run-8/checkpoint-1100/trainer_state.json

{
  "best_metric": 0.8448979591836735,
  "best_model_checkpoint": "rubert-tiny2-finetuned-classification\\run-8\\checkpoint-1100",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1100,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.91,
      "learning_rate": 1.33127981814863e-05,
      "loss": 1.7493,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7346938775510204,
      "eval_loss": 1.10899817943573,
      "eval_runtime": 1.157,
      "eval_samples_per_second": 211.754,
      "eval_steps_per_second": 13.829,
      "step": 550
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.2187996969143835e-06,
      "loss": 1.0489,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8448979591836735,
      "eval_loss": 0.860304057598114,
      "eval_runtime": 1.1,
      "eval_samples_per_second": 222.727,
      "eval_steps_per_second": 14.545,
      "step": 1100
    }
  ],
  "logging_steps": 500,
  "max_steps": 1100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1918563271800.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.4406796666058216e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 4,
    "seed": 23
  }
}