{
  "best_metric": 0.5596868884540117,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-1532",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 6894,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.49913594126701355,
      "learning_rate": 9.527255626859226e-05,
      "loss": 0.162,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5430528375733855,
      "eval_f1": 0.21247892074198985,
      "eval_loss": 0.1496681571006775,
      "eval_precision": 0.7682926829268293,
      "eval_recall": 0.1232876712328767,
      "eval_runtime": 132.7483,
      "eval_samples_per_second": 7.699,
      "eval_steps_per_second": 1.928,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9839492440223694,
      "learning_rate": 8.336348673501824e-05,
      "loss": 0.1535,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5596868884540117,
      "eval_f1": 0.2788461538461538,
      "eval_loss": 0.15276561677455902,
      "eval_precision": 0.7699115044247787,
      "eval_recall": 0.17025440313111545,
      "eval_runtime": 132.9511,
      "eval_samples_per_second": 7.687,
      "eval_steps_per_second": 1.926,
      "step": 1532
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.20836693048477173,
      "learning_rate": 7.145441720144419e-05,
      "loss": 0.1517,
      "step": 2298
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5322896281800391,
      "eval_f1": 0.16433566433566432,
      "eval_loss": 0.15022048354148865,
      "eval_precision": 0.7704918032786885,
      "eval_recall": 0.09197651663405088,
      "eval_runtime": 133.1996,
      "eval_samples_per_second": 7.673,
      "eval_steps_per_second": 1.922,
      "step": 2298
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.4116870164871216,
      "learning_rate": 5.9545347667870174e-05,
      "loss": 0.1508,
      "step": 3064
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.538160469667319,
      "eval_f1": 0.18620689655172412,
      "eval_loss": 0.14854486286640167,
      "eval_precision": 0.782608695652174,
      "eval_recall": 0.10567514677103718,
      "eval_runtime": 135.4399,
      "eval_samples_per_second": 7.546,
      "eval_steps_per_second": 1.89,
      "step": 3064
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8599097728729248,
      "learning_rate": 4.763627813429613e-05,
      "loss": 0.1499,
      "step": 3830
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5547945205479452,
      "eval_f1": 0.26016260162601623,
      "eval_loss": 0.14884987473487854,
      "eval_precision": 0.7692307692307693,
      "eval_recall": 0.15655577299412915,
      "eval_runtime": 134.169,
      "eval_samples_per_second": 7.617,
      "eval_steps_per_second": 1.908,
      "step": 3830
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.176220417022705,
      "learning_rate": 3.5727208600722096e-05,
      "loss": 0.1487,
      "step": 4596
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5538160469667319,
      "eval_f1": 0.25,
      "eval_loss": 0.14816069602966309,
      "eval_precision": 0.7835051546391752,
      "eval_recall": 0.1487279843444227,
      "eval_runtime": 132.3135,
      "eval_samples_per_second": 7.724,
      "eval_steps_per_second": 1.935,
      "step": 4596
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7192456126213074,
      "learning_rate": 2.3818139067148064e-05,
      "loss": 0.1481,
      "step": 5362
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5528375733855186,
      "eval_f1": 0.24212271973466004,
      "eval_loss": 0.1478748470544815,
      "eval_precision": 0.7934782608695652,
      "eval_recall": 0.14285714285714285,
      "eval_runtime": 132.0993,
      "eval_samples_per_second": 7.737,
      "eval_steps_per_second": 1.938,
      "step": 5362
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.8929200172424316,
      "learning_rate": 1.1909069533574032e-05,
      "loss": 0.1476,
      "step": 6128
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.550880626223092,
      "eval_f1": 0.23627287853577372,
      "eval_loss": 0.14707620441913605,
      "eval_precision": 0.7888888888888889,
      "eval_recall": 0.13894324853228962,
      "eval_runtime": 133.3296,
      "eval_samples_per_second": 7.665,
      "eval_steps_per_second": 1.92,
      "step": 6128
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.470459520816803,
      "learning_rate": 0.0,
      "loss": 0.1471,
      "step": 6894
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5528375733855186,
      "eval_f1": 0.24462809917355371,
      "eval_loss": 0.14724816381931305,
      "eval_precision": 0.7872340425531915,
      "eval_recall": 0.14481409001956946,
      "eval_runtime": 133.2048,
      "eval_samples_per_second": 7.672,
      "eval_steps_per_second": 1.922,
      "step": 6894
    }
  ],
  "logging_steps": 500,
  "max_steps": 6894,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 2121256775520.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.18918715587656498,
    "learning_rate": 0.0001071816258021663,
    "num_train_epochs": 9,
    "temperature": 30
  }
}