run-2/checkpoint-1072/trainer_state.json

{
  "best_metric": 0.24305007868019401,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-2/checkpoint-1072",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 1072,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.602029025554657,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.7285,
      "eval_samples_per_second": 1431.715,
      "eval_steps_per_second": 90.598,
      "step": 268
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 1.8423285484313965,
      "learning_rate": 9.08259470497268e-07,
      "loss": 0.605,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5747156739234924,
      "eval_matthews_correlation": 0.0463559874942472,
      "eval_runtime": 0.8181,
      "eval_samples_per_second": 1274.974,
      "eval_steps_per_second": 80.679,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.557612419128418,
      "eval_matthews_correlation": 0.1504867919191043,
      "eval_runtime": 1.1383,
      "eval_samples_per_second": 916.265,
      "eval_steps_per_second": 57.98,
      "step": 804
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 2.6135356426239014,
      "learning_rate": 3.676288332965132e-07,
      "loss": 0.5415,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5473177433013916,
      "eval_matthews_correlation": 0.24305007868019401,
      "eval_runtime": 0.7383,
      "eval_samples_per_second": 1412.795,
      "eval_steps_per_second": 89.4,
      "step": 1072
    }
  ],
  "logging_steps": 500,
  "max_steps": 1340,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 191380204405152.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.4488901076980227e-06,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 32,
    "seed": 3
  }
}