run-3/checkpoint-1605/trainer_state.json

{
  "best_metric": 0.44485820340444643,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-3\\checkpoint-1605",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1605,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.93,
      "grad_norm": 4.072146892547607,
      "learning_rate": 5.718435626528386e-05,
      "loss": 0.5498,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5121288299560547,
      "eval_matthews_correlation": 0.39294365720435087,
      "eval_runtime": 0.8037,
      "eval_samples_per_second": 1297.67,
      "eval_steps_per_second": 82.115,
      "step": 535
    },
    {
      "epoch": 1.87,
      "grad_norm": 6.578577995300293,
      "learning_rate": 3.130908193710112e-05,
      "loss": 0.3359,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6010443568229675,
      "eval_matthews_correlation": 0.4448508885597646,
      "eval_runtime": 1.1063,
      "eval_samples_per_second": 942.781,
      "eval_steps_per_second": 59.658,
      "step": 1070
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.5026248693466187,
      "learning_rate": 5.433807608918376e-06,
      "loss": 0.1888,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8050044178962708,
      "eval_matthews_correlation": 0.44485820340444643,
      "eval_runtime": 1.5026,
      "eval_samples_per_second": 694.112,
      "eval_steps_per_second": 43.923,
      "step": 1605
    }
  ],
  "logging_steps": 500,
  "max_steps": 1605,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 128783925307044.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 8.305963059346661e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 16,
    "seed": 35
  }
}