run-1/checkpoint-1072/trainer_state.json

{
  "best_metric": 0.44997112166326025,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/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": 0.5182795524597168,
      "eval_matthews_correlation": 0.3637991719695883,
      "eval_runtime": 0.8749,
      "eval_samples_per_second": 1192.137,
      "eval_steps_per_second": 75.437,
      "step": 268
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 5.394637107849121,
      "learning_rate": 4.082371431187213e-06,
      "loss": 0.5021,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5294386148452759,
      "eval_matthews_correlation": 0.44429652763655303,
      "eval_runtime": 0.8769,
      "eval_samples_per_second": 1189.431,
      "eval_steps_per_second": 75.266,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4963017404079437,
      "eval_matthews_correlation": 0.446244823436024,
      "eval_runtime": 0.9086,
      "eval_samples_per_second": 1147.897,
      "eval_steps_per_second": 72.638,
      "step": 804
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 8.369268417358398,
      "learning_rate": 5.138649353941947e-07,
      "loss": 0.379,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5046903491020203,
      "eval_matthews_correlation": 0.44997112166326025,
      "eval_runtime": 1.1052,
      "eval_samples_per_second": 943.695,
      "eval_steps_per_second": 59.716,
      "step": 1072
    }
  ],
  "logging_steps": 500,
  "max_steps": 1072,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 191790801595908.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.650877926980232e-06,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 32,
    "seed": 28
  }
}