run-5/checkpoint-42/trainer_state.json

{
  "best_metric": 0.22967602990411826,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-28",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 42,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.6001009941101074,
      "eval_matthews_correlation": 0.15053351978709253,
      "eval_runtime": 1.2928,
      "eval_samples_per_second": 806.797,
      "eval_steps_per_second": 51.053,
      "step": 14
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5574995279312134,
      "eval_matthews_correlation": 0.22967602990411826,
      "eval_runtime": 1.1108,
      "eval_samples_per_second": 938.983,
      "eval_steps_per_second": 59.418,
      "step": 28
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6004344820976257,
      "eval_matthews_correlation": 0.18952742025331043,
      "eval_runtime": 0.7876,
      "eval_samples_per_second": 1324.269,
      "eval_steps_per_second": 83.798,
      "step": 42
    }
  ],
  "logging_steps": 500,
  "max_steps": 42,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.898807790233683e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 64,
    "seed": 13
  }
}