run-4/checkpoint-1605/trainer_state.json

{
  "best_metric": 0.5132274753696389,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-4\\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": 0.93,
      "grad_norm": 6.663467884063721,
      "learning_rate": 1.1599690135657977e-05,
      "loss": 0.5243,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.46048709750175476,
      "eval_matthews_correlation": 0.4791261224300467,
      "eval_runtime": 0.9073,
      "eval_samples_per_second": 1149.535,
      "eval_steps_per_second": 72.741,
      "step": 535
    },
    {
      "epoch": 1.87,
      "grad_norm": 4.0639166831970215,
      "learning_rate": 8.063199240640302e-06,
      "loss": 0.3712,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.47315478324890137,
      "eval_matthews_correlation": 0.5038487938814906,
      "eval_runtime": 1.4506,
      "eval_samples_per_second": 719.021,
      "eval_steps_per_second": 45.499,
      "step": 1070
    },
    {
      "epoch": 2.8,
      "grad_norm": 19.560606002807617,
      "learning_rate": 4.5267083456226254e-06,
      "loss": 0.2652,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5683983564376831,
      "eval_matthews_correlation": 0.5132274753696389,
      "eval_runtime": 1.4702,
      "eval_samples_per_second": 709.424,
      "eval_steps_per_second": 44.892,
      "step": 1605
    }
  ],
  "logging_steps": 500,
  "max_steps": 2140,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 128490789442440.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.5136181030675654e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 16,
    "seed": 7
  }
}