      "loss": 1.386,
      "step": 77500
    },
    {
      "epoch": 6.77,
      "eval_loss": 1.3327932357788086,
      "eval_runtime": 800.4263,
      "eval_samples_per_second": 96.319,
      "eval_steps_per_second": 6.021,
      "step": 77500
    },
    {
      "epoch": 6.82,
      "learning_rate": 3.491765607047108e-05,
      "loss": 1.3941,
      "step": 78000
    },
    {
      "epoch": 6.86,
      "learning_rate": 3.443891229414018e-05,
      "loss": 1.3698,
      "step": 78500
    },
    {
      "epoch": 6.9,
      "learning_rate": 3.396016851780927e-05,
      "loss": 1.3712,
      "step": 79000
    },
    {
      "epoch": 6.95,
      "learning_rate": 3.3481424741478364e-05,
      "loss": 1.3751,
      "step": 79500
    },
    {
      "epoch": 6.99,
      "learning_rate": 3.300268096514745e-05,
      "loss": 1.3704,
      "step": 80000
    },
    {
      "epoch": 6.99,
      "eval_loss": 1.319200038909912,
      "eval_runtime": 800.9368,
      "eval_samples_per_second": 96.257,
      "eval_steps_per_second": 6.017,
      "step": 80000
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.252489467636921e-05,
      "loss": 1.3614,
      "step": 80500
    },
    {
      "epoch": 7.08,
      "learning_rate": 3.20461509000383e-05,
      "loss": 1.3624,
      "step": 81000
    },
    {
      "epoch": 7.12,
      "learning_rate": 3.156740712370739e-05,
      "loss": 1.3534,
      "step": 81500
    },
    {
      "epoch": 7.17,
      "learning_rate": 3.1088663347376484e-05,
      "loss": 1.3537,
      "step": 82000
    },
    {
      "epoch": 7.21,
      "learning_rate": 3.061087705859824e-05,
      "loss": 1.3538,
      "step": 82500
    },
    {
      "epoch": 7.21,
      "eval_loss": 1.3130738735198975,
      "eval_runtime": 801.3578,
      "eval_samples_per_second": 96.207,
      "eval_steps_per_second": 6.014,
      "step": 82500
    },
    {
      "epoch": 7.25,
      "learning_rate": 3.0132133282267332e-05,
      "loss": 1.343,
      "step": 83000
    },
    {
      "epoch": 7.3,
      "learning_rate": 2.9653389505936425e-05,
      "loss": 1.3452,
      "step": 83500
    },
    {
      "epoch": 7.34,
      "learning_rate": 2.9174645729605514e-05,
      "loss": 1.3414,
      "step": 84000
    },
    {
      "epoch": 7.38,
      "learning_rate": 2.8695901953274606e-05,
      "loss": 1.3449,
      "step": 84500
    },
    {
      "epoch": 7.43,
      "learning_rate": 2.8217158176943702e-05,
      "loss": 1.3468,
      "step": 85000
    },
    {
      "epoch": 7.43,
      "eval_loss": 1.2915565967559814,
      "eval_runtime": 801.1997,
      "eval_samples_per_second": 96.226,
      "eval_steps_per_second": 6.015,
      "step": 85000
    },
    {
      "epoch": 7.47,
      "learning_rate": 2.7739371888165455e-05,
      "loss": 1.3406,
      "step": 85500
    },
    {
      "epoch": 7.51,
      "learning_rate": 2.7260628111834547e-05,
      "loss": 1.337,
      "step": 86000
    },
    {
      "epoch": 7.56,
      "learning_rate": 2.6781884335503636e-05,
      "loss": 1.3374,
      "step": 86500
    },
    {
      "epoch": 7.6,
      "learning_rate": 2.6303140559172735e-05,
      "loss": 1.3232,
      "step": 87000
    },
    {
      "epoch": 7.65,
      "learning_rate": 2.5824396782841825e-05,
      "loss": 1.323,
      "step": 87500
    },
    {
      "epoch": 7.65,
      "eval_loss": 1.2871359586715698,
      "eval_runtime": 801.3386,
      "eval_samples_per_second": 96.209,
      "eval_steps_per_second": 6.014,
      "step": 87500
    },
    {
      "epoch": 7.69,
      "learning_rate": 2.5345653006510917e-05,
      "loss": 1.3165,
      "step": 88000
    },
    {
      "epoch": 7.73,
      "learning_rate": 2.4866909230180006e-05,
      "loss": 1.3311,
      "step": 88500
    },
    {
      "epoch": 7.78,
      "learning_rate": 2.4389122941401762e-05,
      "loss": 1.3187,
      "step": 89000
    },
    {
      "epoch": 7.82,
      "learning_rate": 2.3910379165070858e-05,
      "loss": 1.3243,
      "step": 89500
    },
    {
      "epoch": 7.86,
      "learning_rate": 2.3431635388739947e-05,
      "loss": 1.322,
      "step": 90000
    },
    {
      "epoch": 7.86,
      "eval_loss": 1.262225866317749,
      "eval_runtime": 799.8871,
      "eval_samples_per_second": 96.384,
      "eval_steps_per_second": 6.025,
      "step": 90000
    },
    {
      "epoch": 7.91,
      "learning_rate": 2.295289161240904e-05,
      "loss": 1.3195,
      "step": 90500
    },
    {
      "epoch": 7.95,
      "learning_rate": 2.2475105323630792e-05,
      "loss": 1.3097,
      "step": 91000
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.1996361547299888e-05,
      "loss": 1.3154,
      "step": 91500
    },
    {
      "epoch": 8.04,
      "learning_rate": 2.1517617770968977e-05,
      "loss": 1.2994,
      "step": 92000
    },
    {
      "epoch": 8.08,
      "learning_rate": 2.103887399463807e-05,
      "loss": 1.2956,
      "step": 92500
    },
    {
      "epoch": 8.08,
      "eval_loss": 1.2623521089553833,
      "eval_runtime": 801.376,
      "eval_samples_per_second": 96.205,
      "eval_steps_per_second": 6.013,
      "step": 92500
    },
    {
      "epoch": 8.13,
      "learning_rate": 2.0560130218307162e-05,
      "loss": 1.2956,
      "step": 93000
    },
    {
      "epoch": 8.17,
      "learning_rate": 2.0081386441976255e-05,
      "loss": 1.2945,
      "step": 93500
    },
    {
      "epoch": 8.21,
      "learning_rate": 1.960360015319801e-05,
      "loss": 1.2971,
      "step": 94000
    },
    {
      "epoch": 8.26,
      "learning_rate": 1.91248563768671e-05,
      "loss": 1.2941,
      "step": 94500
    },
    {
      "epoch": 8.3,
      "learning_rate": 1.8646112600536196e-05,
      "loss": 1.2869,
      "step": 95000
    },
    {
      "epoch": 8.3,
      "eval_loss": 1.2546786069869995,
      "eval_runtime": 800.9015,
      "eval_samples_per_second": 96.262,
      "eval_steps_per_second": 6.017,
      "step": 95000
    },
    {
      "epoch": 8.34,
      "learning_rate": 1.8167368824205285e-05,
      "loss": 1.2868,
      "step": 95500
    },
    {
      "epoch": 8.39,
      "learning_rate": 1.768958253542704e-05,
      "loss": 1.2939,
      "step": 96000
    },
    {
      "epoch": 8.43,
      "learning_rate": 1.7210838759096133e-05,
      "loss": 1.2876,
      "step": 96500
    },
    {
      "epoch": 8.48,
      "learning_rate": 1.6732094982765226e-05,
      "loss": 1.2843,
      "step": 97000
    },
    {
      "epoch": 8.52,
      "learning_rate": 1.625430869398698e-05,
      "loss": 1.2763,
      "step": 97500
    },
    {
      "epoch": 8.52,
      "eval_loss": 1.2404190301895142,
      "eval_runtime": 800.8453,
      "eval_samples_per_second": 96.268,
      "eval_steps_per_second": 6.017,
      "step": 97500
    },
    {
      "epoch": 8.56,
      "learning_rate": 1.577556491765607e-05,
      "loss": 1.2845,
      "step": 98000
    },
    {
      "epoch": 8.61,
      "learning_rate": 1.5296821141325163e-05,
      "loss": 1.2795,
      "step": 98500
    },
    {
      "epoch": 8.65,
      "learning_rate": 1.4818077364994256e-05,
      "loss": 1.275,
      "step": 99000
    },
    {
      "epoch": 8.69,
      "learning_rate": 1.4339333588663348e-05,
      "loss": 1.2763,
      "step": 99500
    },
    {
      "epoch": 8.74,
      "learning_rate": 1.3860589812332439e-05,
      "loss": 1.275,
      "step": 100000
    },
    {
      "epoch": 8.74,
      "eval_loss": 1.2305477857589722,
      "eval_runtime": 799.9521,
      "eval_samples_per_second": 96.376,
      "eval_steps_per_second": 6.024,
      "step": 100000
    },
    {
      "epoch": 8.78,
      "learning_rate": 1.3381846036001533e-05,
      "loss": 1.2724,
      "step": 100500
    },
    {
      "epoch": 8.83,
      "learning_rate": 1.2903102259670624e-05,
      "loss": 1.2779,
      "step": 101000
    },
    {
      "epoch": 8.87,
      "learning_rate": 1.2424358483339716e-05,
      "loss": 1.2625,
      "step": 101500
    },
    {
      "epoch": 8.91,
      "learning_rate": 1.1946572194561471e-05,
      "loss": 1.2647,
      "step": 102000
    },
    {
      "epoch": 8.96,
      "learning_rate": 1.1467828418230563e-05,
      "loss": 1.2709,
      "step": 102500
    },
    {
      "epoch": 8.96,
      "eval_loss": 1.2300862073898315,
      "eval_runtime": 800.2626,
      "eval_samples_per_second": 96.338,
      "eval_steps_per_second": 6.022,
      "step": 102500
    },
    {
      "epoch": 9.0,
      "learning_rate": 1.0989084641899656e-05,
      "loss": 1.2647,
      "step": 103000
    },
    {
      "epoch": 9.04,
      "learning_rate": 1.0510340865568748e-05,
      "loss": 1.2462,
      "step": 103500
    },
    {
      "epoch": 9.09,
      "learning_rate": 1.003159708923784e-05,
      "loss": 1.2586,
      "step": 104000
    },
    {
      "epoch": 9.13,
      "learning_rate": 9.552853312906933e-06,
      "loss": 1.2455,
      "step": 104500
    },
    {
      "epoch": 9.18,
      "learning_rate": 9.074109536576026e-06,
      "loss": 1.2514,
      "step": 105000
    },
    {
      "epoch": 9.18,
      "eval_loss": 1.2178542613983154,
      "eval_runtime": 829.9046,
      "eval_samples_per_second": 92.897,
      "eval_steps_per_second": 5.807,
      "step": 105000
    },
    {
      "epoch": 9.22,
      "learning_rate": 8.595365760245118e-06,
      "loss": 1.2526,
      "step": 105500
    },
    {
      "epoch": 9.26,
      "learning_rate": 8.11757947146687e-06,
      "loss": 1.2523,
      "step": 106000
    },
    {
      "epoch": 9.31,
      "learning_rate": 7.638835695135963e-06,
      "loss": 1.2504,
      "step": 106500
    },
    {
      "epoch": 9.35,
      "learning_rate": 7.1600919188050565e-06,
      "loss": 1.2516,
      "step": 107000
    },
    {
      "epoch": 9.39,
      "learning_rate": 6.681348142474147e-06,
      "loss": 1.2563,
      "step": 107500
    },
    {
      "epoch": 9.39,
      "eval_loss": 1.2134195566177368,
      "eval_runtime": 802.5122,
      "eval_samples_per_second": 96.068,
      "eval_steps_per_second": 6.005,
      "step": 107500
    },
    {
      "epoch": 9.44,
      "learning_rate": 6.204519341248564e-06,
      "loss": 1.2407,
      "step": 108000
    },
    {
      "epoch": 9.48,
      "learning_rate": 5.725775564917656e-06,
      "loss": 1.243,
      "step": 108500
    },
    {
      "epoch": 9.52,
      "learning_rate": 5.247031788586749e-06,
      "loss": 1.2449,
      "step": 109000
    },
    {
      "epoch": 9.57,
      "learning_rate": 4.768288012255841e-06,
      "loss": 1.2442,
      "step": 109500
    },
    {
      "epoch": 9.61,
      "learning_rate": 4.290501723477595e-06,
      "loss": 1.2487,
      "step": 110000
    },
    {
      "epoch": 9.61,
      "eval_loss": 1.211136817932129,
      "eval_runtime": 800.5992,
      "eval_samples_per_second": 96.298,
      "eval_steps_per_second": 6.019,
      "step": 110000
    },
    {
      "epoch": 9.66,
      "learning_rate": 3.811757947146687e-06,
      "loss": 1.2438,
      "step": 110500
    },
    {
      "epoch": 9.7,
      "learning_rate": 3.33301417081578e-06,
      "loss": 1.2399,
      "step": 111000
    },
    {
      "epoch": 9.74,
      "learning_rate": 2.8542703944848716e-06,
      "loss": 1.236,
      "step": 111500
    },
    {
      "epoch": 9.79,
      "learning_rate": 2.375526618153964e-06,
      "loss": 1.2381,
      "step": 112000
    },
    {
      "epoch": 9.83,
      "learning_rate": 1.8977403293757183e-06,
      "loss": 1.2337,
      "step": 112500
    },
    {
      "epoch": 9.83,
      "eval_loss": 1.2040736675262451,
      "eval_runtime": 801.2606,
      "eval_samples_per_second": 96.218,
      "eval_steps_per_second": 6.014,
      "step": 112500
    },
    {
      "epoch": 9.87,
      "learning_rate": 5.296144005847954e-05,
      "loss": 1.2919,
      "step": 113000
    },
    {
      "epoch": 9.92,
      "learning_rate": 5.273300438596491e-05,
      "loss": 1.3121,
      "step": 113500
    },
    {
      "epoch": 9.96,
      "learning_rate": 5.25045687134503e-05,
      "loss": 1.3188,
      "step": 114000
    },
    {
      "epoch": 10.01,
      "learning_rate": 5.2276589912280705e-05,
      "loss": 1.3266,
      "step": 114500
    },
    {
      "epoch": 10.05,
      "learning_rate": 5.204815423976609e-05,
      "loss": 1.3215,
      "step": 115000
    },
    {
      "epoch": 10.05,
      "eval_loss": 1.2878538370132446,
      "eval_runtime": 801.5103,
      "eval_samples_per_second": 96.188,
      "eval_steps_per_second": 6.012,
      "step": 115000
    },
    {
      "epoch": 10.09,
      "learning_rate": 5.1819718567251463e-05,
      "loss": 1.3241,
      "step": 115500
    },
    {
      "epoch": 10.14,
      "learning_rate": 5.1591282894736846e-05,
      "loss": 1.3216,
      "step": 116000
    },
    {
      "epoch": 10.18,
      "learning_rate": 5.136284722222222e-05,
      "loss": 1.3297,
      "step": 116500
    },
    {
      "epoch": 10.22,
      "learning_rate": 5.1134411549707604e-05,
      "loss": 1.332,
      "step": 117000
    },
    {
      "epoch": 10.27,
      "learning_rate": 5.090597587719298e-05,
      "loss": 1.3364,
      "step": 117500
    },
    {
      "epoch": 10.27,
      "eval_loss": 1.2850462198257446,
      "eval_runtime": 799.6277,
      "eval_samples_per_second": 96.415,
      "eval_steps_per_second": 6.027,
      "step": 117500
    },
    {
      "epoch": 10.31,
      "learning_rate": 5.06779970760234e-05,
      "loss": 1.3357,
      "step": 118000
    },
    {
      "epoch": 10.35,
      "learning_rate": 5.044956140350877e-05,
      "loss": 1.3299,
      "step": 118500
    },
    {
      "epoch": 10.4,
      "learning_rate": 5.022112573099416e-05,
      "loss": 1.3297,
      "step": 119000
    },
    {
      "epoch": 10.44,
      "learning_rate": 4.9992690058479533e-05,
      "loss": 1.3227,
      "step": 119500
    },
    {
      "epoch": 10.49,
      "learning_rate": 4.9764254385964916e-05,
      "loss": 1.3286,
      "step": 120000
    },
    {
      "epoch": 10.49,
      "eval_loss": 1.2779029607772827,
      "eval_runtime": 798.5622,
      "eval_samples_per_second": 96.544,
      "eval_steps_per_second": 6.035,
      "step": 120000
    },
    {
      "epoch": 10.53,
      "learning_rate": 4.953627558479532e-05,
      "loss": 1.3185,
      "step": 120500
    },
    {
      "epoch": 10.57,
      "learning_rate": 4.9307839912280704e-05,
      "loss": 1.3202,
      "step": 121000
    },
    {
      "epoch": 10.62,
      "learning_rate": 4.9079861111111116e-05,
      "loss": 1.3174,
      "step": 121500
    },
    {
      "epoch": 10.66,
      "learning_rate": 4.885142543859649e-05,
      "loss": 1.3146,
      "step": 122000
    },
    {
      "epoch": 10.7,
      "learning_rate": 4.8622989766081875e-05,
      "loss": 1.3202,
      "step": 122500
    },
    {
      "epoch": 10.7,
      "eval_loss": 1.2730140686035156,
      "eval_runtime": 800.6627,
      "eval_samples_per_second": 96.29,
      "eval_steps_per_second": 6.019,
      "step": 122500
    },
    {
      "epoch": 10.75,
      "learning_rate": 4.839455409356725e-05,
      "loss": 1.3214,
      "step": 123000
    },
    {
      "epoch": 10.79,
      "learning_rate": 4.816611842105263e-05,
      "loss": 1.3105,
      "step": 123500
    },
    {
      "epoch": 10.84,
      "learning_rate": 4.7937682748538016e-05,
      "loss": 1.3203,
      "step": 124000
    },
    {
      "epoch": 10.88,
      "learning_rate": 4.770924707602339e-05,
      "loss": 1.3061,
      "step": 124500
    },
    {
      "epoch": 10.92,
      "learning_rate": 4.7480811403508774e-05,
      "loss": 1.3181,
      "step": 125000
    },
    {
      "epoch": 10.92,
      "eval_loss": 1.2651313543319702,
      "eval_runtime": 800.4049,
      "eval_samples_per_second": 96.321,
      "eval_steps_per_second": 6.021,
      "step": 125000
    },
    {
      "epoch": 10.97,
      "learning_rate": 4.725237573099415e-05,
      "loss": 1.3121,
      "step": 125500
    },
    {
      "epoch": 11.01,
      "learning_rate": 4.702439692982457e-05,
      "loss": 1.2988,
      "step": 126000
    },
    {
      "epoch": 11.05,
      "learning_rate": 4.6795961257309945e-05,
      "loss": 1.2894,
      "step": 126500
    },
    {
      "epoch": 11.1,
      "learning_rate": 4.656752558479533e-05,
      "loss": 1.2976,
      "step": 127000
    },
    {
      "epoch": 11.14,
      "learning_rate": 4.63390899122807e-05,
      "loss": 1.2952,
      "step": 127500
    },
    {
      "epoch": 11.14,
      "eval_loss": 1.2544220685958862,
      "eval_runtime": 798.8455,
      "eval_samples_per_second": 96.509,
      "eval_steps_per_second": 6.032,
      "step": 127500
    },
    {
      "epoch": 11.18,
      "learning_rate": 4.6111111111111115e-05,
      "loss": 1.2929,
      "step": 128000
    },
    {
      "epoch": 11.23,
      "learning_rate": 4.58826754385965e-05,
      "loss": 1.2877,
      "step": 128500
    },
    {
      "epoch": 11.27,
      "learning_rate": 4.5654239766081874e-05,
      "loss": 1.2868,
      "step": 129000
    },
    {
      "epoch": 11.32,
      "learning_rate": 4.5425804093567256e-05,
      "loss": 1.2839,
      "step": 129500
    },
    {
      "epoch": 11.36,
      "learning_rate": 4.519736842105263e-05,
      "loss": 1.2889,
      "step": 130000
    },
    {
      "epoch": 11.36,
      "eval_loss": 1.2505890130996704,
      "eval_runtime": 798.9045,
      "eval_samples_per_second": 96.502,
      "eval_steps_per_second": 6.032,
      "step": 130000
    },
    {
      "epoch": 11.4,
      "learning_rate": 4.4969389619883044e-05,
      "loss": 1.2844,
      "step": 130500
    },
    {
      "epoch": 11.45,
      "learning_rate": 4.474095394736843e-05,
      "loss": 1.286,
      "step": 131000
    },
    {
      "epoch": 11.49,
      "learning_rate": 4.45125182748538e-05,
      "loss": 1.2997,
      "step": 131500
    },
    {
      "epoch": 11.53,
      "learning_rate": 4.4284082602339185e-05,
      "loss": 1.2813,
      "step": 132000
    },
    {
      "epoch": 11.58,
      "learning_rate": 4.405610380116959e-05,
      "loss": 1.2747,
      "step": 132500
    },
    {
      "epoch": 11.58,
      "eval_loss": 1.2338725328445435,
      "eval_runtime": 799.247,
      "eval_samples_per_second": 96.461,
      "eval_steps_per_second": 6.029,
      "step": 132500
    },
    {
      "epoch": 11.62,
      "learning_rate": 4.3827668128654973e-05,
      "loss": 1.2803,
      "step": 133000
    },
    {
      "epoch": 11.67,
      "learning_rate": 4.3599232456140356e-05,
      "loss": 1.2869,
      "step": 133500
    },
    {
      "epoch": 11.71,
      "learning_rate": 4.337125365497076e-05,
      "loss": 1.2837,
      "step": 134000
    },
    {
      "epoch": 11.75,
      "learning_rate": 4.3142817982456144e-05,
      "loss": 1.2826,
      "step": 134500
    },
    {
      "epoch": 11.8,
      "learning_rate": 4.291438230994152e-05,
      "loss": 1.2729,
      "step": 135000
    },
    {
      "epoch": 11.8,
      "eval_loss": 1.2277261018753052,
      "eval_runtime": 794.6626,
      "eval_samples_per_second": 97.017,
      "eval_steps_per_second": 6.064,
      "step": 135000
    },
    {
      "epoch": 11.84,
      "learning_rate": 4.26859466374269e-05,
      "loss": 1.2705,
      "step": 135500
    },
    {
      "epoch": 11.88,
      "learning_rate": 4.2457510964912285e-05,
      "loss": 1.2719,
      "step": 136000
    },
    {
      "epoch": 11.93,
      "learning_rate": 4.222907529239766e-05,
      "loss": 1.2672,
      "step": 136500
    },
    {
      "epoch": 11.97,
      "learning_rate": 4.200063961988304e-05,
      "loss": 1.2699,
      "step": 137000
    },
    {
      "epoch": 12.02,
      "learning_rate": 4.177220394736842e-05,
      "loss": 1.2699,
      "step": 137500
    },
    {
      "epoch": 12.02,
      "eval_loss": 1.220065951347351,
      "eval_runtime": 798.5913,
      "eval_samples_per_second": 96.54,
      "eval_steps_per_second": 6.034,
      "step": 137500
    },
    {
      "epoch": 12.06,
      "learning_rate": 4.154422514619883e-05,
      "loss": 1.2554,
      "step": 138000
    },
    {
      "epoch": 12.1,
      "learning_rate": 4.1315789473684214e-05,
      "loss": 1.2609,
      "step": 138500
    },
    {
      "epoch": 12.15,
      "learning_rate": 4.108735380116959e-05,
      "loss": 1.2639,
      "step": 139000
    },
    {
      "epoch": 12.19,
      "learning_rate": 4.085891812865497e-05,
      "loss": 1.2491,
      "step": 139500
    },
    {
      "epoch": 12.23,
      "learning_rate": 4.063048245614035e-05,
      "loss": 1.2508,
      "step": 140000
    },
    {
      "epoch": 12.23,
      "eval_loss": 1.2163188457489014,
      "eval_runtime": 791.6231,
      "eval_samples_per_second": 97.39,
      "eval_steps_per_second": 6.087,
      "step": 140000
    },
    {
      "epoch": 12.28,
      "learning_rate": 4.040250365497076e-05,
      "loss": 1.2521,
      "step": 140500
    },
    {
      "epoch": 12.32,
      "learning_rate": 4.017406798245614e-05,
      "loss": 1.256,
      "step": 141000
    },
    {
      "epoch": 12.36,
      "learning_rate": 3.994563230994152e-05,
      "loss": 1.245,
      "step": 141500
    },
    {
      "epoch": 12.41,
      "learning_rate": 3.97171966374269e-05,
      "loss": 1.2516,
      "step": 142000
    },
    {
      "epoch": 12.45,
      "learning_rate": 3.9489217836257314e-05,
      "loss": 1.2438,
      "step": 142500
    },
    {
      "epoch": 12.45,
      "eval_loss": 1.2091211080551147,
      "eval_runtime": 798.8782,
      "eval_samples_per_second": 96.505,
      "eval_steps_per_second": 6.032,
      "step": 142500
    },
    {
      "epoch": 12.5,
      "learning_rate": 3.9260782163742696e-05,
      "loss": 1.2505,
      "step": 143000
    },
    {
      "epoch": 12.54,
      "learning_rate": 3.903234649122807e-05,
      "loss": 1.2397,
      "step": 143500
    },
    {
      "epoch": 12.58,
      "learning_rate": 3.8803910818713455e-05,
      "loss": 1.248,
      "step": 144000
    },
    {
      "epoch": 12.63,
      "learning_rate": 3.857547514619883e-05,
      "loss": 1.2403,
      "step": 144500
    },
    {
      "epoch": 12.67,
      "learning_rate": 3.834703947368421e-05,
      "loss": 1.2445,
      "step": 145000
    },
    {
      "epoch": 12.67,
      "eval_loss": 1.200273036956787,
      "eval_runtime": 795.7818,
      "eval_samples_per_second": 96.881,
      "eval_steps_per_second": 6.056,
      "step": 145000
    },
    {
      "epoch": 12.71,
      "learning_rate": 3.8118603801169596e-05,
      "loss": 1.246,
      "step": 145500
    },
    {
      "epoch": 12.76,
      "learning_rate": 3.789016812865497e-05,
      "loss": 1.2457,
      "step": 146000
    },
    {
      "epoch": 12.8,
      "learning_rate": 3.7661732456140354e-05,
      "loss": 1.2365,
      "step": 146500
    },
    {
      "epoch": 12.85,
      "learning_rate": 3.743375365497076e-05,
      "loss": 1.2353,
      "step": 147000
    },
    {
      "epoch": 12.89,
      "learning_rate": 3.720531798245614e-05,
      "loss": 1.2314,
      "step": 147500
    },
    {
      "epoch": 12.89,
      "eval_loss": 1.195684790611267,
      "eval_runtime": 798.8193,
      "eval_samples_per_second": 96.512,
      "eval_steps_per_second": 6.033,
      "step": 147500
    },
    {
      "epoch": 12.93,
      "learning_rate": 3.6976882309941525e-05,
      "loss": 1.2283,
      "step": 148000
    },
    {
      "epoch": 12.98,
      "learning_rate": 3.67484466374269e-05,
      "loss": 1.2423,
      "step": 148500
    },
    {
      "epoch": 13.02,
      "learning_rate": 3.652001096491228e-05,
      "loss": 1.2213,
      "step": 149000
    },
    {
      "epoch": 13.06,
      "learning_rate": 3.629203216374269e-05,
      "loss": 1.2237,
      "step": 149500
    },
    {
      "epoch": 13.11,
      "learning_rate": 3.606359649122807e-05,
      "loss": 1.2188,
      "step": 150000
    },
    {
      "epoch": 13.11,
      "eval_loss": 1.1842824220657349,
      "eval_runtime": 798.645,
      "eval_samples_per_second": 96.534,
      "eval_steps_per_second": 6.034,
      "step": 150000
    },
    {
      "epoch": 13.15,
      "learning_rate": 3.5835160818713454e-05,
      "loss": 1.2147,
      "step": 150500
    },
    {
      "epoch": 13.19,
      "learning_rate": 3.560672514619883e-05,
      "loss": 1.2252,
      "step": 151000
    },
    {
      "epoch": 13.24,
      "learning_rate": 3.537874634502924e-05,
      "loss": 1.2265,
      "step": 151500
    },
    {
      "epoch": 13.28,
      "learning_rate": 3.515031067251462e-05,
      "loss": 1.2158,
      "step": 152000
    },
    {
      "epoch": 13.33,
      "learning_rate": 3.492187500000001e-05,
      "loss": 1.2071,
      "step": 152500
    },
    {
      "epoch": 13.33,
      "eval_loss": 1.1805495023727417,
      "eval_runtime": 799.0741,
      "eval_samples_per_second": 96.482,
      "eval_steps_per_second": 6.031,
      "step": 152500
    },
    {
      "epoch": 13.37,
      "learning_rate": 3.469343932748538e-05,
      "loss": 1.2199,
      "step": 153000
    },
    {
      "epoch": 13.41,
      "learning_rate": 3.4465003654970765e-05,
      "loss": 1.2059,
      "step": 153500
    },
    {
      "epoch": 13.46,
      "learning_rate": 3.42374817251462e-05,
      "loss": 1.2106,
      "step": 154000
    },
    {
      "epoch": 13.5,
      "learning_rate": 3.400904605263158e-05,
      "loss": 1.2129,
      "step": 154500
    },
    {
      "epoch": 13.54,
      "learning_rate": 3.378061038011696e-05,
      "loss": 1.2123,
      "step": 155000
    },
    {
      "epoch": 13.54,
      "eval_loss": 1.1766316890716553,
      "eval_runtime": 795.4577,
      "eval_samples_per_second": 96.92,
      "eval_steps_per_second": 6.058,
      "step": 155000
    },
    {
      "epoch": 13.59,
      "learning_rate": 3.355217470760234e-05,
      "loss": 1.2058,
      "step": 155500
    },
    {
      "epoch": 13.63,
      "learning_rate": 3.332373903508772e-05,
      "loss": 1.204,
      "step": 156000
    },
    {
      "epoch": 13.68,
      "learning_rate": 3.30953033625731e-05,
      "loss": 1.1967,
      "step": 156500
    },
    {
      "epoch": 13.72,
      "learning_rate": 3.286686769005848e-05,
      "loss": 1.2056,
      "step": 157000
    },
    {
      "epoch": 13.76,
      "learning_rate": 3.2638432017543865e-05,
      "loss": 1.2016,
      "step": 157500
    },
    {
      "epoch": 13.76,
      "eval_loss": 1.166089653968811,
      "eval_runtime": 799.396,
      "eval_samples_per_second": 96.443,
      "eval_steps_per_second": 6.028,
      "step": 157500
    },
    {
      "epoch": 13.81,
      "learning_rate": 3.241045321637427e-05,
      "loss": 1.2101,
      "step": 158000
    },
    {
      "epoch": 13.85,
      "learning_rate": 3.2182017543859646e-05,
      "loss": 1.2096,
      "step": 158500
    },
    {
      "epoch": 13.89,
      "learning_rate": 3.195358187134503e-05,
      "loss": 1.1916,
      "step": 159000
    },
    {
      "epoch": 13.94,
      "learning_rate": 3.172514619883041e-05,
      "loss": 1.2013,
      "step": 159500
    },
    {
      "epoch": 13.98,
      "learning_rate": 3.1496710526315794e-05,
      "loss": 1.2079,
      "step": 160000
    },
    {
      "epoch": 13.98,
      "eval_loss": 1.1625038385391235,
      "eval_runtime": 801.5699,
      "eval_samples_per_second": 96.181,
      "eval_steps_per_second": 6.012,
      "step": 160000
    },
    {
      "epoch": 14.02,
      "learning_rate": 3.12687317251462e-05,
      "loss": 1.1861,
      "step": 160500
    },
    {
      "epoch": 14.07,
      "learning_rate": 3.104029605263158e-05,
      "loss": 1.1892,
      "step": 161000
    },
    {
      "epoch": 14.11,
      "learning_rate": 3.081186038011696e-05,
      "loss": 1.1907,
      "step": 161500
    },
    {
      "epoch": 14.16,
      "learning_rate": 3.058342470760234e-05,
      "loss": 1.1865,
      "step": 162000
    },
    {
      "epoch": 14.2,
      "learning_rate": 3.035544590643275e-05,
      "loss": 1.1884,
      "step": 162500
    },
    {
      "epoch": 14.2,
      "eval_loss": 1.1524909734725952,
      "eval_runtime": 798.9251,
      "eval_samples_per_second": 96.5,
      "eval_steps_per_second": 6.032,
      "step": 162500
    },
    {
      "epoch": 14.24,
      "learning_rate": 3.012701023391813e-05,
      "loss": 1.1843,
      "step": 163000
    },
    {
      "epoch": 14.29,
      "learning_rate": 2.9898574561403508e-05,
      "loss": 1.1913,
      "step": 163500
    },
    {
      "epoch": 14.33,
      "learning_rate": 2.9670138888888887e-05,
      "loss": 1.1804,
      "step": 164000
    },
    {
      "epoch": 14.37,
      "learning_rate": 2.9441703216374273e-05,
      "loss": 1.1778,
      "step": 164500
    },
    {
      "epoch": 14.42,
      "learning_rate": 2.921372441520468e-05,
      "loss": 1.177,
      "step": 165000
    },
    {
      "epoch": 14.42,
      "eval_loss": 1.14187490940094,
      "eval_runtime": 799.1692,
      "eval_samples_per_second": 96.47,
      "eval_steps_per_second": 6.03,
      "step": 165000
    },
    {
      "epoch": 14.46,
      "learning_rate": 2.8985288742690058e-05,
      "loss": 1.1769,
      "step": 165500
    },
    {
      "epoch": 14.51,
      "learning_rate": 2.8756853070175437e-05,
      "loss": 1.1869,
      "step": 166000
    },
    {
      "epoch": 14.55,
      "learning_rate": 2.8528417397660823e-05,
      "loss": 1.1713,
      "step": 166500
    },
    {
      "epoch": 14.59,
      "learning_rate": 2.830043859649123e-05,
      "loss": 1.1749,
      "step": 167000
    },
    {
      "epoch": 14.64,
      "learning_rate": 2.807200292397661e-05,
      "loss": 1.1793,
      "step": 167500
    },
    {
      "epoch": 14.64,
      "eval_loss": 1.1453900337219238,
      "eval_runtime": 795.1979,
      "eval_samples_per_second": 96.952,
      "eval_steps_per_second": 6.06,
      "step": 167500
    },
    {
      "epoch": 14.68,
      "learning_rate": 2.784356725146199e-05,
      "loss": 1.1743,
      "step": 168000
    },
    {
      "epoch": 14.72,
      "learning_rate": 2.7615131578947366e-05,
      "loss": 1.1727,
      "step": 168500
    },
    {
      "epoch": 14.77,
      "learning_rate": 2.738715277777778e-05,
      "loss": 1.1743,
      "step": 169000
    },
    {
      "epoch": 14.81,
      "learning_rate": 2.715871710526316e-05,
      "loss": 1.1702,
      "step": 169500
    },
    {
      "epoch": 14.85,
      "learning_rate": 2.693028143274854e-05,
      "loss": 1.173,
      "step": 170000
    },
    {
      "epoch": 14.85,
      "eval_loss": 1.1378823518753052,
      "eval_runtime": 799.1257,
      "eval_samples_per_second": 96.475,
      "eval_steps_per_second": 6.03,
      "step": 170000
    },
    {
      "epoch": 14.9,
      "learning_rate": 2.670184576023392e-05,
      "loss": 1.1762,
      "step": 170500
    },
    {
      "epoch": 14.94,
      "learning_rate": 2.6473410087719298e-05,
      "loss": 1.1578,
      "step": 171000
    },
    {
      "epoch": 14.99,
      "learning_rate": 2.624497441520468e-05,
      "loss": 1.1615,
      "step": 171500
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.6017452485380116e-05,
      "loss": 1.1529,
      "step": 172000
    },
    {
      "epoch": 15.07,
      "learning_rate": 2.5789016812865495e-05,
      "loss": 1.1502,
      "step": 172500
    },
    {
      "epoch": 15.07,
      "eval_loss": 1.1370670795440674,
      "eval_runtime": 798.9647,
      "eval_samples_per_second": 96.495,
      "eval_steps_per_second": 6.032,
      "step": 172500
    },
    {
      "epoch": 15.12,
      "learning_rate": 2.556058114035088e-05,
      "loss": 1.1504,
      "step": 173000
    },
    {
      "epoch": 15.16,
      "learning_rate": 2.533214546783626e-05,
      "loss": 1.1534,
      "step": 173500
    },
    {
      "epoch": 15.2,
      "learning_rate": 2.510370979532164e-05,
      "loss": 1.1597,
      "step": 174000
    },
    {
      "epoch": 15.25,
      "learning_rate": 2.4875730994152045e-05,
      "loss": 1.1549,
      "step": 174500
    },
    {
      "epoch": 15.29,
      "learning_rate": 2.4647295321637428e-05,
      "loss": 1.1504,
      "step": 175000
    },
    {
      "epoch": 15.29,
      "eval_loss": 1.1295256614685059,
      "eval_runtime": 798.4053,
      "eval_samples_per_second": 96.562,
      "eval_steps_per_second": 6.036,
      "step": 175000
    },
    {
      "epoch": 15.34,
      "learning_rate": 2.4418859649122807e-05,
      "loss": 1.1583,
      "step": 175500
    },
    {
      "epoch": 15.38,
      "learning_rate": 2.4190423976608186e-05,
      "loss": 1.153,
      "step": 176000
    },
    {
      "epoch": 15.42,
      "learning_rate": 2.3962445175438598e-05,
      "loss": 1.1572,
      "step": 176500
    },
    {
      "epoch": 15.47,
      "learning_rate": 2.3734009502923977e-05,
      "loss": 1.1526,
      "step": 177000
    },
    {
      "epoch": 15.51,
      "learning_rate": 2.3505573830409357e-05,
      "loss": 1.146,
      "step": 177500
    },
    {
      "epoch": 15.51,
      "eval_loss": 1.1202627420425415,
      "eval_runtime": 793.634,
      "eval_samples_per_second": 97.143,
      "eval_steps_per_second": 6.072,
      "step": 177500
    },
    {
      "epoch": 15.55,
      "learning_rate": 2.3277138157894736e-05,
      "loss": 1.1595,
      "step": 178000
    },
    {
      "epoch": 15.6,
      "learning_rate": 2.304870248538012e-05,
      "loss": 1.1496,
      "step": 178500
    },
    {
      "epoch": 15.64,
      "learning_rate": 2.2820266812865498e-05,
      "loss": 1.1466,
      "step": 179000
    },
    {
      "epoch": 15.69,
      "learning_rate": 2.2591831140350877e-05,
      "loss": 1.1483,
      "step": 179500
    },
    {
      "epoch": 15.73,
      "learning_rate": 2.236339546783626e-05,
      "loss": 1.1487,
      "step": 180000
    },
    {
      "epoch": 15.73,
      "eval_loss": 1.1136623620986938,
      "eval_runtime": 798.4421,
      "eval_samples_per_second": 96.558,
      "eval_steps_per_second": 6.036,
      "step": 180000
    },
    {
      "epoch": 15.77,
      "learning_rate": 2.2135416666666668e-05,
      "loss": 1.1383,
      "step": 180500
    },
    {
      "epoch": 15.82,
      "learning_rate": 2.1906980994152047e-05,
      "loss": 1.1463,
      "step": 181000
    },
    {
      "epoch": 15.86,
      "learning_rate": 2.1678545321637427e-05,
      "loss": 1.1351,
      "step": 181500
    },
    {
      "epoch": 15.9,
      "learning_rate": 2.1450109649122806e-05,
      "loss": 1.1432,
      "step": 182000
    },
    {
      "epoch": 15.95,
      "learning_rate": 2.122167397660819e-05,
      "loss": 1.1329,
      "step": 182500
    },
    {
      "epoch": 15.95,
      "eval_loss": 1.1195679903030396,
      "eval_runtime": 798.9095,
      "eval_samples_per_second": 96.502,
      "eval_steps_per_second": 6.032,
      "step": 182500
    },
    {
      "epoch": 15.99,
      "learning_rate": 2.0993695175438597e-05,
      "loss": 1.1394,
      "step": 183000
    },
    {
      "epoch": 16.03,
      "learning_rate": 2.0765716374269006e-05,
      "loss": 1.1325,
      "step": 183500
    },
    {
      "epoch": 16.08,
      "learning_rate": 2.0537280701754385e-05,
      "loss": 1.1326,
      "step": 184000
    },
    {
      "epoch": 16.12,
      "learning_rate": 2.0308845029239768e-05,
      "loss": 1.1353,
      "step": 184500
    },
    {
      "epoch": 16.17,
      "learning_rate": 2.0080409356725147e-05,
      "loss": 1.1259,
      "step": 185000
    },
    {
      "epoch": 16.17,
      "eval_loss": 1.1075031757354736,
      "eval_runtime": 793.7166,
      "eval_samples_per_second": 97.133,
      "eval_steps_per_second": 6.071,
      "step": 185000
    },
    {
      "epoch": 16.21,
      "learning_rate": 1.985197368421053e-05,
      "loss": 1.129,
      "step": 185500
    },
    {
      "epoch": 16.25,
      "learning_rate": 1.962353801169591e-05,
      "loss": 1.1308,
      "step": 186000
    },
    {
      "epoch": 16.3,
      "learning_rate": 1.9395102339181288e-05,
      "loss": 1.1335,
      "step": 186500
    },
    {
      "epoch": 16.34,
      "learning_rate": 1.9166666666666667e-05,
      "loss": 1.1374,
      "step": 187000
    },
    {
      "epoch": 16.38,
      "learning_rate": 1.8938687865497076e-05,
      "loss": 1.1287,
      "step": 187500
    },
    {
      "epoch": 16.38,
      "eval_loss": 1.1037468910217285,
      "eval_runtime": 802.495,
      "eval_samples_per_second": 96.07,
      "eval_steps_per_second": 6.005,
      "step": 187500
    },
    {
      "epoch": 16.43,
      "learning_rate": 1.871025219298246e-05,
      "loss": 1.1144,
      "step": 188000
    },
    {
      "epoch": 16.47,
      "learning_rate": 1.8481816520467838e-05,
      "loss": 1.1244,
      "step": 188500
    },
    {
      "epoch": 16.52,
      "learning_rate": 1.8253380847953217e-05,
      "loss": 1.1198,
      "step": 189000
    },
    {
      "epoch": 16.56,
      "learning_rate": 1.80249451754386e-05,
      "loss": 1.118,
      "step": 189500
    },
    {
      "epoch": 16.6,
      "learning_rate": 1.7796966374269005e-05,
      "loss": 1.126,
      "step": 190000
    },
    {
      "epoch": 16.6,
      "eval_loss": 1.1041918992996216,
      "eval_runtime": 798.7101,
      "eval_samples_per_second": 96.526,
      "eval_steps_per_second": 6.033,
      "step": 190000
    },
    {
      "epoch": 16.65,
      "learning_rate": 1.7568987573099414e-05,
      "loss": 1.115,
      "step": 190500
    },
    {
      "epoch": 16.69,
      "learning_rate": 1.7340551900584793e-05,
      "loss": 1.1211,
      "step": 191000
    },
    {
      "epoch": 16.73,
      "learning_rate": 1.7112116228070176e-05,
      "loss": 1.1191,
      "step": 191500
    },
    {
      "epoch": 16.78,
      "learning_rate": 1.6883680555555555e-05,
      "loss": 1.1095,
      "step": 192000
    },
    {
      "epoch": 16.82,
      "learning_rate": 1.6655244883040934e-05,
      "loss": 1.1199,
      "step": 192500
    },
    {
      "epoch": 16.82,
      "eval_loss": 1.0953432321548462,
      "eval_runtime": 793.9545,
      "eval_samples_per_second": 97.104,
      "eval_steps_per_second": 6.07,
      "step": 192500
    },
    {
      "epoch": 16.86,
      "learning_rate": 1.6426809210526317e-05,
      "loss": 1.1304,
      "step": 193000
    },
    {
      "epoch": 16.91,
      "learning_rate": 1.6198373538011696e-05,
      "loss": 1.1208,
      "step": 193500
    },
    {
      "epoch": 16.95,
      "learning_rate": 1.5969937865497075e-05,
      "loss": 1.1123,
      "step": 194000
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.5741502192982458e-05,
      "loss": 1.1146,
      "step": 194500
    },
    {
      "epoch": 17.04,
      "learning_rate": 1.5513523391812867e-05,
      "loss": 1.1072,
      "step": 195000
    },
    {
      "epoch": 17.04,
      "eval_loss": 1.0885491371154785,
      "eval_runtime": 799.287,
      "eval_samples_per_second": 96.456,
      "eval_steps_per_second": 6.029,
      "step": 195000
    },
    {
      "epoch": 17.08,
      "learning_rate": 1.5285544590643276e-05,
      "loss": 1.1109,
      "step": 195500
    },
    {
      "epoch": 17.13,
      "learning_rate": 1.5057108918128656e-05,
      "loss": 1.1012,
      "step": 196000
    },
    {
      "epoch": 17.17,
      "learning_rate": 1.4828673245614036e-05,
      "loss": 1.112,
      "step": 196500
    },
    {
      "epoch": 17.21,
      "learning_rate": 1.4600237573099415e-05,
      "loss": 1.1014,
      "step": 197000
    },
    {
      "epoch": 17.26,
      "learning_rate": 1.4371801900584797e-05,
      "loss": 1.1043,
      "step": 197500
    },
    {
      "epoch": 17.26,
      "eval_loss": 1.0877279043197632,
      "eval_runtime": 799.1548,
      "eval_samples_per_second": 96.472,
      "eval_steps_per_second": 6.03,
      "step": 197500
    },
    {
      "epoch": 17.3,
      "learning_rate": 1.4143366228070177e-05,
      "loss": 1.0985,
      "step": 198000
    },
    {
      "epoch": 17.35,
      "learning_rate": 1.3914930555555556e-05,
      "loss": 1.1058,
      "step": 198500
    },
    {
      "epoch": 17.39,
      "learning_rate": 1.3686494883040937e-05,
      "loss": 1.0983,
      "step": 199000
    },
    {
      "epoch": 17.43,
      "learning_rate": 1.3458516081871344e-05,
      "loss": 1.1015,
      "step": 199500
    },
    {
      "epoch": 17.48,
      "learning_rate": 1.3230080409356726e-05,
      "loss": 1.1007,
      "step": 200000
    },
    {
      "epoch": 17.48,
      "eval_loss": 1.0834912061691284,
      "eval_runtime": 795.2591,
      "eval_samples_per_second": 96.945,
      "eval_steps_per_second": 6.06,
      "step": 200000
    },
    {
      "epoch": 17.52,
      "learning_rate": 1.3001644736842106e-05,
      "loss": 1.1048,
      "step": 200500
    },
    {
      "epoch": 17.56,
      "learning_rate": 1.2773209064327485e-05,
      "loss": 1.1018,
      "step": 201000
    },
    {
      "epoch": 17.61,
      "learning_rate": 1.2544773391812867e-05,
      "loss": 1.1071,
      "step": 201500
    },
    {
      "epoch": 17.65,
      "learning_rate": 1.2316794590643276e-05,
      "loss": 1.1002,
      "step": 202000
    },
    {
      "epoch": 17.69,
      "learning_rate": 1.2088358918128655e-05,
      "loss": 1.0879,
      "step": 202500
    },
    {
      "epoch": 17.69,
      "eval_loss": 1.0818740129470825,
      "eval_runtime": 801.1636,
      "eval_samples_per_second": 96.23,
      "eval_steps_per_second": 6.015,
      "step": 202500
    },
    {
      "epoch": 17.74,
      "learning_rate": 1.1859923245614035e-05,
      "loss": 1.097,
      "step": 203000
    },
    {
      "epoch": 17.78,
      "learning_rate": 1.1631487573099415e-05,
      "loss": 1.0993,
      "step": 203500
    },
    {
      "epoch": 17.83,
      "learning_rate": 1.1403051900584796e-05,
      "loss": 1.0991,
      "step": 204000
    },
    {
      "epoch": 17.87,
      "learning_rate": 1.1174616228070176e-05,
      "loss": 1.0953,
      "step": 204500
    },
    {
      "epoch": 17.91,
      "learning_rate": 1.0946637426900586e-05,
      "loss": 1.1,
      "step": 205000
    },
    {
      "epoch": 17.91,
      "eval_loss": 1.0743597745895386,
      "eval_runtime": 798.4761,
      "eval_samples_per_second": 96.554,
      "eval_steps_per_second": 6.035,
      "step": 205000
    },
    {
      "epoch": 17.96,
      "learning_rate": 1.0718201754385965e-05,
      "loss": 1.0917,
      "step": 205500
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.0489766081871345e-05,
      "loss": 1.0929,
      "step": 206000
    },
    {
      "epoch": 18.04,
      "learning_rate": 1.0261330409356725e-05,
      "loss": 1.089,
      "step": 206500
    },
    {
      "epoch": 18.09,
      "learning_rate": 1.0033351608187134e-05,
      "loss": 1.0919,
      "step": 207000
    },
    {
      "epoch": 18.13,
      "learning_rate": 9.804915935672515e-06,
      "loss": 1.0863,
      "step": 207500
    },
    {
      "epoch": 18.13,
      "eval_loss": 1.0773507356643677,
      "eval_runtime": 798.9691,
      "eval_samples_per_second": 96.494,
      "eval_steps_per_second": 6.032,
      "step": 207500
    },
    {
      "epoch": 18.18,
      "learning_rate": 9.576480263157896e-06,
      "loss": 1.0895,
      "step": 208000
    },
    {
      "epoch": 18.22,
      "learning_rate": 9.348044590643275e-06,
      "loss": 1.0782,
      "step": 208500
    },
    {
      "epoch": 18.26,
      "learning_rate": 9.120065789473684e-06,
      "loss": 1.0847,
      "step": 209000
    },
    {
      "epoch": 18.31,
      "learning_rate": 8.891630116959063e-06,
      "loss": 1.0819,
      "step": 209500
    },
    {
      "epoch": 18.35,
      "learning_rate": 8.663194444444444e-06,
      "loss": 1.087,
      "step": 210000
    },
    {
      "epoch": 18.35,
      "eval_loss": 1.0759390592575073,
      "eval_runtime": 799.2137,
      "eval_samples_per_second": 96.465,
      "eval_steps_per_second": 6.03,
      "step": 210000
    },
    {
      "epoch": 18.39,
      "learning_rate": 8.434758771929825e-06,
      "loss": 1.0857,
      "step": 210500
    },
    {
      "epoch": 18.44,
      "learning_rate": 8.206779970760234e-06,
      "loss": 1.0831,
      "step": 211000
    },
    {
      "epoch": 18.48,
      "learning_rate": 7.978344298245615e-06,
      "loss": 1.0752,
      "step": 211500
    },
    {
      "epoch": 18.52,
      "learning_rate": 7.749908625730996e-06,
      "loss": 1.0865,
      "step": 212000
    },
    {
      "epoch": 18.57,
      "learning_rate": 7.521472953216374e-06,
      "loss": 1.0755,
      "step": 212500
    },
    {
      "epoch": 18.57,
      "eval_loss": 1.061818242073059,
      "eval_runtime": 795.7294,
      "eval_samples_per_second": 96.887,
      "eval_steps_per_second": 6.056,
      "step": 212500
    },
    {
      "epoch": 18.61,
      "learning_rate": 7.293037280701755e-06,
      "loss": 1.0806,
      "step": 213000
    },
    {
      "epoch": 18.66,
      "learning_rate": 7.064601608187135e-06,
      "loss": 1.0826,
      "step": 213500
    },
    {
      "epoch": 18.7,
      "learning_rate": 6.836165935672514e-06,
      "loss": 1.0854,
      "step": 214000
    },
    {
      "epoch": 18.74,
      "learning_rate": 6.607730263157895e-06,
      "loss": 1.0783,
      "step": 214500
    },
    {
      "epoch": 18.79,
      "learning_rate": 6.379294590643276e-06,
      "loss": 1.0832,
      "step": 215000
    },
    {
      "epoch": 18.79,
      "eval_loss": 1.0627787113189697,
      "eval_runtime": 799.1419,
      "eval_samples_per_second": 96.473,
      "eval_steps_per_second": 6.03,
      "step": 215000
    },
    {
      "epoch": 18.83,
      "learning_rate": 6.151315789473684e-06,
      "loss": 1.0786,
      "step": 215500
    },
    {
      "epoch": 18.87,
      "learning_rate": 5.922880116959065e-06,
      "loss": 1.0719,
      "step": 216000
    },
    {
      "epoch": 18.92,
      "learning_rate": 5.694444444444445e-06,
      "loss": 1.0768,
      "step": 216500
    },
    {
      "epoch": 18.96,
      "learning_rate": 5.466008771929825e-06,
      "loss": 1.075,
      "step": 217000
    },
    {
      "epoch": 19.01,
      "learning_rate": 5.237573099415205e-06,
      "loss": 1.0771,
      "step": 217500
    },
    {
      "epoch": 19.01,
      "eval_loss": 1.061143398284912,
      "eval_runtime": 799.1588,
      "eval_samples_per_second": 96.471,
      "eval_steps_per_second": 6.03,
      "step": 217500
    },
    {
      "epoch": 19.05,
      "learning_rate": 5.009137426900585e-06,
      "loss": 1.0815,
      "step": 218000
    },
    {
      "epoch": 19.09,
      "learning_rate": 4.781158625730994e-06,
      "loss": 1.0656,
      "step": 218500
    },
    {
      "epoch": 19.14,
      "learning_rate": 4.552722953216375e-06,
      "loss": 1.0718,
      "step": 219000
    },
    {
      "epoch": 19.18,
      "learning_rate": 4.324287280701755e-06,
      "loss": 1.0697,
      "step": 219500
    },
    {
      "epoch": 19.22,
      "learning_rate": 4.095851608187135e-06,
      "loss": 1.0703,
      "step": 220000
    },
    {
      "epoch": 19.22,
      "eval_loss": 1.0554795265197754,
      "eval_runtime": 793.6761,
      "eval_samples_per_second": 97.138,
      "eval_steps_per_second": 6.072,
      "step": 220000
    },
    {
      "epoch": 19.27,
      "learning_rate": 3.867415935672515e-06,
      "loss": 1.0681,
      "step": 220500
    },
    {
      "epoch": 19.31,
      "learning_rate": 3.638980263157895e-06,
      "loss": 1.0706,
      "step": 221000
    },
    {
      "epoch": 19.36,
      "learning_rate": 3.4110014619883042e-06,
      "loss": 1.0671,
      "step": 221500
    },
    {
      "epoch": 19.4,
      "learning_rate": 3.1825657894736842e-06,
      "loss": 1.063,
      "step": 222000
    },
    {
      "epoch": 19.44,
      "learning_rate": 2.9541301169590643e-06,
      "loss": 1.069,
      "step": 222500
    },
    {
      "epoch": 19.44,
      "eval_loss": 1.0551636219024658,
      "eval_runtime": 799.5142,
      "eval_samples_per_second": 96.429,
      "eval_steps_per_second": 6.027,
      "step": 222500
    },
    {
      "epoch": 19.49,
      "learning_rate": 2.7256944444444447e-06,
      "loss": 1.0662,
      "step": 223000
    },
    {
      "epoch": 19.53,
      "learning_rate": 2.4972587719298248e-06,
      "loss": 1.0646,
      "step": 223500
    },
    {
      "epoch": 19.57,
      "learning_rate": 2.2688230994152048e-06,
      "loss": 1.0651,
      "step": 224000
    },
    {
      "epoch": 19.62,
      "learning_rate": 2.040844298245614e-06,
      "loss": 1.0729,
      "step": 224500
    },
    {
      "epoch": 19.66,
      "learning_rate": 1.8124086257309944e-06,
      "loss": 1.0706,
      "step": 225000
    },
    {
      "epoch": 19.66,
      "eval_loss": 1.0508904457092285,
      "eval_runtime": 799.3295,
      "eval_samples_per_second": 96.451,
      "eval_steps_per_second": 6.029,
      "step": 225000
    },
    {
      "epoch": 19.7,
      "learning_rate": 1.5839729532163744e-06,
      "loss": 1.0639,
      "step": 225500
    },
    {
      "epoch": 19.75,
      "learning_rate": 1.3555372807017544e-06,
      "loss": 1.0681,
      "step": 226000
    },
    {
      "epoch": 19.79,
      "learning_rate": 1.1271016081871344e-06,
      "loss": 1.0707,
      "step": 226500
    },
    {
      "epoch": 19.84,
      "learning_rate": 8.991228070175439e-07,
      "loss": 1.0674,
      "step": 227000
    },
    {
      "epoch": 19.88,
      "learning_rate": 6.70687134502924e-07,
      "loss": 1.0633,
      "step": 227500
    },
    {
      "epoch": 19.88,
      "eval_loss": 1.0465439558029175,
      "eval_runtime": 799.9551,
      "eval_samples_per_second": 96.375,
      "eval_steps_per_second": 6.024,
      "step": 227500
    },
    {
      "epoch": 19.92,
      "learning_rate": 4.4225146198830405e-07,
      "loss": 1.06,
      "step": 228000
    },
    {
      "epoch": 19.97,
      "learning_rate": 2.1381578947368424e-07,
      "loss": 1.0695,
      "step": 228500
    },
    {
      "epoch": 20.0,
      "step": 228880,
      "total_flos": 2.8703856730616376e+18,
      "train_loss": 0.5992164485148723,
      "train_runtime": 340057.0464,
      "train_samples_per_second": 86.151,
      "train_steps_per_second": 0.673
    }
  ],
  "max_steps": 228880,
  "num_train_epochs": 20,
  "total_flos": 2.8703856730616376e+18,
  "trial_name": null,
  "trial_params": null
}