{
  "best_metric": 0.8314220183486238,
  "best_model_checkpoint": "tiny-bert-sst2-distilled\\run-1\\checkpoint-2108",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 2635,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 7.193342816186545e-05,
      "loss": 1.1191,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.9445732831954956,
      "eval_runtime": 1.8865,
      "eval_samples_per_second": 462.238,
      "eval_steps_per_second": 3.711,
      "step": 527
    },
    {
      "epoch": 2.0,
      "learning_rate": 5.994452346822122e-05,
      "loss": 0.6346,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8279816513761468,
      "eval_loss": 0.8901896476745605,
      "eval_runtime": 1.7665,
      "eval_samples_per_second": 493.618,
      "eval_steps_per_second": 3.963,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.795561877457697e-05,
      "loss": 0.4919,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8211009174311926,
      "eval_loss": 0.9235907196998596,
      "eval_runtime": 1.9102,
      "eval_samples_per_second": 456.496,
      "eval_steps_per_second": 3.665,
      "step": 1581
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.5966714080932725e-05,
      "loss": 0.4192,
      "step": 2108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8314220183486238,
      "eval_loss": 0.9573733806610107,
      "eval_runtime": 1.9272,
      "eval_samples_per_second": 452.465,
      "eval_steps_per_second": 3.632,
      "step": 2108
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.3977809387288486e-05,
      "loss": 0.377,
      "step": 2635
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8268348623853211,
      "eval_loss": 0.9544126987457275,
      "eval_runtime": 1.9749,
      "eval_samples_per_second": 441.534,
      "eval_steps_per_second": 3.544,
      "step": 2635
    }
  ],
  "logging_steps": 500,
  "max_steps": 3689,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 40452267650880.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6227846651931359,
    "learning_rate": 8.39223328555097e-05,
    "num_train_epochs": 7,
    "temperature": 4
  }
}