trainer_state.json

{
  "best_metric": 0.9451375626679077,
  "best_model_checkpoint": "../../checkpoint/unlearn_so/imdb/bert-base/random_label/2.0/87/checkpoint-2346",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 4692,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "df_accuracy": 0.838,
      "dt_accuracy": 0.85948,
      "epoch": 1.0,
      "eval_accuracy": 0.85948,
      "eval_loss": 0.4096975028514862,
      "eval_runtime": 37.8899,
      "eval_samples_per_second": 659.806,
      "eval_steps_per_second": 2.586,
      "eval_unlearn_overall_accuracy": 0.7880528311116546,
      "step": 782,
      "unlearn_overall_accuracy": 0.7880528311116546,
      "unlearn_time": null
    },
    {
      "epoch": 1.278772378516624,
      "grad_norm": 3.2033872604370117,
      "learning_rate": 3.934356351236147e-05,
      "loss": 0.2324,
      "step": 1000
    },
    {
      "df_accuracy": 0.728,
      "dt_accuracy": 0.8454,
      "epoch": 2.0,
      "eval_accuracy": 0.8454,
      "eval_loss": 0.44255802035331726,
      "eval_runtime": 37.8444,
      "eval_samples_per_second": 660.6,
      "eval_steps_per_second": 2.59,
      "eval_unlearn_overall_accuracy": 0.8555264653559134,
      "step": 1564,
      "unlearn_overall_accuracy": 0.8555264653559134,
      "unlearn_time": null
    },
    {
      "epoch": 2.557544757033248,
      "grad_norm": 5.137396812438965,
      "learning_rate": 2.8687127024722932e-05,
      "loss": 0.1991,
      "step": 2000
    },
    {
      "df_accuracy": 0.572,
      "dt_accuracy": 0.849,
      "epoch": 3.0,
      "eval_accuracy": 0.849,
      "eval_loss": 0.5101344585418701,
      "eval_runtime": 37.8417,
      "eval_samples_per_second": 660.646,
      "eval_steps_per_second": 2.59,
      "eval_unlearn_overall_accuracy": 0.9451375626679077,
      "step": 2346,
      "unlearn_overall_accuracy": 0.9451375626679077,
      "unlearn_time": null
    },
    {
      "epoch": 3.836317135549872,
      "grad_norm": 1.9892826080322266,
      "learning_rate": 1.80306905370844e-05,
      "loss": 0.1411,
      "step": 3000
    },
    {
      "df_accuracy": 0.408,
      "dt_accuracy": 0.83248,
      "epoch": 4.0,
      "eval_accuracy": 0.83248,
      "eval_loss": 0.5394909381866455,
      "eval_runtime": 37.8652,
      "eval_samples_per_second": 660.237,
      "eval_steps_per_second": 2.588,
      "eval_unlearn_overall_accuracy": 0.9268024158757548,
      "step": 3128,
      "unlearn_overall_accuracy": 0.9268024158757548,
      "unlearn_time": null
    },
    {
      "df_accuracy": 0.25,
      "dt_accuracy": 0.83,
      "epoch": 5.0,
      "eval_accuracy": 0.83,
      "eval_loss": 0.6352426409721375,
      "eval_runtime": 37.3455,
      "eval_samples_per_second": 669.424,
      "eval_steps_per_second": 2.624,
      "eval_unlearn_overall_accuracy": 0,
      "step": 3910,
      "unlearn_overall_accuracy": 0,
      "unlearn_time": null
    },
    {
      "epoch": 5.115089514066496,
      "grad_norm": 0.045985039323568344,
      "learning_rate": 7.374254049445865e-06,
      "loss": 0.0815,
      "step": 4000
    },
    {
      "df_accuracy": 0.162,
      "dt_accuracy": 0.82516,
      "epoch": 6.0,
      "eval_accuracy": 0.82516,
      "eval_loss": 0.8689067363739014,
      "eval_runtime": 37.3192,
      "eval_samples_per_second": 669.896,
      "eval_steps_per_second": 2.626,
      "eval_unlearn_overall_accuracy": 0,
      "step": 4692,
      "unlearn_overall_accuracy": 0,
      "unlearn_time": null
    },
    {
      "epoch": 6.0,
      "step": 4692,
      "total_flos": 9866664576000000.0,
      "train_loss": 0.1455642276611068,
      "train_runtime": 1145.4271,
      "train_samples_per_second": 130.956,
      "train_steps_per_second": 4.096
    }
  ],
  "logging_steps": 1000,
  "max_steps": 4692,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9866664576000000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}