{
  "best_metric": 0.7697715289982425,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-1926",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 1926,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.634866237640381,
      "learning_rate": 0.00022230485769079655,
      "loss": 0.4904,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7223198594024605,
      "eval_f1": 0.4261501210653753,
      "eval_loss": 0.456611692905426,
      "eval_mcc": 0.31390897314151045,
      "eval_precision": 0.6848249027237354,
      "eval_recall": 0.3093145869947276,
      "eval_runtime": 3.1485,
      "eval_samples_per_second": 542.168,
      "eval_steps_per_second": 17.151,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5362380743026733,
      "learning_rate": 0.000194516750479447,
      "loss": 0.4471,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7375512595196251,
      "eval_f1": 0.4628297362110312,
      "eval_loss": 0.44590938091278076,
      "eval_mcc": 0.3591774667498448,
      "eval_precision": 0.7283018867924528,
      "eval_recall": 0.3391915641476274,
      "eval_runtime": 3.1481,
      "eval_samples_per_second": 542.232,
      "eval_steps_per_second": 17.153,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.2486709356307983,
      "learning_rate": 0.00016672864326809742,
      "loss": 0.4324,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7510251903925015,
      "eval_f1": 0.4860943168077388,
      "eval_loss": 0.43371039628982544,
      "eval_mcc": 0.3989881998382762,
      "eval_precision": 0.7790697674418605,
      "eval_recall": 0.3532513181019332,
      "eval_runtime": 3.1303,
      "eval_samples_per_second": 545.315,
      "eval_steps_per_second": 17.251,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0944862365722656,
      "learning_rate": 0.00013894053605674787,
      "loss": 0.4261,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7574692442882249,
      "eval_f1": 0.5035971223021583,
      "eval_loss": 0.4297623634338379,
      "eval_mcc": 0.4175152081646285,
      "eval_precision": 0.7924528301886793,
      "eval_recall": 0.36906854130052724,
      "eval_runtime": 3.1386,
      "eval_samples_per_second": 543.865,
      "eval_steps_per_second": 17.205,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.000136852264404,
      "learning_rate": 0.00011115242884539828,
      "loss": 0.4173,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7685998828353837,
      "eval_f1": 0.5586592178770949,
      "eval_loss": 0.42163264751434326,
      "eval_mcc": 0.44683297923513404,
      "eval_precision": 0.7668711656441718,
      "eval_recall": 0.43936731107205623,
      "eval_runtime": 3.1323,
      "eval_samples_per_second": 544.972,
      "eval_steps_per_second": 17.24,
      "step": 1070
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.801867961883545,
      "learning_rate": 8.336432163404871e-05,
      "loss": 0.4119,
      "step": 1284
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7656707674282367,
      "eval_f1": 0.5893223819301848,
      "eval_loss": 0.41923508048057556,
      "eval_mcc": 0.44403462928634485,
      "eval_precision": 0.7086419753086419,
      "eval_recall": 0.5043936731107206,
      "eval_runtime": 3.1503,
      "eval_samples_per_second": 541.845,
      "eval_steps_per_second": 17.141,
      "step": 1284
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.013810157775879,
      "learning_rate": 5.557621442269914e-05,
      "loss": 0.4058,
      "step": 1498
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7685998828353837,
      "eval_f1": 0.5496009122006841,
      "eval_loss": 0.418170690536499,
      "eval_mcc": 0.4470425722955601,
      "eval_precision": 0.7824675324675324,
      "eval_recall": 0.4235500878734622,
      "eval_runtime": 3.1741,
      "eval_samples_per_second": 537.794,
      "eval_steps_per_second": 17.013,
      "step": 1498
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.8315951824188232,
      "learning_rate": 2.778810721134957e-05,
      "loss": 0.3998,
      "step": 1712
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7674282366725249,
      "eval_f1": 0.576307363927428,
      "eval_loss": 0.4167364537715912,
      "eval_mcc": 0.4452388718633528,
      "eval_precision": 0.7336956521739131,
      "eval_recall": 0.47451669595782076,
      "eval_runtime": 3.1403,
      "eval_samples_per_second": 543.573,
      "eval_steps_per_second": 17.196,
      "step": 1712
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.7433109283447266,
      "learning_rate": 0.0,
      "loss": 0.3989,
      "step": 1926
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7697715289982425,
      "eval_f1": 0.5901981230448384,
      "eval_loss": 0.4155946969985962,
      "eval_mcc": 0.4528688994135847,
      "eval_precision": 0.7256410256410256,
      "eval_recall": 0.4973637961335677,
      "eval_runtime": 3.1443,
      "eval_samples_per_second": 542.891,
      "eval_steps_per_second": 17.174,
      "step": 1926
    }
  ],
  "logging_steps": 500,
  "max_steps": 1926,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 4722980979960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7932456774257669,
    "learning_rate": 0.00025009296490214613,
    "num_train_epochs": 9,
    "temperature": 23
  }
}