{
  "best_metric": 0.8268348623853211,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 20.18393898010254,
      "learning_rate": 0.0004400348679504635,
      "loss": 1.3749,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8268348623853211,
      "eval_loss": 1.6724135875701904,
      "eval_runtime": 2.5582,
      "eval_samples_per_second": 340.862,
      "eval_steps_per_second": 2.736,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.33894681930542,
      "learning_rate": 0.00022001743397523176,
      "loss": 0.5714,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8222477064220184,
      "eval_loss": 1.7203704118728638,
      "eval_runtime": 2.5544,
      "eval_samples_per_second": 341.369,
      "eval_steps_per_second": 2.74,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 1581,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.22319309854701086,
    "learning_rate": 0.0006600523019256953,
    "num_train_epochs": 3,
    "temperature": 16
  }
}