{
  "best_metric": 0.740725657807644,
  "best_model_checkpoint": "audio/unlearn/checkpoint/hubert-large/neggrad/8.0/superb_ks_42/checkpoint-128",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 1280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 1.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -9.011107444763184,
      "eval_runtime": 27.1427,
      "eval_samples_per_second": 250.454,
      "eval_steps_per_second": 31.316,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 128,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 2.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -20.995826721191406,
      "eval_runtime": 26.9746,
      "eval_samples_per_second": 252.015,
      "eval_steps_per_second": 31.511,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 256,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 3.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -35.408653259277344,
      "eval_runtime": 27.2799,
      "eval_samples_per_second": 249.195,
      "eval_steps_per_second": 31.158,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 384,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 4.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -51.39006042480469,
      "eval_runtime": 27.3702,
      "eval_samples_per_second": 248.373,
      "eval_steps_per_second": 31.056,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 512,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 5.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -67.7049789428711,
      "eval_runtime": 27.1352,
      "eval_samples_per_second": 250.524,
      "eval_steps_per_second": 31.325,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 640,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 6.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -83.08731079101562,
      "eval_runtime": 27.1817,
      "eval_samples_per_second": 250.095,
      "eval_steps_per_second": 31.271,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 768,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 7.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -96.35177612304688,
      "eval_runtime": 27.125,
      "eval_samples_per_second": 250.617,
      "eval_steps_per_second": 31.336,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 896,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "epoch": 7.81,
      "grad_norm": 179.9547882080078,
      "learning_rate": 1.2152777777777779e-05,
      "loss": -113.6834,
      "step": 1000
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 8.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -106.53089141845703,
      "eval_runtime": 27.1699,
      "eval_samples_per_second": 250.204,
      "eval_steps_per_second": 31.285,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 1024,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 9.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -112.93724822998047,
      "eval_runtime": 27.1372,
      "eval_samples_per_second": 250.505,
      "eval_steps_per_second": 31.322,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 1152,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13946660141913383,
      "epoch": 10.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -115.13362121582031,
      "eval_runtime": 27.0193,
      "eval_samples_per_second": 251.598,
      "eval_steps_per_second": 31.459,
      "eval_unlearn_overall_accuracy": 0.740725657807644,
      "step": 1280,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.740725657807644,
      "unlearn_time": -1
    },
    {
      "epoch": 10.0,
      "step": 1280,
      "total_flos": 1.23867175847808e+18,
      "train_loss": -143.77431030273436,
      "train_runtime": 928.7475,
      "train_samples_per_second": 44.006,
      "train_steps_per_second": 1.378
    }
  ],
  "logging_steps": 1000,
  "max_steps": 1280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.23867175847808e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}