trainer_state.json

{
  "best_metric": 0.04778485745191574,
  "best_model_checkpoint": "/kaggle/working/BERT-BASE-vulgar/checkpoint-4468",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 8936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 1.8880931065353626e-05,
      "loss": 0.0943,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.7761862130707253e-05,
      "loss": 0.065,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.6642793196060877e-05,
      "loss": 0.0595,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.5523724261414504e-05,
      "loss": 0.0552,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.4404655326768131e-05,
      "loss": 0.0629,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.3285586392121757e-05,
      "loss": 0.061,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.2166517457475382e-05,
      "loss": 0.0563,
      "step": 3500
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.1047448522829008e-05,
      "loss": 0.0581,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.04778485745191574,
      "eval_roc_auc": 0.9320462909174155,
      "eval_runtime": 219.0453,
      "eval_samples_per_second": 81.59,
      "eval_steps_per_second": 5.099,
      "step": 4468
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.928379588182634e-06,
      "loss": 0.0558,
      "step": 4500
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.809310653536257e-06,
      "loss": 0.0389,
      "step": 5000
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.690241718889885e-06,
      "loss": 0.0395,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.57117278424351e-06,
      "loss": 0.044,
      "step": 6000
    },
    {
      "epoch": 1.45,
      "learning_rate": 5.452103849597136e-06,
      "loss": 0.0381,
      "step": 6500
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.3330349149507615e-06,
      "loss": 0.0396,
      "step": 7000
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.213965980304387e-06,
      "loss": 0.0349,
      "step": 7500
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.0948970456580127e-06,
      "loss": 0.0362,
      "step": 8000
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.758281110116385e-07,
      "loss": 0.0389,
      "step": 8500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.052548788487911224,
      "eval_roc_auc": 0.933822394895322,
      "eval_runtime": 218.4118,
      "eval_samples_per_second": 81.827,
      "eval_steps_per_second": 5.114,
      "step": 8936
    }
  ],
  "logging_steps": 500,
  "max_steps": 8936,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 2.592649627027338e+16,
  "trial_name": null,
  "trial_params": null
}