run-1/checkpoint-856/trainer_state.json

{
  "best_metric": 0.3432921557472473,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-428",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 856,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5924310088157654,
      "eval_matthews_correlation": 0.06558874629318973,
      "eval_runtime": 0.7705,
      "eval_samples_per_second": 1353.67,
      "eval_steps_per_second": 85.659,
      "step": 214
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5929568409919739,
      "eval_matthews_correlation": 0.3432921557472473,
      "eval_runtime": 1.2272,
      "eval_samples_per_second": 849.906,
      "eval_steps_per_second": 53.781,
      "step": 428
    },
    {
      "epoch": 2.34,
      "grad_norm": 10.632513999938965,
      "learning_rate": 6.962872544536971e-06,
      "loss": 0.48,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.0858466625213623,
      "eval_matthews_correlation": 0.3319729838669631,
      "eval_runtime": 0.8034,
      "eval_samples_per_second": 1298.204,
      "eval_steps_per_second": 82.149,
      "step": 642
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.2085105180740356,
      "eval_matthews_correlation": 0.3372791957746894,
      "eval_runtime": 0.913,
      "eval_samples_per_second": 1142.392,
      "eval_steps_per_second": 72.289,
      "step": 856
    }
  ],
  "logging_steps": 500,
  "max_steps": 856,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 8373388457232.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.674218791607766e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 4,
    "seed": 16
  }
}