run-27/checkpoint-1100/trainer_state.json

{
  "best_metric": 0.9020408163265307,
  "best_model_checkpoint": "rubert-tiny2-finetuned-classification\\run-27\\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": 4.106777464711046e-05,
      "loss": 1.2397,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8775510204081632,
      "eval_loss": 0.5188024640083313,
      "eval_runtime": 1.334,
      "eval_samples_per_second": 183.658,
      "eval_steps_per_second": 11.994,
      "step": 550
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.3212220452714603e-05,
      "loss": 0.4622,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9020408163265307,
      "eval_loss": 0.38360315561294556,
      "eval_runtime": 1.182,
      "eval_samples_per_second": 207.276,
      "eval_steps_per_second": 13.536,
      "step": 1100
    }
  ],
  "logging_steps": 500,
  "max_steps": 1650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2022384751800.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.89233288415063e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 4,
    "seed": 32
  }
}