{
  "best_metric": 0.38646939396858215,
  "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_logit_kd_qnli/checkpoint-820",
  "epoch": 7.0,
  "global_step": 2870,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.4069,
      "step": 410
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5680029287937031,
      "eval_loss": 0.3913678824901581,
      "eval_runtime": 8.451,
      "eval_samples_per_second": 646.432,
      "eval_steps_per_second": 2.603,
      "step": 410
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.3877,
      "step": 820
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5905180303862346,
      "eval_loss": 0.38646939396858215,
      "eval_runtime": 8.4822,
      "eval_samples_per_second": 644.053,
      "eval_steps_per_second": 2.594,
      "step": 820
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.3741,
      "step": 1230
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5971078162181951,
      "eval_loss": 0.39170926809310913,
      "eval_runtime": 8.5451,
      "eval_samples_per_second": 639.31,
      "eval_steps_per_second": 2.575,
      "step": 1230
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3604,
      "step": 1640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5928976752699981,
      "eval_loss": 0.38925468921661377,
      "eval_runtime": 8.5294,
      "eval_samples_per_second": 640.488,
      "eval_steps_per_second": 2.579,
      "step": 1640
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.3432,
      "step": 2050
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5921654768442248,
      "eval_loss": 0.39078015089035034,
      "eval_runtime": 8.4767,
      "eval_samples_per_second": 644.474,
      "eval_steps_per_second": 2.595,
      "step": 2050
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3194,
      "step": 2460
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5861248398315944,
      "eval_loss": 0.4250967800617218,
      "eval_runtime": 8.8358,
      "eval_samples_per_second": 618.277,
      "eval_steps_per_second": 2.49,
      "step": 2460
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.2938,
      "step": 2870
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5939959729086582,
      "eval_loss": 0.4414215385913849,
      "eval_runtime": 8.4799,
      "eval_samples_per_second": 644.23,
      "eval_steps_per_second": 2.594,
      "step": 2870
    },
    {
      "epoch": 7.0,
      "step": 2870,
      "total_flos": 4.85626137846743e+16,
      "train_loss": 0.3550731512727638,
      "train_runtime": 1910.5551,
      "train_samples_per_second": 2741.167,
      "train_steps_per_second": 10.73
    }
  ],
  "max_steps": 20500,
  "num_train_epochs": 50,
  "total_flos": 4.85626137846743e+16,
  "trial_name": null,
  "trial_params": null
}