{
  "best_metric": 0.8999612252811167,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-34/checkpoint-2580",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 2580,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.7244538068771362,
      "learning_rate": 5.056379477363956e-05,
      "loss": 0.3213,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8898797983714618,
      "eval_f1": 0.003508771929824561,
      "eval_loss": 0.2966959476470947,
      "eval_mcc": 0.039546160983878816,
      "eval_precision": 1.0,
      "eval_recall": 0.0017574692442882249,
      "eval_runtime": 9.9281,
      "eval_samples_per_second": 519.536,
      "eval_steps_per_second": 16.317,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9419800043106079,
      "learning_rate": 4.213649564469964e-05,
      "loss": 0.2912,
      "step": 1290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8947266382318728,
      "eval_f1": 0.15552099533437014,
      "eval_loss": 0.29070472717285156,
      "eval_mcc": 0.21772363297082026,
      "eval_precision": 0.6756756756756757,
      "eval_recall": 0.08787346221441125,
      "eval_runtime": 9.9164,
      "eval_samples_per_second": 520.151,
      "eval_steps_per_second": 16.337,
      "step": 1290
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8201465606689453,
      "learning_rate": 3.3709196515759705e-05,
      "loss": 0.2839,
      "step": 1935
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8980224893369523,
      "eval_f1": 0.21492537313432836,
      "eval_loss": 0.27995234727859497,
      "eval_mcc": 0.2718182891460322,
      "eval_precision": 0.7128712871287128,
      "eval_recall": 0.1265377855887522,
      "eval_runtime": 9.9148,
      "eval_samples_per_second": 520.234,
      "eval_steps_per_second": 16.339,
      "step": 1935
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.445148468017578,
      "learning_rate": 2.528189738681978e-05,
      "loss": 0.2815,
      "step": 2580
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8999612252811167,
      "eval_f1": 0.33333333333333337,
      "eval_loss": 0.2778097689151764,
      "eval_mcc": 0.33700627546745177,
      "eval_precision": 0.6292682926829268,
      "eval_recall": 0.22671353251318102,
      "eval_runtime": 9.9331,
      "eval_samples_per_second": 519.275,
      "eval_steps_per_second": 16.309,
      "step": 2580
    }
  ],
  "logging_steps": 500,
  "max_steps": 4515,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 6347770130400.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8969386841514778,
    "learning_rate": 5.899109390257949e-05,
    "num_train_epochs": 7,
    "temperature": 28
  }
}