{
  "best_metric": 0.6076320939334638,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-576",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 576,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8380242586135864,
      "learning_rate": 8.6265645867868e-06,
      "loss": 0.6538,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4980430528375734,
      "eval_f1": 0.06215722120658134,
      "eval_loss": 0.6460775136947632,
      "eval_precision": 0.4722222222222222,
      "eval_recall": 0.033268101761252444,
      "eval_runtime": 32.7473,
      "eval_samples_per_second": 31.209,
      "eval_steps_per_second": 0.977,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.982408046722412,
      "learning_rate": 6.901251669429439e-06,
      "loss": 0.6453,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.019193857965451058,
      "eval_loss": 0.6427881717681885,
      "eval_precision": 0.5,
      "eval_recall": 0.009784735812133072,
      "eval_runtime": 30.8528,
      "eval_samples_per_second": 33.125,
      "eval_steps_per_second": 1.037,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.274803638458252,
      "learning_rate": 5.17593875207208e-06,
      "loss": 0.6429,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5313111545988258,
      "eval_f1": 0.18950930626057527,
      "eval_loss": 0.6384090781211853,
      "eval_precision": 0.7,
      "eval_recall": 0.1095890410958904,
      "eval_runtime": 29.2081,
      "eval_samples_per_second": 34.99,
      "eval_steps_per_second": 1.096,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.62313711643219,
      "learning_rate": 3.4506258347147196e-06,
      "loss": 0.6388,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5831702544031311,
      "eval_f1": 0.3622754491017964,
      "eval_loss": 0.6332760453224182,
      "eval_precision": 0.7707006369426752,
      "eval_recall": 0.23679060665362034,
      "eval_runtime": 28.4104,
      "eval_samples_per_second": 35.973,
      "eval_steps_per_second": 1.126,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.8507488965988159,
      "learning_rate": 1.7253129173573598e-06,
      "loss": 0.634,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6046966731898239,
      "eval_f1": 0.43258426966292135,
      "eval_loss": 0.629119336605072,
      "eval_precision": 0.7661691542288557,
      "eval_recall": 0.3013698630136986,
      "eval_runtime": 28.4687,
      "eval_samples_per_second": 35.899,
      "eval_steps_per_second": 1.124,
      "step": 480
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.1146538257598877,
      "learning_rate": 0.0,
      "loss": 0.6325,
      "step": 576
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6076320939334638,
      "eval_f1": 0.4438280166435506,
      "eval_loss": 0.6275492906570435,
      "eval_precision": 0.7619047619047619,
      "eval_recall": 0.3131115459882583,
      "eval_runtime": 28.3455,
      "eval_samples_per_second": 36.055,
      "eval_steps_per_second": 1.129,
      "step": 576
    }
  ],
  "logging_steps": 500,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1414171183680.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9136100763812092,
    "learning_rate": 1.035187750414416e-05,
    "num_train_epochs": 6,
    "temperature": 5
  }
}