run-0/checkpoint-642/trainer_state.json

{
  "best_metric": 0.0,
  "best_model_checkpoint": "deberta-v3-large-finetuned-cola-midterm/run-0/checkpoint-214",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 642,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.629180371761322,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 8.504,
      "eval_samples_per_second": 122.648,
      "eval_steps_per_second": 7.761,
      "step": 214
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6256847977638245,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 8.3756,
      "eval_samples_per_second": 124.528,
      "eval_steps_per_second": 7.88,
      "step": 428
    },
    {
      "epoch": 2.34,
      "grad_norm": 4.587509632110596,
      "learning_rate": 1.75137837661781e-05,
      "loss": 0.6414,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6201117038726807,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 14.3558,
      "eval_samples_per_second": 72.653,
      "eval_steps_per_second": 4.597,
      "step": 642
    }
  ],
  "logging_steps": 500,
  "max_steps": 642,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 56862758467680.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.918203646398832e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 4,
    "seed": 10
  }
}