{
  "best_metric": 0.8356164383561644,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-4596",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 4596,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.943489909172058,
      "learning_rate": 9.076816597122249e-05,
      "loss": 0.4823,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7896281800391389,
      "eval_f1": 0.8125544899738447,
      "eval_loss": 0.43095502257347107,
      "eval_precision": 0.7327044025157232,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 132.0712,
      "eval_samples_per_second": 7.738,
      "eval_steps_per_second": 1.938,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.456132888793945,
      "learning_rate": 7.261453277697798e-05,
      "loss": 0.4312,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.824853228962818,
      "eval_f1": 0.8347183748845799,
      "eval_loss": 0.40763548016548157,
      "eval_precision": 0.7902097902097902,
      "eval_recall": 0.8845401174168297,
      "eval_runtime": 132.1546,
      "eval_samples_per_second": 7.733,
      "eval_steps_per_second": 1.937,
      "step": 1532
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.0639514923095703,
      "learning_rate": 5.446089958273349e-05,
      "loss": 0.4141,
      "step": 2298
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8189823874755382,
      "eval_f1": 0.821944177093359,
      "eval_loss": 0.43046870827674866,
      "eval_precision": 0.8087121212121212,
      "eval_recall": 0.8356164383561644,
      "eval_runtime": 130.1314,
      "eval_samples_per_second": 7.854,
      "eval_steps_per_second": 1.967,
      "step": 2298
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.4112496376037598,
      "learning_rate": 3.630726638848899e-05,
      "loss": 0.4066,
      "step": 3064
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8317025440313112,
      "eval_f1": 0.8374291115311909,
      "eval_loss": 0.4001712203025818,
      "eval_precision": 0.8098720292504571,
      "eval_recall": 0.8669275929549902,
      "eval_runtime": 132.1937,
      "eval_samples_per_second": 7.731,
      "eval_steps_per_second": 1.937,
      "step": 3064
    },
    {
      "epoch": 5.0,
      "grad_norm": 28.75933265686035,
      "learning_rate": 1.8153633194244494e-05,
      "loss": 0.3942,
      "step": 3830
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.824853228962818,
      "eval_f1": 0.8334883720930233,
      "eval_loss": 0.4072827398777008,
      "eval_precision": 0.7943262411347518,
      "eval_recall": 0.8767123287671232,
      "eval_runtime": 133.1862,
      "eval_samples_per_second": 7.673,
      "eval_steps_per_second": 1.922,
      "step": 3830
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.708053588867188,
      "learning_rate": 0.0,
      "loss": 0.3933,
      "step": 4596
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8356164383561644,
      "eval_f1": 0.8444444444444444,
      "eval_loss": 0.40063977241516113,
      "eval_precision": 0.8014059753954306,
      "eval_recall": 0.8923679060665362,
      "eval_runtime": 131.5339,
      "eval_samples_per_second": 7.77,
      "eval_steps_per_second": 1.946,
      "step": 4596
    }
  ],
  "logging_steps": 500,
  "max_steps": 4596,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1414171183680.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7464330126775673,
    "learning_rate": 0.00010892179916546698,
    "num_train_epochs": 6,
    "temperature": 11
  }
}