run-2/checkpoint-252/trainer_state.json

{
  "best_metric": 0.5667870036101083,
  "best_model_checkpoint": "deberta-v3-base-finetuned-rte/run-2/checkpoint-252",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 252,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.6936208009719849,
      "eval_runtime": 7.9894,
      "eval_samples_per_second": 34.671,
      "eval_steps_per_second": 4.381,
      "step": 63
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.6934747695922852,
      "eval_runtime": 6.6241,
      "eval_samples_per_second": 41.817,
      "eval_steps_per_second": 5.284,
      "step": 126
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.6927840709686279,
      "eval_runtime": 6.4364,
      "eval_samples_per_second": 43.036,
      "eval_steps_per_second": 5.438,
      "step": 189
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5667870036101083,
      "eval_loss": 0.6924492716789246,
      "eval_runtime": 5.8031,
      "eval_samples_per_second": 47.733,
      "eval_steps_per_second": 6.031,
      "step": 252
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 4.16979144320955e-06,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 4,
    "seed": 12
  }
}