{
  "best_metric": 0.7188118811881188,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-23/checkpoint-1485",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2079,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.666294813156128,
      "learning_rate": 0.00034226276676608983,
      "loss": 0.688,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5900990099009901,
      "eval_f1": 0.5389755011135857,
      "eval_loss": 0.6901505589485168,
      "eval_mcc": 0.18426952240167496,
      "eval_precision": 0.6142131979695431,
      "eval_recall": 0.4801587301587302,
      "eval_runtime": 0.9313,
      "eval_samples_per_second": 542.229,
      "eval_steps_per_second": 17.18,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4357942342758179,
      "learning_rate": 0.00029336808579950554,
      "loss": 0.6503,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6237623762376238,
      "eval_f1": 0.6895424836601307,
      "eval_loss": 0.641117513179779,
      "eval_mcc": 0.27448731892832257,
      "eval_precision": 0.5861111111111111,
      "eval_recall": 0.8373015873015873,
      "eval_runtime": 0.9311,
      "eval_samples_per_second": 542.381,
      "eval_steps_per_second": 17.184,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.663438081741333,
      "learning_rate": 0.0002444734048329213,
      "loss": 0.6219,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6455445544554456,
      "eval_f1": 0.7098865478119936,
      "eval_loss": 0.6294957995414734,
      "eval_mcc": 0.32613098301912463,
      "eval_precision": 0.6,
      "eval_recall": 0.8690476190476191,
      "eval_runtime": 0.9408,
      "eval_samples_per_second": 536.754,
      "eval_steps_per_second": 17.006,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.565063238143921,
      "learning_rate": 0.00019557872386633703,
      "loss": 0.5927,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.700990099009901,
      "eval_f1": 0.7392055267702936,
      "eval_loss": 0.59776771068573,
      "eval_mcc": 0.42132165499205354,
      "eval_precision": 0.654434250764526,
      "eval_recall": 0.8492063492063492,
      "eval_runtime": 0.9309,
      "eval_samples_per_second": 542.515,
      "eval_steps_per_second": 17.189,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.263581275939941,
      "learning_rate": 0.00014668404289975277,
      "loss": 0.5661,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7188118811881188,
      "eval_f1": 0.7455197132616487,
      "eval_loss": 0.560869574546814,
      "eval_mcc": 0.44822084848200644,
      "eval_precision": 0.6797385620915033,
      "eval_recall": 0.8253968253968254,
      "eval_runtime": 0.9261,
      "eval_samples_per_second": 545.289,
      "eval_steps_per_second": 17.276,
      "step": 1485
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.6163554191589355,
      "learning_rate": 9.778936193316852e-05,
      "loss": 0.5531,
      "step": 1782
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7128712871287128,
      "eval_f1": 0.7415329768270944,
      "eval_loss": 0.5461980104446411,
      "eval_mcc": 0.4372741452666515,
      "eval_precision": 0.6731391585760518,
      "eval_recall": 0.8253968253968254,
      "eval_runtime": 0.9305,
      "eval_samples_per_second": 542.701,
      "eval_steps_per_second": 17.194,
      "step": 1782
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.3622937202453613,
      "learning_rate": 4.889468096658426e-05,
      "loss": 0.5399,
      "step": 2079
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7148514851485148,
      "eval_f1": 0.7464788732394366,
      "eval_loss": 0.5450253486633301,
      "eval_mcc": 0.44448710523745394,
      "eval_precision": 0.6708860759493671,
      "eval_recall": 0.8412698412698413,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 539.724,
      "eval_steps_per_second": 17.1,
      "step": 2079
    }
  ],
  "logging_steps": 500,
  "max_steps": 2376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 5114908861860.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9955244977733414,
    "learning_rate": 0.00039115744773267407,
    "num_train_epochs": 8,
    "temperature": 44
  }
}