trainer_state.json

{
  "best_metric": 0.7428781553830288,
  "best_model_checkpoint": "audio/unlearn/checkpoint/hubert-large/neggrad/2.0/superb_ks_42/checkpoint-64",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "df_accuracy": 0.9598432908912831,
      "epoch": 1.0,
      "eval_accuracy": 0.9801412180052956,
      "eval_loss": -0.12229806184768677,
      "eval_runtime": 28.8279,
      "eval_samples_per_second": 235.813,
      "eval_steps_per_second": 29.485,
      "eval_unlearn_overall_accuracy": 0.5101489635570062,
      "step": 32,
      "test_accuracy": 0.9801412180052956,
      "unlearn_overall_accuracy": 0.5101489635570062,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 2.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -6.875938892364502,
      "eval_runtime": 29.0591,
      "eval_samples_per_second": 233.937,
      "eval_steps_per_second": 29.251,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 64,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 3.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -9.186688423156738,
      "eval_runtime": 28.9567,
      "eval_samples_per_second": 234.764,
      "eval_steps_per_second": 29.354,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 96,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 4.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -11.137105941772461,
      "eval_runtime": 28.8853,
      "eval_samples_per_second": 235.345,
      "eval_steps_per_second": 29.427,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 128,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 5.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -12.868447303771973,
      "eval_runtime": 28.8975,
      "eval_samples_per_second": 235.245,
      "eval_steps_per_second": 29.414,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 160,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 6.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -14.352008819580078,
      "eval_runtime": 29.1333,
      "eval_samples_per_second": 233.341,
      "eval_steps_per_second": 29.176,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 192,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 7.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -15.559453010559082,
      "eval_runtime": 28.8044,
      "eval_samples_per_second": 236.006,
      "eval_steps_per_second": 29.509,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 224,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 8.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -16.451257705688477,
      "eval_runtime": 28.9153,
      "eval_samples_per_second": 235.101,
      "eval_steps_per_second": 29.396,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 256,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 9.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -16.99852752685547,
      "eval_runtime": 28.6371,
      "eval_samples_per_second": 237.385,
      "eval_steps_per_second": 29.682,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 288,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "df_accuracy": 0.13516160626836435,
      "epoch": 10.0,
      "eval_accuracy": 0.6209179170344219,
      "eval_loss": -17.18701934814453,
      "eval_runtime": 28.717,
      "eval_samples_per_second": 236.724,
      "eval_steps_per_second": 29.599,
      "eval_unlearn_overall_accuracy": 0.7428781553830288,
      "step": 320,
      "test_accuracy": 0.6209179170344219,
      "unlearn_overall_accuracy": 0.7428781553830288,
      "unlearn_time": -1
    },
    {
      "epoch": 10.0,
      "step": 320,
      "total_flos": 3.0944063259264e+17,
      "train_loss": -25.27679901123047,
      "train_runtime": 651.3076,
      "train_samples_per_second": 15.676,
      "train_steps_per_second": 0.491
    }
  ],
  "logging_steps": 1000,
  "max_steps": 320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 3.0944063259264e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}