run-1/checkpoint-1605/trainer_state.json

{
  "best_metric": 0.4557726198308099,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-1605",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1605,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 5.798934459686279,
      "learning_rate": 3.564815491501093e-06,
      "loss": 0.5631,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.538830041885376,
      "eval_matthews_correlation": 0.3489786582473965,
      "eval_runtime": 0.3409,
      "eval_samples_per_second": 3059.333,
      "eval_steps_per_second": 193.592,
      "step": 535
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 13.630791664123535,
      "learning_rate": 1.951776807563947e-06,
      "loss": 0.4611,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5263909697532654,
      "eval_matthews_correlation": 0.40967417350821667,
      "eval_runtime": 0.3618,
      "eval_samples_per_second": 2882.412,
      "eval_steps_per_second": 182.396,
      "step": 1070
    },
    {
      "epoch": 2.803738317757009,
      "grad_norm": 11.390581130981445,
      "learning_rate": 3.3873812362680067e-07,
      "loss": 0.4175,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5153056979179382,
      "eval_matthews_correlation": 0.4557726198308099,
      "eval_runtime": 0.3914,
      "eval_samples_per_second": 2664.719,
      "eval_steps_per_second": 168.621,
      "step": 1605
    }
  ],
  "logging_steps": 500,
  "max_steps": 1605,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 128366859981588.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.177854175438239e-06,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 16,
    "seed": 31
  }
}