run-22/checkpoint-1100/trainer_state.json

{
  "best_metric": 0.9020408163265307,
  "best_model_checkpoint": "rubert-tiny2-finetuned-classification\\run-22\\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": 4.3997824385548506e-05,
      "loss": 1.247,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8775510204081632,
      "eval_loss": 0.5126648545265198,
      "eval_runtime": 1.224,
      "eval_samples_per_second": 200.163,
      "eval_steps_per_second": 13.072,
      "step": 550
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.1057287801563645e-05,
      "loss": 0.4548,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9020408163265307,
      "eval_loss": 0.37136828899383545,
      "eval_runtime": 2.899,
      "eval_samples_per_second": 84.512,
      "eval_steps_per_second": 5.519,
      "step": 1100
    }
  ],
  "logging_steps": 500,
  "max_steps": 2200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1932233100000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.693836096953336e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 4,
    "seed": 32
  }
}