{
  "best_metric": 0.8245412844036697,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-4216",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 4216,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 5.104732990264893,
      "learning_rate": 2.9453688815098063e-05,
      "loss": 1.6038,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7752293577981652,
      "eval_loss": 1.2298829555511475,
      "eval_runtime": 2.8117,
      "eval_samples_per_second": 310.129,
      "eval_steps_per_second": 2.49,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.5778963882284596e-05,
      "loss": 1.0293,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8084862385321101,
      "eval_loss": 1.0575315952301025,
      "eval_runtime": 2.8112,
      "eval_samples_per_second": 310.184,
      "eval_steps_per_second": 2.49,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.63373565673828,
      "learning_rate": 2.209725278039734e-05,
      "loss": 0.814,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8004587155963303,
      "eval_loss": 1.0314538478851318,
      "eval_runtime": 2.813,
      "eval_samples_per_second": 309.985,
      "eval_steps_per_second": 2.488,
      "step": 1581
    },
    {
      "epoch": 4.0,
      "grad_norm": 14.250652313232422,
      "learning_rate": 1.841554167851008e-05,
      "loss": 0.6947,
      "step": 2108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8165137614678899,
      "eval_loss": 0.9996209740638733,
      "eval_runtime": 2.826,
      "eval_samples_per_second": 308.569,
      "eval_steps_per_second": 2.477,
      "step": 2108
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.728443145751953,
      "learning_rate": 1.4733830576622822e-05,
      "loss": 0.6281,
      "step": 2635
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.9810827374458313,
      "eval_runtime": 2.8344,
      "eval_samples_per_second": 307.654,
      "eval_steps_per_second": 2.47,
      "step": 2635
    },
    {
      "epoch": 6.0,
      "grad_norm": 33.9716911315918,
      "learning_rate": 1.1052119474735564e-05,
      "loss": 0.5753,
      "step": 3162
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 1.0152171850204468,
      "eval_runtime": 2.8231,
      "eval_samples_per_second": 308.876,
      "eval_steps_per_second": 2.48,
      "step": 3162
    },
    {
      "epoch": 7.0,
      "grad_norm": 38.60210418701172,
      "learning_rate": 7.370408372848306e-06,
      "loss": 0.5522,
      "step": 3689
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8222477064220184,
      "eval_loss": 1.008946418762207,
      "eval_runtime": 2.8157,
      "eval_samples_per_second": 309.697,
      "eval_steps_per_second": 2.486,
      "step": 3689
    },
    {
      "epoch": 8.0,
      "grad_norm": 33.359806060791016,
      "learning_rate": 3.6886972709610482e-06,
      "loss": 0.5295,
      "step": 4216
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8245412844036697,
      "eval_loss": 1.0115498304367065,
      "eval_runtime": 2.8207,
      "eval_samples_per_second": 309.143,
      "eval_steps_per_second": 2.482,
      "step": 4216
    }
  ],
  "logging_steps": 500,
  "max_steps": 4743,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 64687167744780.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.698062417678107,
    "learning_rate": 3.3135399916985324e-05,
    "num_train_epochs": 9,
    "temperature": 29
  }
}