trainer_state.json

{
  "best_metric": 0.6153448820114136,
  "best_model_checkpoint": "hBERTv1_new_pretrain_48_cola/checkpoint-335",
  "epoch": 10.0,
  "global_step": 670,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6296,
      "step": 67
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6342563033103943,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8565,
      "eval_samples_per_second": 561.803,
      "eval_steps_per_second": 4.848,
      "step": 67
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.6166,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6422747373580933,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8631,
      "eval_samples_per_second": 559.823,
      "eval_steps_per_second": 4.831,
      "step": 134
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.611,
      "step": 201
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6202267408370972,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8656,
      "eval_samples_per_second": 559.059,
      "eval_steps_per_second": 4.824,
      "step": 201
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6085,
      "step": 268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6188409924507141,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8563,
      "eval_samples_per_second": 561.857,
      "eval_steps_per_second": 4.848,
      "step": 268
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6107,
      "step": 335
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6153448820114136,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8588,
      "eval_samples_per_second": 561.112,
      "eval_steps_per_second": 4.842,
      "step": 335
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.5988,
      "step": 402
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6259793639183044,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8582,
      "eval_samples_per_second": 561.308,
      "eval_steps_per_second": 4.843,
      "step": 402
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.5826,
      "step": 469
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6353920102119446,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8582,
      "eval_samples_per_second": 561.287,
      "eval_steps_per_second": 4.843,
      "step": 469
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.5536,
      "step": 536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6538830399513245,
      "eval_loss": 0.6534190773963928,
      "eval_matthews_correlation": 0.030412427474982722,
      "eval_runtime": 1.8566,
      "eval_samples_per_second": 561.79,
      "eval_steps_per_second": 4.848,
      "step": 536
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 0.5402,
      "step": 603
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5953978896141052,
      "eval_loss": 0.6797628998756409,
      "eval_matthews_correlation": 0.09165495543797302,
      "eval_runtime": 1.8544,
      "eval_samples_per_second": 562.44,
      "eval_steps_per_second": 4.853,
      "step": 603
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.5341,
      "step": 670
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6836050152778625,
      "eval_loss": 0.7669716477394104,
      "eval_matthews_correlation": 0.05132011407765458,
      "eval_runtime": 1.8548,
      "eval_samples_per_second": 562.325,
      "eval_steps_per_second": 4.852,
      "step": 670
    },
    {
      "epoch": 10.0,
      "step": 670,
      "total_flos": 1.266375536410624e+16,
      "train_loss": 0.5885643973279355,
      "train_runtime": 498.4841,
      "train_samples_per_second": 857.7,
      "train_steps_per_second": 6.72
    }
  ],
  "max_steps": 3350,
  "num_train_epochs": 50,
  "total_flos": 1.266375536410624e+16,
  "trial_name": null,
  "trial_params": null
}