run-1/checkpoint-3176/trainer_state.json

{
  "best_metric": 0.6874704863843853,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-3176",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 3176,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0594383478164673,
      "learning_rate": 8.902447569795288e-05,
      "loss": 0.5725,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6804659216118369,
      "eval_f1": 0.3575949367088608,
      "eval_loss": 0.5597381591796875,
      "eval_mcc": 0.19612047304332003,
      "eval_precision": 0.5419664268585132,
      "eval_recall": 0.2668240850059032,
      "eval_runtime": 23.4624,
      "eval_samples_per_second": 541.547,
      "eval_steps_per_second": 16.963,
      "step": 1588
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.540066123008728,
      "learning_rate": 7.789641623570877e-05,
      "loss": 0.5555,
      "step": 3176
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6874704863843853,
      "eval_f1": 0.2603836841124977,
      "eval_loss": 0.5536655783653259,
      "eval_mcc": 0.18799631789309962,
      "eval_precision": 0.6164021164021164,
      "eval_recall": 0.16505312868949232,
      "eval_runtime": 23.2494,
      "eval_samples_per_second": 546.509,
      "eval_steps_per_second": 17.119,
      "step": 3176
    }
  ],
  "logging_steps": 500,
  "max_steps": 14292,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 7817788152600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9165014388792823,
    "learning_rate": 0.00010015253516019699,
    "num_train_epochs": 9,
    "temperature": 13
  }
}