run-8/checkpoint-804/trainer_state.json

{
  "best_metric": 0.5179258445384731,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-8/checkpoint-804",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 804,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.4677087068557739,
      "eval_matthews_correlation": 0.47972131189227946,
      "eval_runtime": 0.7433,
      "eval_samples_per_second": 1403.203,
      "eval_steps_per_second": 88.793,
      "step": 268
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.3516971949951169e-05,
      "loss": 0.4212,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4737430512905121,
      "eval_matthews_correlation": 0.49867676415016876,
      "eval_runtime": 1.3743,
      "eval_samples_per_second": 758.917,
      "eval_steps_per_second": 48.023,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5736908316612244,
      "eval_matthews_correlation": 0.5179258445384731,
      "eval_runtime": 0.8113,
      "eval_samples_per_second": 1285.54,
      "eval_steps_per_second": 81.348,
      "step": 804
    }
  ],
  "logging_steps": 500,
  "max_steps": 804,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 95049498141888.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3.574883370973928e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 32,
    "seed": 15
  }
}