{
  "best_metric": 0.7633274751025191,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-1284",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1498,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 5.252918720245361,
      "learning_rate": 0.00015581725710675,
      "loss": 0.4793,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6994727592267135,
      "eval_f1": 0.2982216142270862,
      "eval_loss": 0.45286598801612854,
      "eval_mcc": 0.23321001518594284,
      "eval_precision": 0.6728395061728395,
      "eval_recall": 0.19156414762741653,
      "eval_runtime": 3.1418,
      "eval_samples_per_second": 543.327,
      "eval_steps_per_second": 17.188,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6013866662979126,
      "learning_rate": 0.00013850422853933333,
      "loss": 0.4377,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7393087287639133,
      "eval_f1": 0.44165621079046424,
      "eval_loss": 0.4278346300125122,
      "eval_mcc": 0.36530426808949645,
      "eval_precision": 0.7719298245614035,
      "eval_recall": 0.3093145869947276,
      "eval_runtime": 3.129,
      "eval_samples_per_second": 545.542,
      "eval_steps_per_second": 17.258,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8795703649520874,
      "learning_rate": 0.00012119119997191666,
      "loss": 0.4198,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7410661980082015,
      "eval_f1": 0.45566502463054187,
      "eval_loss": 0.4231508672237396,
      "eval_mcc": 0.36988431980614395,
      "eval_precision": 0.7613168724279835,
      "eval_recall": 0.3251318101933216,
      "eval_runtime": 3.1432,
      "eval_samples_per_second": 543.083,
      "eval_steps_per_second": 17.18,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.028773784637451,
      "learning_rate": 0.00010387817140449999,
      "loss": 0.4158,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7492677211482133,
      "eval_f1": 0.48681055155875297,
      "eval_loss": 0.41446971893310547,
      "eval_mcc": 0.3934937852291293,
      "eval_precision": 0.7660377358490567,
      "eval_recall": 0.35676625659050965,
      "eval_runtime": 3.1319,
      "eval_samples_per_second": 545.042,
      "eval_steps_per_second": 17.242,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.238177537918091,
      "learning_rate": 8.656514283708333e-05,
      "loss": 0.4093,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7545401288810779,
      "eval_f1": 0.5144843568945539,
      "eval_loss": 0.41232869029045105,
      "eval_mcc": 0.40811560014590953,
      "eval_precision": 0.7551020408163265,
      "eval_recall": 0.39015817223198596,
      "eval_runtime": 3.1304,
      "eval_samples_per_second": 545.304,
      "eval_steps_per_second": 17.25,
      "step": 1070
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.982135772705078,
      "learning_rate": 6.925211426966667e-05,
      "loss": 0.406,
      "step": 1284
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7633274751025191,
      "eval_f1": 0.5817805383022774,
      "eval_loss": 0.4112951457500458,
      "eval_mcc": 0.4373095480521575,
      "eval_precision": 0.707808564231738,
      "eval_recall": 0.4938488576449912,
      "eval_runtime": 3.1421,
      "eval_samples_per_second": 543.275,
      "eval_steps_per_second": 17.186,
      "step": 1284
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.224627494812012,
      "learning_rate": 5.1939085702249996e-05,
      "loss": 0.4012,
      "step": 1498
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.760398359695372,
      "eval_f1": 0.5193889541715628,
      "eval_loss": 0.40900105237960815,
      "eval_mcc": 0.4249892956955117,
      "eval_precision": 0.7836879432624113,
      "eval_recall": 0.3884007029876977,
      "eval_runtime": 3.1417,
      "eval_samples_per_second": 543.33,
      "eval_steps_per_second": 17.188,
      "step": 1498
    }
  ],
  "logging_steps": 500,
  "max_steps": 2140,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 3673429651080.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7600265334081553,
    "learning_rate": 0.00017313028567416666,
    "num_train_epochs": 10,
    "temperature": 20
  }
}