run-3/checkpoint-536/trainer_state.json

{
  "best_metric": 0.4906666052000313,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/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": 0.514577329158783,
      "eval_matthews_correlation": 0.39789056142736157,
      "eval_runtime": 0.6982,
      "eval_samples_per_second": 1493.818,
      "eval_steps_per_second": 94.527,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.49808281660079956,
      "eval_matthews_correlation": 0.43428613653509085,
      "eval_runtime": 0.754,
      "eval_samples_per_second": 1383.244,
      "eval_steps_per_second": 87.53,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5177927613258362,
      "eval_matthews_correlation": 0.4621927387299967,
      "eval_runtime": 0.7201,
      "eval_samples_per_second": 1448.493,
      "eval_steps_per_second": 91.659,
      "step": 402
    },
    {
      "epoch": 3.73,
      "grad_norm": 7.471534252166748,
      "learning_rate": 3.126905805070171e-06,
      "loss": 0.4284,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5037042498588562,
      "eval_matthews_correlation": 0.4906666052000313,
      "eval_runtime": 1.0921,
      "eval_samples_per_second": 954.999,
      "eval_steps_per_second": 60.431,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 670,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 210585131231004.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.232368758468832e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64,
    "seed": 29
  }
}