{
  "best_metric": 0.6514851485148515,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-12/checkpoint-594",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1485,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.2921676635742188,
      "learning_rate": 0.0002222584415009539,
      "loss": 0.686,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5702970297029702,
      "eval_f1": 0.44501278772378516,
      "eval_loss": 0.6809988617897034,
      "eval_mcc": 0.15639504109088162,
      "eval_precision": 0.6258992805755396,
      "eval_recall": 0.34523809523809523,
      "eval_runtime": 0.9404,
      "eval_samples_per_second": 537.021,
      "eval_steps_per_second": 17.015,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.4260432720184326,
      "learning_rate": 0.0001666938311257154,
      "loss": 0.6423,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6514851485148515,
      "eval_f1": 0.7095709570957095,
      "eval_loss": 0.6317262053489685,
      "eval_mcc": 0.33175059993848016,
      "eval_precision": 0.6073446327683616,
      "eval_recall": 0.8531746031746031,
      "eval_runtime": 0.9481,
      "eval_samples_per_second": 532.626,
      "eval_steps_per_second": 16.875,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.5611865520477295,
      "learning_rate": 0.00011112922075047695,
      "loss": 0.6124,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6316831683168317,
      "eval_f1": 0.7056962025316456,
      "eval_loss": 0.6464924216270447,
      "eval_mcc": 0.3062819624263322,
      "eval_precision": 0.5868421052631579,
      "eval_recall": 0.8849206349206349,
      "eval_runtime": 0.9328,
      "eval_samples_per_second": 541.37,
      "eval_steps_per_second": 17.152,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.205620288848877,
      "learning_rate": 5.5564610375238476e-05,
      "loss": 0.5982,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6475247524752475,
      "eval_f1": 0.7062706270627062,
      "eval_loss": 0.6347035765647888,
      "eval_mcc": 0.3231001092600616,
      "eval_precision": 0.6045197740112994,
      "eval_recall": 0.8492063492063492,
      "eval_runtime": 0.9417,
      "eval_samples_per_second": 536.25,
      "eval_steps_per_second": 16.99,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.662528038024902,
      "learning_rate": 0.0,
      "loss": 0.5829,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6514851485148515,
      "eval_f1": 0.6912280701754385,
      "eval_loss": 0.6265904307365417,
      "eval_mcc": 0.31424156599226655,
      "eval_precision": 0.6194968553459119,
      "eval_recall": 0.7817460317460317,
      "eval_runtime": 0.9503,
      "eval_samples_per_second": 531.404,
      "eval_steps_per_second": 16.837,
      "step": 1485
    }
  ],
  "logging_steps": 500,
  "max_steps": 1485,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3653506329900.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9940046606774494,
    "learning_rate": 0.0002778230518761924,
    "num_train_epochs": 5,
    "temperature": 40
  }
}