{
  "best_metric": 0.8325688073394495,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-3162",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 3162,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 31.228694915771484,
      "learning_rate": 7.086160775881376e-05,
      "loss": 2.5988,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 1.924224615097046,
      "eval_runtime": 2.5772,
      "eval_samples_per_second": 338.351,
      "eval_steps_per_second": 2.716,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 51.49909591674805,
      "learning_rate": 5.9055073680756574e-05,
      "loss": 1.2849,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8176605504587156,
      "eval_loss": 1.7333961725234985,
      "eval_runtime": 2.54,
      "eval_samples_per_second": 343.313,
      "eval_steps_per_second": 2.756,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 84.45722961425781,
      "learning_rate": 4.72485396026994e-05,
      "loss": 0.9617,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8245412844036697,
      "eval_loss": 1.7150405645370483,
      "eval_runtime": 2.5578,
      "eval_samples_per_second": 340.916,
      "eval_steps_per_second": 2.737,
      "step": 1581
    },
    {
      "epoch": 4.0,
      "grad_norm": 30.98444938659668,
      "learning_rate": 3.5442005524642224e-05,
      "loss": 0.8154,
      "step": 2108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8222477064220184,
      "eval_loss": 1.6820627450942993,
      "eval_runtime": 2.5749,
      "eval_samples_per_second": 338.654,
      "eval_steps_per_second": 2.719,
      "step": 2108
    },
    {
      "epoch": 5.0,
      "grad_norm": 18.188648223876953,
      "learning_rate": 2.3635471446585048e-05,
      "loss": 0.7233,
      "step": 2635
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8245412844036697,
      "eval_loss": 1.7338383197784424,
      "eval_runtime": 2.5776,
      "eval_samples_per_second": 338.302,
      "eval_steps_per_second": 2.716,
      "step": 2635
    },
    {
      "epoch": 6.0,
      "grad_norm": Infinity,
      "learning_rate": 1.1851340658998569e-05,
      "loss": 0.6653,
      "step": 3162
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8325688073394495,
      "eval_loss": 1.7560914754867554,
      "eval_runtime": 2.5981,
      "eval_samples_per_second": 335.631,
      "eval_steps_per_second": 2.694,
      "step": 3162
    }
  ],
  "logging_steps": 500,
  "max_steps": 3689,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 48527917525620.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.1351029311037084,
    "learning_rate": 8.264573854640023e-05,
    "num_train_epochs": 7,
    "temperature": 16
  }
}