{
  "best_metric": 0.8209393346379648,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-12/checkpoint-768",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1344,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9969735741615295,
      "learning_rate": 0.0003519988357465585,
      "loss": 0.4838,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7915851272015656,
      "eval_f1": 0.8149435273675064,
      "eval_loss": 0.4351368844509125,
      "eval_mcc": 0.6026908465743466,
      "eval_precision": 0.7328125,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 66.7179,
      "eval_samples_per_second": 15.318,
      "eval_steps_per_second": 0.48,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.071135520935059,
      "learning_rate": 0.0002933323631221321,
      "loss": 0.435,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7954990215264188,
      "eval_f1": 0.7956989247311829,
      "eval_loss": 0.42000794410705566,
      "eval_mcc": 0.5909991747116098,
      "eval_precision": 0.794921875,
      "eval_recall": 0.7964774951076321,
      "eval_runtime": 66.4436,
      "eval_samples_per_second": 15.381,
      "eval_steps_per_second": 0.482,
      "step": 384
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.5558242797851562,
      "learning_rate": 0.00023466589049770566,
      "loss": 0.4107,
      "step": 576
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8101761252446184,
      "eval_f1": 0.8255395683453237,
      "eval_loss": 0.4215203523635864,
      "eval_mcc": 0.6302037430382941,
      "eval_precision": 0.7637271214642263,
      "eval_recall": 0.898238747553816,
      "eval_runtime": 66.249,
      "eval_samples_per_second": 15.427,
      "eval_steps_per_second": 0.483,
      "step": 576
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.956291437149048,
      "learning_rate": 0.00017599941787327925,
      "loss": 0.4095,
      "step": 768
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8209393346379648,
      "eval_f1": 0.8271954674220963,
      "eval_loss": 0.4081785976886749,
      "eval_mcc": 0.6435679288765915,
      "eval_precision": 0.7992700729927007,
      "eval_recall": 0.8571428571428571,
      "eval_runtime": 66.2882,
      "eval_samples_per_second": 15.418,
      "eval_steps_per_second": 0.483,
      "step": 768
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.7001824378967285,
      "learning_rate": 0.00011733294524885283,
      "loss": 0.3959,
      "step": 960
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8170254403131115,
      "eval_f1": 0.826046511627907,
      "eval_loss": 0.41252297163009644,
      "eval_mcc": 0.6374890350533482,
      "eval_precision": 0.7872340425531915,
      "eval_recall": 0.8688845401174168,
      "eval_runtime": 66.4475,
      "eval_samples_per_second": 15.381,
      "eval_steps_per_second": 0.482,
      "step": 960
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.9468530416488647,
      "learning_rate": 5.8666472624426415e-05,
      "loss": 0.3889,
      "step": 1152
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8160469667318982,
      "eval_f1": 0.8324420677361855,
      "eval_loss": 0.40997278690338135,
      "eval_mcc": 0.6445565206113043,
      "eval_precision": 0.7643207855973814,
      "eval_recall": 0.913894324853229,
      "eval_runtime": 66.9893,
      "eval_samples_per_second": 15.256,
      "eval_steps_per_second": 0.478,
      "step": 1152
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.8196734189987183,
      "learning_rate": 0.0,
      "loss": 0.3804,
      "step": 1344
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8209393346379648,
      "eval_f1": 0.8297674418604651,
      "eval_loss": 0.405351847410202,
      "eval_mcc": 0.6453592700540068,
      "eval_precision": 0.7907801418439716,
      "eval_recall": 0.87279843444227,
      "eval_runtime": 66.7109,
      "eval_samples_per_second": 15.32,
      "eval_steps_per_second": 0.48,
      "step": 1344
    }
  ],
  "logging_steps": 500,
  "max_steps": 1344,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1649866380960.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7654129557516562,
    "learning_rate": 0.00041066530837098494,
    "num_train_epochs": 7,
    "per_device_train_batch_size": 16,
    "temperature": 40
  }
}