{
  "best_metric": 1.0977559089660645,
  "best_model_checkpoint": "hBERTv1_new_pretrain_48_KD_mnli/checkpoint-15340",
  "epoch": 10.0,
  "global_step": 30680,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 1.1011,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3175751400916964,
      "eval_loss": 1.100122094154358,
      "eval_runtime": 20.8828,
      "eval_samples_per_second": 470.005,
      "eval_steps_per_second": 3.687,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 1.0991,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.32949566989302087,
      "eval_loss": 1.0978107452392578,
      "eval_runtime": 20.8137,
      "eval_samples_per_second": 471.565,
      "eval_steps_per_second": 3.699,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 1.0983,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3350993377483444,
      "eval_loss": 1.0985406637191772,
      "eval_runtime": 20.7951,
      "eval_samples_per_second": 471.986,
      "eval_steps_per_second": 3.703,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 1.0984,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3293937850229241,
      "eval_loss": 1.0982733964920044,
      "eval_runtime": 20.631,
      "eval_samples_per_second": 475.74,
      "eval_steps_per_second": 3.732,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 1.098,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3264391237901172,
      "eval_loss": 1.0977559089660645,
      "eval_runtime": 20.5966,
      "eval_samples_per_second": 476.536,
      "eval_steps_per_second": 3.738,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 1.098,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.32847682119205296,
      "eval_loss": 1.0978683233261108,
      "eval_runtime": 20.4228,
      "eval_samples_per_second": 480.59,
      "eval_steps_per_second": 3.77,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 1.098,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3271523178807947,
      "eval_loss": 1.0980249643325806,
      "eval_runtime": 20.4047,
      "eval_samples_per_second": 481.017,
      "eval_steps_per_second": 3.774,
      "step": 21476
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 1.098,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.32664289353031073,
      "eval_loss": 1.0981284379959106,
      "eval_runtime": 20.3776,
      "eval_samples_per_second": 481.656,
      "eval_steps_per_second": 3.779,
      "step": 24544
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 1.098,
      "step": 27612
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3256240448293428,
      "eval_loss": 1.0980428457260132,
      "eval_runtime": 20.2631,
      "eval_samples_per_second": 484.378,
      "eval_steps_per_second": 3.8,
      "step": 27612
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.098,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.33204279164544065,
      "eval_loss": 1.0985139608383179,
      "eval_runtime": 20.4136,
      "eval_samples_per_second": 480.806,
      "eval_steps_per_second": 3.772,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "step": 30680,
      "total_flos": 5.81583626085335e+17,
      "train_loss": 1.098501455612879,
      "train_runtime": 24976.3218,
      "train_samples_per_second": 786.149,
      "train_steps_per_second": 6.142
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 5.81583626085335e+17,
  "trial_name": null,
  "trial_params": null
}