run-0/checkpoint-536/trainer_state.json

{
  "best_metric": 0.19816060019544374,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-536",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 536,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.6058756113052368,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.1108,
      "eval_samples_per_second": 938.924,
      "eval_steps_per_second": 59.414,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5703422427177429,
      "eval_matthews_correlation": 0.0463559874942472,
      "eval_runtime": 1.1844,
      "eval_samples_per_second": 880.626,
      "eval_steps_per_second": 55.725,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5537189245223999,
      "eval_matthews_correlation": 0.13847035035325458,
      "eval_runtime": 1.2327,
      "eval_samples_per_second": 846.092,
      "eval_steps_per_second": 53.54,
      "step": 402
    },
    {
      "epoch": 3.73,
      "grad_norm": 2.2481775283813477,
      "learning_rate": 1.4813595850914947e-07,
      "loss": 0.5692,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5494163632392883,
      "eval_matthews_correlation": 0.19816060019544374,
      "eval_runtime": 0.7154,
      "eval_samples_per_second": 1457.894,
      "eval_steps_per_second": 92.254,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 536,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 209831981626536.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.205579826691781e-06,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 64,
    "seed": 12
  }
}