{
  "best_metric": 0.8180039138943248,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-3830",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 4596,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.48061203956604,
      "learning_rate": 4.112696231809907e-05,
      "loss": 0.4707,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7847358121330724,
      "eval_f1": 0.8018018018018018,
      "eval_loss": 0.4169985055923462,
      "eval_precision": 0.7429048414023373,
      "eval_recall": 0.8708414872798435,
      "eval_runtime": 130.438,
      "eval_samples_per_second": 7.835,
      "eval_steps_per_second": 1.963,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.212409019470215,
      "learning_rate": 3.290156985447925e-05,
      "loss": 0.4238,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.799412915851272,
      "eval_f1": 0.8038277511961722,
      "eval_loss": 0.40348485112190247,
      "eval_precision": 0.7865168539325843,
      "eval_recall": 0.821917808219178,
      "eval_runtime": 131.3303,
      "eval_samples_per_second": 7.782,
      "eval_steps_per_second": 1.949,
      "step": 1532
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.2959482669830322,
      "learning_rate": 2.4676177390859443e-05,
      "loss": 0.413,
      "step": 2298
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.812133072407045,
      "eval_f1": 0.8212290502793296,
      "eval_loss": 0.4022705852985382,
      "eval_precision": 0.783303730017762,
      "eval_recall": 0.863013698630137,
      "eval_runtime": 131.8559,
      "eval_samples_per_second": 7.751,
      "eval_steps_per_second": 1.942,
      "step": 2298
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.280675411224365,
      "learning_rate": 1.6450784927239627e-05,
      "loss": 0.4075,
      "step": 3064
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8091976516634051,
      "eval_f1": 0.8196114708603145,
      "eval_loss": 0.39988988637924194,
      "eval_precision": 0.7771929824561403,
      "eval_recall": 0.8669275929549902,
      "eval_runtime": 129.5664,
      "eval_samples_per_second": 7.888,
      "eval_steps_per_second": 1.976,
      "step": 3064
    },
    {
      "epoch": 5.0,
      "grad_norm": 18.869022369384766,
      "learning_rate": 8.225392463619813e-06,
      "loss": 0.3963,
      "step": 3830
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8180039138943248,
      "eval_f1": 0.8327338129496402,
      "eval_loss": 0.3992269039154053,
      "eval_precision": 0.7703826955074875,
      "eval_recall": 0.9060665362035225,
      "eval_runtime": 131.856,
      "eval_samples_per_second": 7.751,
      "eval_steps_per_second": 1.942,
      "step": 3830
    },
    {
      "epoch": 6.0,
      "grad_norm": 16.301708221435547,
      "learning_rate": 0.0,
      "loss": 0.3934,
      "step": 4596
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8180039138943248,
      "eval_f1": 0.8299817184643509,
      "eval_loss": 0.39675089716911316,
      "eval_precision": 0.7787307032590052,
      "eval_recall": 0.8884540117416829,
      "eval_runtime": 129.9145,
      "eval_samples_per_second": 7.867,
      "eval_steps_per_second": 1.971,
      "step": 4596
    }
  ],
  "logging_steps": 500,
  "max_steps": 4596,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1414171183680.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.694449127189597,
    "learning_rate": 4.935235478171889e-05,
    "num_train_epochs": 6,
    "temperature": 10
  }
}