{
  "best_metric": 0.8656126482213438,
  "best_model_checkpoint": "final_models/transformer_base_final_2/finetune/sst2/checkpoint-4000",
  "epoch": 10.0,
  "global_step": 4220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.95,
      "eval_accuracy": 0.8307086825370789,
      "eval_f1": 0.837121212121212,
      "eval_loss": 0.4124297797679901,
      "eval_mcc": 0.6640213069719639,
      "eval_runtime": 0.6988,
      "eval_samples_per_second": 727.0,
      "eval_steps_per_second": 91.591,
      "step": 400
    },
    {
      "epoch": 1.18,
      "learning_rate": 4.407582938388626e-05,
      "loss": 0.4047,
      "step": 500
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.8444882035255432,
      "eval_f1": 0.8429423459244534,
      "eval_loss": 0.4102684557437897,
      "eval_mcc": 0.6890051590962813,
      "eval_runtime": 0.697,
      "eval_samples_per_second": 728.844,
      "eval_steps_per_second": 91.823,
      "step": 800
    },
    {
      "epoch": 2.37,
      "learning_rate": 3.815165876777251e-05,
      "loss": 0.2104,
      "step": 1000
    },
    {
      "epoch": 2.84,
      "eval_accuracy": 0.8602362275123596,
      "eval_f1": 0.855397148676171,
      "eval_loss": 0.43660804629325867,
      "eval_mcc": 0.7216632281287576,
      "eval_runtime": 0.695,
      "eval_samples_per_second": 730.986,
      "eval_steps_per_second": 92.093,
      "step": 1200
    },
    {
      "epoch": 3.55,
      "learning_rate": 3.222748815165877e-05,
      "loss": 0.1344,
      "step": 1500
    },
    {
      "epoch": 3.79,
      "eval_accuracy": 0.8523622155189514,
      "eval_f1": 0.8565965583173996,
      "eval_loss": 0.5639896988868713,
      "eval_mcc": 0.7063807287425677,
      "eval_runtime": 0.7011,
      "eval_samples_per_second": 724.576,
      "eval_steps_per_second": 91.285,
      "step": 1600
    },
    {
      "epoch": 4.74,
      "learning_rate": 2.6303317535545023e-05,
      "loss": 0.0905,
      "step": 2000
    },
    {
      "epoch": 4.74,
      "eval_accuracy": 0.8484252095222473,
      "eval_f1": 0.8538899430740038,
      "eval_loss": 0.5765847563743591,
      "eval_mcc": 0.6993371175737106,
      "eval_runtime": 0.707,
      "eval_samples_per_second": 718.53,
      "eval_steps_per_second": 90.523,
      "step": 2000
    },
    {
      "epoch": 5.69,
      "eval_accuracy": 0.8582677245140076,
      "eval_f1": 0.8599221789883268,
      "eval_loss": 0.6214715242385864,
      "eval_mcc": 0.716921790662046,
      "eval_runtime": 0.7075,
      "eval_samples_per_second": 718.06,
      "eval_steps_per_second": 90.464,
      "step": 2400
    },
    {
      "epoch": 5.92,
      "learning_rate": 2.037914691943128e-05,
      "loss": 0.0594,
      "step": 2500
    },
    {
      "epoch": 6.64,
      "eval_accuracy": 0.8641732335090637,
      "eval_f1": 0.8622754491017963,
      "eval_loss": 0.7745445370674133,
      "eval_mcc": 0.7284623703160482,
      "eval_runtime": 0.6986,
      "eval_samples_per_second": 727.201,
      "eval_steps_per_second": 91.616,
      "step": 2800
    },
    {
      "epoch": 7.11,
      "learning_rate": 1.4454976303317535e-05,
      "loss": 0.0401,
      "step": 3000
    },
    {
      "epoch": 7.58,
      "eval_accuracy": 0.8602362275123596,
      "eval_f1": 0.8599605522682446,
      "eval_loss": 0.7496480941772461,
      "eval_mcc": 0.7204780246823778,
      "eval_runtime": 0.7044,
      "eval_samples_per_second": 721.208,
      "eval_steps_per_second": 90.861,
      "step": 3200
    },
    {
      "epoch": 8.29,
      "learning_rate": 8.530805687203793e-06,
      "loss": 0.0243,
      "step": 3500
    },
    {
      "epoch": 8.53,
      "eval_accuracy": 0.8582677245140076,
      "eval_f1": 0.8604651162790697,
      "eval_loss": 0.7993507981300354,
      "eval_mcc": 0.717130808811096,
      "eval_runtime": 0.6967,
      "eval_samples_per_second": 729.112,
      "eval_steps_per_second": 91.857,
      "step": 3600
    },
    {
      "epoch": 9.48,
      "learning_rate": 2.6066350710900472e-06,
      "loss": 0.0177,
      "step": 4000
    },
    {
      "epoch": 9.48,
      "eval_accuracy": 0.8661417365074158,
      "eval_f1": 0.8656126482213438,
      "eval_loss": 0.8229035139083862,
      "eval_mcc": 0.7322793148880106,
      "eval_runtime": 0.7047,
      "eval_samples_per_second": 720.921,
      "eval_steps_per_second": 90.825,
      "step": 4000
    },
    {
      "epoch": 10.0,
      "step": 4220,
      "total_flos": 3.32486062743552e+16,
      "train_loss": 0.11697568825635865,
      "train_runtime": 1671.74,
      "train_samples_per_second": 302.248,
      "train_steps_per_second": 2.524
    }
  ],
  "max_steps": 4220,
  "num_train_epochs": 10,
  "total_flos": 3.32486062743552e+16,
  "trial_name": null,
  "trial_params": null
}