{
  "best_metric": 0.21237443387508392,
  "best_model_checkpoint": "finetuned_entity_categorical_classification/checkpoint-4728",
  "epoch": 15.0,
  "eval_steps": 500,
  "global_step": 23640,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "learning_rate": 1.957698815566836e-05,
      "loss": 1.5567,
      "step": 500
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.915397631133672e-05,
      "loss": 0.3944,
      "step": 1000
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.873096446700508e-05,
      "loss": 0.2773,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9374900840869427,
      "eval_loss": 0.2187376469373703,
      "eval_runtime": 2.2114,
      "eval_samples_per_second": 2850.256,
      "eval_steps_per_second": 178.169,
      "step": 1576
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.830795262267344e-05,
      "loss": 0.1997,
      "step": 2000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.7884940778341796e-05,
      "loss": 0.202,
      "step": 2500
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.7461928934010152e-05,
      "loss": 0.1797,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9452641599238458,
      "eval_loss": 0.22844311594963074,
      "eval_runtime": 2.2403,
      "eval_samples_per_second": 2813.437,
      "eval_steps_per_second": 175.868,
      "step": 3152
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.7038917089678512e-05,
      "loss": 0.148,
      "step": 3500
    },
    {
      "epoch": 2.54,
      "learning_rate": 1.661590524534687e-05,
      "loss": 0.1357,
      "step": 4000
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.619289340101523e-05,
      "loss": 0.1525,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9481199428843408,
      "eval_loss": 0.21237443387508392,
      "eval_runtime": 2.2281,
      "eval_samples_per_second": 2828.834,
      "eval_steps_per_second": 176.83,
      "step": 4728
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.576988155668359e-05,
      "loss": 0.1218,
      "step": 5000
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.5346869712351946e-05,
      "loss": 0.1147,
      "step": 5500
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.4923857868020306e-05,
      "loss": 0.1195,
      "step": 6000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.94859590671109,
      "eval_loss": 0.2216227501630783,
      "eval_runtime": 2.2255,
      "eval_samples_per_second": 2832.161,
      "eval_steps_per_second": 177.038,
      "step": 6304
    },
    {
      "epoch": 4.12,
      "learning_rate": 1.4500846023688663e-05,
      "loss": 0.114,
      "step": 6500
    },
    {
      "epoch": 4.44,
      "learning_rate": 1.4077834179357023e-05,
      "loss": 0.104,
      "step": 7000
    },
    {
      "epoch": 4.76,
      "learning_rate": 1.3654822335025382e-05,
      "loss": 0.0936,
      "step": 7500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9462160875773441,
      "eval_loss": 0.24313929677009583,
      "eval_runtime": 2.1752,
      "eval_samples_per_second": 2897.6,
      "eval_steps_per_second": 181.129,
      "step": 7880
    },
    {
      "epoch": 5.08,
      "learning_rate": 1.323181049069374e-05,
      "loss": 0.099,
      "step": 8000
    },
    {
      "epoch": 5.39,
      "learning_rate": 1.28087986463621e-05,
      "loss": 0.0929,
      "step": 8500
    },
    {
      "epoch": 5.71,
      "learning_rate": 1.2385786802030457e-05,
      "loss": 0.0872,
      "step": 9000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9465333967951769,
      "eval_loss": 0.25265371799468994,
      "eval_runtime": 2.2222,
      "eval_samples_per_second": 2836.351,
      "eval_steps_per_second": 177.3,
      "step": 9456
    },
    {
      "epoch": 6.03,
      "learning_rate": 1.1962774957698817e-05,
      "loss": 0.0963,
      "step": 9500
    },
    {
      "epoch": 6.35,
      "learning_rate": 1.1539763113367176e-05,
      "loss": 0.0733,
      "step": 10000
    },
    {
      "epoch": 6.66,
      "learning_rate": 1.1116751269035532e-05,
      "loss": 0.0812,
      "step": 10500
    },
    {
      "epoch": 6.98,
      "learning_rate": 1.0693739424703892e-05,
      "loss": 0.0929,
      "step": 11000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9490718705378391,
      "eval_loss": 0.23417465388774872,
      "eval_runtime": 2.1699,
      "eval_samples_per_second": 2904.721,
      "eval_steps_per_second": 181.574,
      "step": 11032
    },
    {
      "epoch": 7.3,
      "learning_rate": 1.0270727580372251e-05,
      "loss": 0.0629,
      "step": 11500
    },
    {
      "epoch": 7.61,
      "learning_rate": 9.84771573604061e-06,
      "loss": 0.0802,
      "step": 12000
    },
    {
      "epoch": 7.93,
      "learning_rate": 9.424703891708968e-06,
      "loss": 0.0757,
      "step": 12500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9490718705378391,
      "eval_loss": 0.24814845621585846,
      "eval_runtime": 2.241,
      "eval_samples_per_second": 2812.616,
      "eval_steps_per_second": 175.816,
      "step": 12608
    },
    {
      "epoch": 8.25,
      "learning_rate": 9.001692047377328e-06,
      "loss": 0.063,
      "step": 13000
    },
    {
      "epoch": 8.57,
      "learning_rate": 8.578680203045686e-06,
      "loss": 0.0619,
      "step": 13500
    },
    {
      "epoch": 8.88,
      "learning_rate": 8.155668358714045e-06,
      "loss": 0.0673,
      "step": 14000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9474853244486753,
      "eval_loss": 0.2526280879974365,
      "eval_runtime": 2.1778,
      "eval_samples_per_second": 2894.229,
      "eval_steps_per_second": 180.918,
      "step": 14184
    },
    {
      "epoch": 9.2,
      "learning_rate": 7.732656514382403e-06,
      "loss": 0.0599,
      "step": 14500
    },
    {
      "epoch": 9.52,
      "learning_rate": 7.309644670050762e-06,
      "loss": 0.0605,
      "step": 15000
    },
    {
      "epoch": 9.84,
      "learning_rate": 6.886632825719121e-06,
      "loss": 0.0563,
      "step": 15500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9482785974932572,
      "eval_loss": 0.2582007646560669,
      "eval_runtime": 2.1873,
      "eval_samples_per_second": 2881.619,
      "eval_steps_per_second": 180.13,
      "step": 15760
    },
    {
      "epoch": 10.15,
      "learning_rate": 6.4636209813874795e-06,
      "loss": 0.0581,
      "step": 16000
    },
    {
      "epoch": 10.47,
      "learning_rate": 6.040609137055839e-06,
      "loss": 0.0614,
      "step": 16500
    },
    {
      "epoch": 10.79,
      "learning_rate": 5.617597292724196e-06,
      "loss": 0.0488,
      "step": 17000
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9487545613200064,
      "eval_loss": 0.2686789333820343,
      "eval_runtime": 2.2111,
      "eval_samples_per_second": 2850.648,
      "eval_steps_per_second": 178.194,
      "step": 17336
    },
    {
      "epoch": 11.1,
      "learning_rate": 5.194585448392555e-06,
      "loss": 0.054,
      "step": 17500
    },
    {
      "epoch": 11.42,
      "learning_rate": 4.771573604060914e-06,
      "loss": 0.0552,
      "step": 18000
    },
    {
      "epoch": 11.74,
      "learning_rate": 4.3485617597292725e-06,
      "loss": 0.0516,
      "step": 18500
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9497064889735046,
      "eval_loss": 0.26535487174987793,
      "eval_runtime": 2.1929,
      "eval_samples_per_second": 2874.229,
      "eval_steps_per_second": 179.668,
      "step": 18912
    },
    {
      "epoch": 12.06,
      "learning_rate": 3.925549915397631e-06,
      "loss": 0.0543,
      "step": 19000
    },
    {
      "epoch": 12.37,
      "learning_rate": 3.5025380710659903e-06,
      "loss": 0.05,
      "step": 19500
    },
    {
      "epoch": 12.69,
      "learning_rate": 3.079526226734349e-06,
      "loss": 0.0494,
      "step": 20000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9487545613200064,
      "eval_loss": 0.2722657322883606,
      "eval_runtime": 2.2256,
      "eval_samples_per_second": 2832.044,
      "eval_steps_per_second": 177.031,
      "step": 20488
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.656514382402707e-06,
      "loss": 0.0443,
      "step": 20500
    },
    {
      "epoch": 13.32,
      "learning_rate": 2.233502538071066e-06,
      "loss": 0.0355,
      "step": 21000
    },
    {
      "epoch": 13.64,
      "learning_rate": 1.810490693739425e-06,
      "loss": 0.0502,
      "step": 21500
    },
    {
      "epoch": 13.96,
      "learning_rate": 1.3874788494077834e-06,
      "loss": 0.0478,
      "step": 22000
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9500237981913374,
      "eval_loss": 0.2725750505924225,
      "eval_runtime": 2.2066,
      "eval_samples_per_second": 2856.45,
      "eval_steps_per_second": 178.556,
      "step": 22064
    },
    {
      "epoch": 14.28,
      "learning_rate": 9.644670050761422e-07,
      "loss": 0.0423,
      "step": 22500
    },
    {
      "epoch": 14.59,
      "learning_rate": 5.414551607445009e-07,
      "loss": 0.0415,
      "step": 23000
    },
    {
      "epoch": 14.91,
      "learning_rate": 1.1844331641285957e-07,
      "loss": 0.0367,
      "step": 23500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9501824528002538,
      "eval_loss": 0.27884194254875183,
      "eval_runtime": 2.2284,
      "eval_samples_per_second": 2828.43,
      "eval_steps_per_second": 176.805,
      "step": 23640
    }
  ],
  "logging_steps": 500,
  "max_steps": 23640,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 1508157302578992.0,
  "trial_name": null,
  "trial_params": null
}