{
  "best_metric": 0.824853228962818,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-49/checkpoint-192",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0952401161193848,
      "learning_rate": 0.000329696350517435,
      "loss": 0.5506,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7788649706457925,
      "eval_f1": 0.8055077452667815,
      "eval_loss": 0.46728333830833435,
      "eval_precision": 0.7188940092165899,
      "eval_recall": 0.9158512720156555,
      "eval_runtime": 28.1443,
      "eval_samples_per_second": 36.313,
      "eval_steps_per_second": 1.137,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.149355888366699,
      "learning_rate": 0.0002884843067027556,
      "loss": 0.4507,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.824853228962818,
      "eval_f1": 0.8328664799253034,
      "eval_loss": 0.411376416683197,
      "eval_precision": 0.7964285714285714,
      "eval_recall": 0.87279843444227,
      "eval_runtime": 28.7591,
      "eval_samples_per_second": 35.537,
      "eval_steps_per_second": 1.113,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.500710487365723,
      "learning_rate": 0.0002472722628880762,
      "loss": 0.4184,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8189823874755382,
      "eval_f1": 0.8313582497721058,
      "eval_loss": 0.43684908747673035,
      "eval_precision": 0.7781569965870307,
      "eval_recall": 0.8923679060665362,
      "eval_runtime": 28.8386,
      "eval_samples_per_second": 35.439,
      "eval_steps_per_second": 1.11,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.119166374206543,
      "learning_rate": 0.00020606021907339686,
      "loss": 0.4097,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8189823874755382,
      "eval_f1": 0.8177339901477833,
      "eval_loss": 0.4454614818096161,
      "eval_precision": 0.8234126984126984,
      "eval_recall": 0.812133072407045,
      "eval_runtime": 28.4183,
      "eval_samples_per_second": 35.963,
      "eval_steps_per_second": 1.126,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 942780789120.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8860751863253173,
    "learning_rate": 0.00037090839433211436,
    "num_train_epochs": 9,
    "temperature": 11
  }
}