run-5/checkpoint-1070/trainer_state.json

{
  "best_metric": 0.0,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-535",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1070,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.93,
      "grad_norm": 2.0617318153381348,
      "learning_rate": 1.1358200826758552e-06,
      "loss": 0.6112,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5897562503814697,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.7817,
      "eval_samples_per_second": 1334.213,
      "eval_steps_per_second": 84.428,
      "step": 535
    },
    {
      "epoch": 1.87,
      "grad_norm": 4.581611633300781,
      "learning_rate": 1.3948667681984187e-07,
      "loss": 0.5628,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5660226941108704,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.8172,
      "eval_samples_per_second": 1276.381,
      "eval_steps_per_second": 80.768,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 1070,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 85847929717668.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.132153488531869e-06,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 16,
    "seed": 2
  }
}