{
  "best_metric": 0.508567750453949,
  "best_model_checkpoint": "bert_final_model/checkpoint-430",
  "epoch": 20.0,
  "global_step": 8600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7917374080362196,
      "eval_loss": 0.508567750453949,
      "eval_runtime": 14.6046,
      "eval_samples_per_second": 241.978,
      "eval_steps_per_second": 3.834,
      "step": 430
    },
    {
      "epoch": 1.16,
      "learning_rate": 9.418604651162791e-06,
      "loss": 0.539,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7959818902093945,
      "eval_loss": 0.5117859840393066,
      "eval_runtime": 14.7107,
      "eval_samples_per_second": 240.233,
      "eval_steps_per_second": 3.807,
      "step": 860
    },
    {
      "epoch": 2.33,
      "learning_rate": 8.837209302325582e-06,
      "loss": 0.4463,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7891907187323146,
      "eval_loss": 0.5270956158638,
      "eval_runtime": 14.86,
      "eval_samples_per_second": 237.819,
      "eval_steps_per_second": 3.769,
      "step": 1290
    },
    {
      "epoch": 3.49,
      "learning_rate": 8.255813953488374e-06,
      "loss": 0.3718,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7897566496887379,
      "eval_loss": 0.5500622987747192,
      "eval_runtime": 14.8926,
      "eval_samples_per_second": 237.3,
      "eval_steps_per_second": 3.76,
      "step": 1720
    },
    {
      "epoch": 4.65,
      "learning_rate": 7.674418604651164e-06,
      "loss": 0.3066,
      "step": 2000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7900396151669496,
      "eval_loss": 0.5998191237449646,
      "eval_runtime": 14.9204,
      "eval_samples_per_second": 236.857,
      "eval_steps_per_second": 3.753,
      "step": 2150
    },
    {
      "epoch": 5.81,
      "learning_rate": 7.0930232558139545e-06,
      "loss": 0.2395,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7843803056027164,
      "eval_loss": 0.6628095507621765,
      "eval_runtime": 14.9429,
      "eval_samples_per_second": 236.5,
      "eval_steps_per_second": 3.748,
      "step": 2580
    },
    {
      "epoch": 6.98,
      "learning_rate": 6.511627906976745e-06,
      "loss": 0.1955,
      "step": 3000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7826825127334465,
      "eval_loss": 0.723823070526123,
      "eval_runtime": 14.9266,
      "eval_samples_per_second": 236.758,
      "eval_steps_per_second": 3.752,
      "step": 3010
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7821165817770233,
      "eval_loss": 0.7725191116333008,
      "eval_runtime": 14.9366,
      "eval_samples_per_second": 236.601,
      "eval_steps_per_second": 3.749,
      "step": 3440
    },
    {
      "epoch": 8.14,
      "learning_rate": 5.930232558139536e-06,
      "loss": 0.1485,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7852292020373515,
      "eval_loss": 0.8427122831344604,
      "eval_runtime": 14.994,
      "eval_samples_per_second": 235.695,
      "eval_steps_per_second": 3.735,
      "step": 3870
    },
    {
      "epoch": 9.3,
      "learning_rate": 5.348837209302326e-06,
      "loss": 0.1196,
      "step": 4000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7818336162988115,
      "eval_loss": 0.9093418121337891,
      "eval_runtime": 15.0048,
      "eval_samples_per_second": 235.524,
      "eval_steps_per_second": 3.732,
      "step": 4300
    },
    {
      "epoch": 10.47,
      "learning_rate": 4.767441860465117e-06,
      "loss": 0.0998,
      "step": 4500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.77985285795133,
      "eval_loss": 0.9718943238258362,
      "eval_runtime": 14.9478,
      "eval_samples_per_second": 236.422,
      "eval_steps_per_second": 3.746,
      "step": 4730
    },
    {
      "epoch": 11.63,
      "learning_rate": 4.186046511627907e-06,
      "loss": 0.0838,
      "step": 5000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7809847198641766,
      "eval_loss": 0.9967551827430725,
      "eval_runtime": 15.0053,
      "eval_samples_per_second": 235.517,
      "eval_steps_per_second": 3.732,
      "step": 5160
    },
    {
      "epoch": 12.79,
      "learning_rate": 3.6046511627906977e-06,
      "loss": 0.0717,
      "step": 5500
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7829654782116582,
      "eval_loss": 1.0477949380874634,
      "eval_runtime": 14.9708,
      "eval_samples_per_second": 236.059,
      "eval_steps_per_second": 3.741,
      "step": 5590
    },
    {
      "epoch": 13.95,
      "learning_rate": 3.0232558139534885e-06,
      "loss": 0.0615,
      "step": 6000
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7818336162988115,
      "eval_loss": 1.085091471672058,
      "eval_runtime": 15.0224,
      "eval_samples_per_second": 235.249,
      "eval_steps_per_second": 3.728,
      "step": 6020
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7784380305602716,
      "eval_loss": 1.1496540307998657,
      "eval_runtime": 14.9724,
      "eval_samples_per_second": 236.034,
      "eval_steps_per_second": 3.74,
      "step": 6450
    },
    {
      "epoch": 15.12,
      "learning_rate": 2.4418604651162793e-06,
      "loss": 0.0554,
      "step": 6500
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7869269949066214,
      "eval_loss": 1.1740212440490723,
      "eval_runtime": 14.9588,
      "eval_samples_per_second": 236.25,
      "eval_steps_per_second": 3.744,
      "step": 6880
    },
    {
      "epoch": 16.28,
      "learning_rate": 1.86046511627907e-06,
      "loss": 0.0459,
      "step": 7000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.7846632710809281,
      "eval_loss": 1.2093095779418945,
      "eval_runtime": 14.779,
      "eval_samples_per_second": 239.124,
      "eval_steps_per_second": 3.789,
      "step": 7310
    },
    {
      "epoch": 17.44,
      "learning_rate": 1.2790697674418605e-06,
      "loss": 0.0421,
      "step": 7500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.787209960384833,
      "eval_loss": 1.2152066230773926,
      "eval_runtime": 14.8521,
      "eval_samples_per_second": 237.946,
      "eval_steps_per_second": 3.771,
      "step": 7740
    },
    {
      "epoch": 18.6,
      "learning_rate": 6.976744186046513e-07,
      "loss": 0.042,
      "step": 8000
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7849462365591398,
      "eval_loss": 1.2264506816864014,
      "eval_runtime": 14.8089,
      "eval_samples_per_second": 238.64,
      "eval_steps_per_second": 3.781,
      "step": 8170
    },
    {
      "epoch": 19.77,
      "learning_rate": 1.1627906976744187e-07,
      "loss": 0.0359,
      "step": 8500
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7855121675155631,
      "eval_loss": 1.2309114933013916,
      "eval_runtime": 14.8778,
      "eval_samples_per_second": 237.536,
      "eval_steps_per_second": 3.764,
      "step": 8600
    }
  ],
  "max_steps": 8600,
  "num_train_epochs": 20,
  "total_flos": 3.10690695931308e+16,
  "trial_name": null,
  "trial_params": null
}