{
  "best_metric": 0.7129466900995899,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-1070",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1070,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.930738925933838,
      "learning_rate": 5.414014250397772e-05,
      "loss": 0.6091,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6865846514352666,
      "eval_f1": 0.37572928821470253,
      "eval_loss": 0.564587414264679,
      "eval_mcc": 0.2156910734882714,
      "eval_precision": 0.5590277777777778,
      "eval_recall": 0.28295254833040423,
      "eval_runtime": 3.1483,
      "eval_samples_per_second": 542.201,
      "eval_steps_per_second": 17.152,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.107285499572754,
      "learning_rate": 4.64058364319809e-05,
      "loss": 0.5591,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6994727592267135,
      "eval_f1": 0.5535248041775457,
      "eval_loss": 0.5493428111076355,
      "eval_mcc": 0.3271000794140795,
      "eval_precision": 0.5482758620689655,
      "eval_recall": 0.5588752196836555,
      "eval_runtime": 3.1416,
      "eval_samples_per_second": 543.358,
      "eval_steps_per_second": 17.189,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.368832588195801,
      "learning_rate": 3.867153035998409e-05,
      "loss": 0.5413,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7053309900410076,
      "eval_f1": 0.523222748815166,
      "eval_loss": 0.5401688814163208,
      "eval_mcc": 0.31393167816118617,
      "eval_precision": 0.5679012345679012,
      "eval_recall": 0.4850615114235501,
      "eval_runtime": 3.1381,
      "eval_samples_per_second": 543.954,
      "eval_steps_per_second": 17.208,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.47548770904541,
      "learning_rate": 3.093722428798727e-05,
      "loss": 0.5333,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7053309900410076,
      "eval_f1": 0.5381083562901744,
      "eval_loss": 0.5343005061149597,
      "eval_mcc": 0.32311150297426244,
      "eval_precision": 0.5634615384615385,
      "eval_recall": 0.5149384885764499,
      "eval_runtime": 3.1464,
      "eval_samples_per_second": 542.53,
      "eval_steps_per_second": 17.163,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.320547103881836,
      "learning_rate": 2.320291821599045e-05,
      "loss": 0.5244,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7129466900995899,
      "eval_f1": 0.5167652859960552,
      "eval_loss": 0.5274397730827332,
      "eval_mcc": 0.32175825520567214,
      "eval_precision": 0.5887640449438202,
      "eval_recall": 0.46045694200351495,
      "eval_runtime": 3.1507,
      "eval_samples_per_second": 541.788,
      "eval_steps_per_second": 17.139,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 1712,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 2623878322200.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9740992948842065,
    "learning_rate": 6.187444857597454e-05,
    "num_train_epochs": 8,
    "temperature": 3
  }
}