{
  "best_metric": 0.686368644734771,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-7940",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 7940,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0009326934814453,
      "learning_rate": 3.0645727808436554e-05,
      "loss": 0.5665,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6762159609633244,
      "eval_f1": 0.23045267489711932,
      "eval_loss": 0.5539034605026245,
      "eval_mcc": 0.1452179000925905,
      "eval_precision": 0.5544554455445545,
      "eval_recall": 0.14545454545454545,
      "eval_runtime": 24.2515,
      "eval_samples_per_second": 523.927,
      "eval_steps_per_second": 16.411,
      "step": 1588
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.73245108127594,
      "learning_rate": 2.4516582246749238e-05,
      "loss": 0.5518,
      "step": 3176
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6808594364866992,
      "eval_f1": 0.21338506304558677,
      "eval_loss": 0.5477257370948792,
      "eval_mcc": 0.15677700026841318,
      "eval_precision": 0.5978260869565217,
      "eval_recall": 0.12987012987012986,
      "eval_runtime": 23.7421,
      "eval_samples_per_second": 535.167,
      "eval_steps_per_second": 16.763,
      "step": 3176
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6722546815872192,
      "learning_rate": 1.838743668506193e-05,
      "loss": 0.5457,
      "step": 4764
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6843223673854871,
      "eval_f1": 0.27794779477947795,
      "eval_loss": 0.5387066602706909,
      "eval_mcc": 0.18168774215500283,
      "eval_precision": 0.5848484848484848,
      "eval_recall": 0.18229043683589138,
      "eval_runtime": 23.3085,
      "eval_samples_per_second": 545.123,
      "eval_steps_per_second": 17.075,
      "step": 4764
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.5396648645401,
      "learning_rate": 1.2258291123374619e-05,
      "loss": 0.542,
      "step": 6352
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6862899417597985,
      "eval_f1": 0.33875248838752486,
      "eval_loss": 0.5369384288787842,
      "eval_mcc": 0.20304079258119898,
      "eval_precision": 0.5694366982710541,
      "eval_recall": 0.241086186540732,
      "eval_runtime": 23.3813,
      "eval_samples_per_second": 543.425,
      "eval_steps_per_second": 17.022,
      "step": 6352
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.607357978820801,
      "learning_rate": 6.1291455616873095e-06,
      "loss": 0.5402,
      "step": 7940
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.686368644734771,
      "eval_f1": 0.29431556578714363,
      "eval_loss": 0.5342038869857788,
      "eval_mcc": 0.1914355744675391,
      "eval_precision": 0.5885269121813032,
      "eval_recall": 0.1962219598583235,
      "eval_runtime": 23.3468,
      "eval_samples_per_second": 544.229,
      "eval_steps_per_second": 17.047,
      "step": 7940
    }
  ],
  "logging_steps": 500,
  "max_steps": 9528,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 19544470381500.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8938428030285009,
    "learning_rate": 3.677487337012386e-05,
    "num_train_epochs": 6,
    "temperature": 20
  }
}