{
  "best_metric": 0.7592156862745097,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-800",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3765578269958496,
      "learning_rate": 0.0003184457865836907,
      "loss": 0.3598,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.3431719243526459,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.88,
      "eval_samples_per_second": 678.204,
      "eval_steps_per_second": 21.277,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.66363924741745,
      "learning_rate": 0.0002653714888197423,
      "loss": 0.3406,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6964705882352941,
      "eval_f1": 0.2275449101796407,
      "eval_loss": 0.3270348310470581,
      "eval_mcc": 0.22253187118870327,
      "eval_precision": 0.75,
      "eval_recall": 0.13411764705882354,
      "eval_runtime": 1.9253,
      "eval_samples_per_second": 662.218,
      "eval_steps_per_second": 20.775,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6294198036193848,
      "learning_rate": 0.00021229719105579382,
      "loss": 0.3329,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7333333333333333,
      "eval_f1": 0.4370860927152318,
      "eval_loss": 0.32712480425834656,
      "eval_mcc": 0.34642796561282824,
      "eval_precision": 0.7374301675977654,
      "eval_recall": 0.31058823529411766,
      "eval_runtime": 1.9116,
      "eval_samples_per_second": 666.968,
      "eval_steps_per_second": 20.924,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.48543694615364075,
      "learning_rate": 0.00015922289329184535,
      "loss": 0.3281,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7011764705882353,
      "eval_f1": 0.24554455445544554,
      "eval_loss": 0.3240555226802826,
      "eval_mcc": 0.2424163574855646,
      "eval_precision": 0.775,
      "eval_recall": 0.14588235294117646,
      "eval_runtime": 1.8705,
      "eval_samples_per_second": 681.63,
      "eval_steps_per_second": 21.384,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6081313490867615,
      "learning_rate": 0.00010614859552789691,
      "loss": 0.3231,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7592156862745097,
      "eval_f1": 0.5491923641703378,
      "eval_loss": 0.3244645297527313,
      "eval_mcc": 0.4222578838064514,
      "eval_precision": 0.73046875,
      "eval_recall": 0.44,
      "eval_runtime": 1.8559,
      "eval_samples_per_second": 687.004,
      "eval_steps_per_second": 21.553,
      "step": 800
    }
  ],
  "logging_steps": 500,
  "max_steps": 1120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1458777189600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.543790056219147,
    "learning_rate": 0.0003715200843476392,
    "num_train_epochs": 7,
    "temperature": 25
  }
}