{
  "best_metric": 0.6831683168316832,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-10/checkpoint-1782",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1782,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.256939649581909,
      "learning_rate": 0.00020003476406791257,
      "loss": 0.6792,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5623762376237624,
      "eval_f1": 0.4199475065616798,
      "eval_loss": 0.6835970878601074,
      "eval_mcc": 0.14191809339823447,
      "eval_precision": 0.6201550387596899,
      "eval_recall": 0.31746031746031744,
      "eval_runtime": 0.9302,
      "eval_samples_per_second": 542.921,
      "eval_steps_per_second": 17.201,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.075068712234497,
      "learning_rate": 0.00016002781125433002,
      "loss": 0.6413,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6475247524752475,
      "eval_f1": 0.6920415224913495,
      "eval_loss": 0.6267873644828796,
      "eval_mcc": 0.3090077890032306,
      "eval_precision": 0.6134969325153374,
      "eval_recall": 0.7936507936507936,
      "eval_runtime": 0.9329,
      "eval_samples_per_second": 541.318,
      "eval_steps_per_second": 17.151,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.5431761741638184,
      "learning_rate": 0.00012002085844074753,
      "loss": 0.6128,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6376237623762376,
      "eval_f1": 0.7127158555729984,
      "eval_loss": 0.6455628275871277,
      "eval_mcc": 0.3245645841843151,
      "eval_precision": 0.5896103896103896,
      "eval_recall": 0.9007936507936508,
      "eval_runtime": 0.9321,
      "eval_samples_per_second": 541.772,
      "eval_steps_per_second": 17.165,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.826803207397461,
      "learning_rate": 8.001390562716501e-05,
      "loss": 0.5934,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6554455445544555,
      "eval_f1": 0.7156862745098039,
      "eval_loss": 0.6337190866470337,
      "eval_mcc": 0.3445175537543676,
      "eval_precision": 0.6083333333333333,
      "eval_recall": 0.8690476190476191,
      "eval_runtime": 0.9303,
      "eval_samples_per_second": 542.814,
      "eval_steps_per_second": 17.198,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.4030377864837646,
      "learning_rate": 4.0006952813582505e-05,
      "loss": 0.5801,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6792079207920793,
      "eval_f1": 0.7054545454545453,
      "eval_loss": 0.6095411777496338,
      "eval_mcc": 0.3647441045625896,
      "eval_precision": 0.6510067114093959,
      "eval_recall": 0.7698412698412699,
      "eval_runtime": 0.9333,
      "eval_samples_per_second": 541.076,
      "eval_steps_per_second": 17.143,
      "step": 1485
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.6899707317352295,
      "learning_rate": 0.0,
      "loss": 0.5709,
      "step": 1782
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6831683168316832,
      "eval_f1": 0.7090909090909091,
      "eval_loss": 0.6130138039588928,
      "eval_mcc": 0.3727967308109112,
      "eval_precision": 0.6543624161073825,
      "eval_recall": 0.7738095238095238,
      "eval_runtime": 0.9378,
      "eval_samples_per_second": 538.498,
      "eval_steps_per_second": 17.061,
      "step": 1782
    }
  ],
  "logging_steps": 500,
  "max_steps": 1782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 4384207595880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9802546625572438,
    "learning_rate": 0.00024004171688149506,
    "num_train_epochs": 6,
    "temperature": 49
  }
}