run-1/checkpoint-594/trainer_state.json

{
  "best_metric": 0.5425742574257426,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-594",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 594,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9758312702178955,
      "learning_rate": 2.4789676674761582e-05,
      "loss": 0.4903,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.502970297029703,
      "eval_f1": 0.08058608058608059,
      "eval_loss": 0.48342981934547424,
      "eval_mcc": 0.010331488326433175,
      "eval_precision": 0.5238095238095238,
      "eval_recall": 0.04365079365079365,
      "eval_runtime": 0.9315,
      "eval_samples_per_second": 542.124,
      "eval_steps_per_second": 17.176,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7211948037147522,
      "learning_rate": 1.2394838337380791e-05,
      "loss": 0.4812,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5425742574257426,
      "eval_f1": 0.3773584905660377,
      "eval_loss": 0.4795511066913605,
      "eval_mcc": 0.09908299956847767,
      "eval_precision": 0.5882352941176471,
      "eval_recall": 0.2777777777777778,
      "eval_runtime": 0.9303,
      "eval_samples_per_second": 542.812,
      "eval_steps_per_second": 17.198,
      "step": 594
    }
  ],
  "logging_steps": 500,
  "max_steps": 891,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1461402531960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6389006598546823,
    "learning_rate": 3.718451501214238e-05,
    "num_train_epochs": 3,
    "temperature": 17
  }
}