{
  "best_metric": 0.8339622641509433,
  "best_model_checkpoint": "final_models/structroberta_sx2_final/finetune/sst2/checkpoint-2000",
  "epoch": 4.739336492890995,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.95,
      "eval_accuracy": 0.7992125749588013,
      "eval_f1": 0.7976190476190474,
      "eval_loss": 0.42879804968833923,
      "eval_mcc": 0.598425075943802,
      "eval_runtime": 1.0717,
      "eval_samples_per_second": 474.004,
      "eval_steps_per_second": 59.717,
      "step": 400
    },
    {
      "epoch": 1.18,
      "learning_rate": 4.407582938388626e-05,
      "loss": 0.3876,
      "step": 500
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.789370059967041,
      "eval_f1": 0.7897838899803538,
      "eval_loss": 0.4634992778301239,
      "eval_mcc": 0.5787935859950727,
      "eval_runtime": 1.069,
      "eval_samples_per_second": 475.219,
      "eval_steps_per_second": 59.87,
      "step": 800
    },
    {
      "epoch": 2.37,
      "learning_rate": 3.815165876777251e-05,
      "loss": 0.2086,
      "step": 1000
    },
    {
      "epoch": 2.84,
      "eval_accuracy": 0.8149606585502625,
      "eval_f1": 0.8,
      "eval_loss": 0.5669229626655579,
      "eval_mcc": 0.6361381051113675,
      "eval_runtime": 1.0678,
      "eval_samples_per_second": 475.734,
      "eval_steps_per_second": 59.935,
      "step": 1200
    },
    {
      "epoch": 3.55,
      "learning_rate": 3.222748815165877e-05,
      "loss": 0.1375,
      "step": 1500
    },
    {
      "epoch": 3.79,
      "eval_accuracy": 0.8090550899505615,
      "eval_f1": 0.8079207920792079,
      "eval_loss": 0.594048023223877,
      "eval_mcc": 0.6181031878719743,
      "eval_runtime": 1.0784,
      "eval_samples_per_second": 471.09,
      "eval_steps_per_second": 59.35,
      "step": 1600
    },
    {
      "epoch": 4.74,
      "learning_rate": 2.6303317535545023e-05,
      "loss": 0.091,
      "step": 2000
    },
    {
      "epoch": 4.74,
      "eval_accuracy": 0.8267716765403748,
      "eval_f1": 0.8339622641509433,
      "eval_loss": 0.6517126560211182,
      "eval_mcc": 0.6566023271980082,
      "eval_runtime": 1.0643,
      "eval_samples_per_second": 477.324,
      "eval_steps_per_second": 60.135,
      "step": 2000
    }
  ],
  "max_steps": 4220,
  "num_train_epochs": 10,
  "total_flos": 2.1943908958961664e+16,
  "trial_name": null,
  "trial_params": null
}