{
  "best_global_step": 156,
  "best_metric": 0.7184115523465704,
  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-156",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 1.662625789642334,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.696,
      "step": 10
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 2.0300142765045166,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.6793,
      "step": 20
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 4.4956440925598145,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.6499,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6173285198555957,
      "eval_loss": 0.6330550909042358,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 418.98,
      "eval_steps_per_second": 7.563,
      "step": 39
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 3.44403338432312,
      "learning_rate": 7.948717948717948e-05,
      "loss": 0.6782,
      "step": 40
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 5.425619602203369,
      "learning_rate": 7.435897435897436e-05,
      "loss": 0.5831,
      "step": 50
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 11.772316932678223,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.5307,
      "step": 60
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 3.624584674835205,
      "learning_rate": 6.410256410256412e-05,
      "loss": 0.5396,
      "step": 70
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6859205776173285,
      "eval_loss": 0.6415449380874634,
      "eval_runtime": 0.6613,
      "eval_samples_per_second": 418.848,
      "eval_steps_per_second": 7.56,
      "step": 78
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 3.940481662750244,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.455,
      "step": 80
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 5.196528434753418,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.3532,
      "step": 90
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 4.290834903717041,
      "learning_rate": 4.871794871794872e-05,
      "loss": 0.3482,
      "step": 100
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 4.902711391448975,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.3223,
      "step": 110
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6931407942238267,
      "eval_loss": 0.7661525011062622,
      "eval_runtime": 0.6709,
      "eval_samples_per_second": 412.863,
      "eval_steps_per_second": 7.452,
      "step": 117
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 4.413120269775391,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.2385,
      "step": 120
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 4.013213157653809,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1954,
      "step": 130
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 5.019925117492676,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.1789,
      "step": 140
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 4.320362091064453,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.1594,
      "step": 150
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7184115523465704,
      "eval_loss": 0.8570895791053772,
      "eval_runtime": 0.6607,
      "eval_samples_per_second": 419.255,
      "eval_steps_per_second": 7.568,
      "step": 156
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 971430683050560.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "dropout_rate": 0.01,
    "learning_rate": 0.0001,
    "max_length": 32,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64
  }
}