run-0/checkpoint-20/trainer_state.json

{
  "best_metric": 0.5992779783393501,
  "best_model_checkpoint": "microsoft/deberta-v3-smallsed_rte-finetuned-rte/run-0/checkpoint-12",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.6915086507797241,
      "eval_runtime": 1.9301,
      "eval_samples_per_second": 143.52,
      "eval_steps_per_second": 9.326,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.6917890906333923,
      "eval_runtime": 1.768,
      "eval_samples_per_second": 156.678,
      "eval_steps_per_second": 10.181,
      "step": 8
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5992779783393501,
      "eval_loss": 0.692664384841919,
      "eval_runtime": 1.7537,
      "eval_samples_per_second": 157.948,
      "eval_steps_per_second": 10.264,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 0.6929299831390381,
      "eval_runtime": 1.7438,
      "eval_samples_per_second": 158.846,
      "eval_steps_per_second": 10.322,
      "step": 16
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.48014440433212996,
      "eval_loss": 0.6928216218948364,
      "eval_runtime": 2.4069,
      "eval_samples_per_second": 115.088,
      "eval_steps_per_second": 7.479,
      "step": 20
    }
  ],
  "logging_steps": 500,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.0951184212103965e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64,
    "seed": 39
  }
}