trainer_state.json

{
  "best_metric": 0.9066210706300144,
  "best_model_checkpoint": "../../checkpoint/unlearn/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": 0.908,
      "dt_accuracy": 0.87476,
      "epoch": 1.0,
      "eval_accuracy": 0.87476,
      "eval_loss": 0.45199230313301086,
      "eval_runtime": 39.0075,
      "eval_samples_per_second": 640.902,
      "eval_steps_per_second": 2.512,
      "eval_unlearn_overall_accuracy": 0.7405655287597742,
      "step": 782,
      "unlearn_overall_accuracy": 0.7405655287597742,
      "unlearn_time": null
    },
    {
      "epoch": 1.278772378516624,
      "grad_norm": 0.889214038848877,
      "learning_rate": 3.934356351236147e-05,
      "loss": 0.1206,
      "step": 1000
    },
    {
      "df_accuracy": 0.752,
      "dt_accuracy": 0.87788,
      "epoch": 2.0,
      "eval_accuracy": 0.87788,
      "eval_loss": 0.43953636288642883,
      "eval_runtime": 38.1024,
      "eval_samples_per_second": 656.126,
      "eval_steps_per_second": 2.572,
      "eval_unlearn_overall_accuracy": 0.8528180250603518,
      "step": 1564,
      "unlearn_overall_accuracy": 0.8528180250603518,
      "unlearn_time": null
    },
    {
      "epoch": 2.557544757033248,
      "grad_norm": 0.24288327991962433,
      "learning_rate": 2.8687127024722932e-05,
      "loss": 0.1011,
      "step": 2000
    },
    {
      "df_accuracy": 0.66,
      "dt_accuracy": 0.8708,
      "epoch": 3.0,
      "eval_accuracy": 0.8708,
      "eval_loss": 0.5831520557403564,
      "eval_runtime": 37.4762,
      "eval_samples_per_second": 667.09,
      "eval_steps_per_second": 2.615,
      "eval_unlearn_overall_accuracy": 0.9066210706300144,
      "step": 2346,
      "unlearn_overall_accuracy": 0.9066210706300144,
      "unlearn_time": null
    },
    {
      "epoch": 3.836317135549872,
      "grad_norm": 0.07790789008140564,
      "learning_rate": 1.80306905370844e-05,
      "loss": 0.068,
      "step": 3000
    },
    {
      "df_accuracy": 0.234,
      "dt_accuracy": 0.87012,
      "epoch": 4.0,
      "eval_accuracy": 0.87012,
      "eval_loss": 0.7269598245620728,
      "eval_runtime": 37.4269,
      "eval_samples_per_second": 667.968,
      "eval_steps_per_second": 2.618,
      "eval_unlearn_overall_accuracy": 0,
      "step": 3128,
      "unlearn_overall_accuracy": 0,
      "unlearn_time": null
    },
    {
      "df_accuracy": 0.018,
      "dt_accuracy": 0.86948,
      "epoch": 5.0,
      "eval_accuracy": 0.86948,
      "eval_loss": 0.8713451027870178,
      "eval_runtime": 38.2922,
      "eval_samples_per_second": 652.875,
      "eval_steps_per_second": 2.559,
      "eval_unlearn_overall_accuracy": 0,
      "step": 3910,
      "unlearn_overall_accuracy": 0,
      "unlearn_time": null
    },
    {
      "epoch": 5.115089514066496,
      "grad_norm": 0.034020449966192245,
      "learning_rate": 7.374254049445865e-06,
      "loss": 0.026,
      "step": 4000
    },
    {
      "df_accuracy": 0.01,
      "dt_accuracy": 0.87316,
      "epoch": 6.0,
      "eval_accuracy": 0.87316,
      "eval_loss": 0.9447321891784668,
      "eval_runtime": 37.3517,
      "eval_samples_per_second": 669.314,
      "eval_steps_per_second": 2.624,
      "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": 0.06846464147754744,
      "train_runtime": 1043.7186,
      "train_samples_per_second": 143.717,
      "train_steps_per_second": 4.495
    }
  ],
  "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
}