run-20/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8072407045009785,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-20/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.3483004570007324,
      "learning_rate": 0.00013855533928571864,
      "loss": 0.5685,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7671232876712328,
      "eval_f1": 0.8019966722129783,
      "eval_loss": 0.477300226688385,
      "eval_precision": 0.6975397973950795,
      "eval_recall": 0.9432485322896281,
      "eval_runtime": 25.843,
      "eval_samples_per_second": 39.546,
      "eval_steps_per_second": 1.238,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.558192729949951,
      "learning_rate": 0.00011546278273809888,
      "loss": 0.4736,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8072407045009785,
      "eval_f1": 0.8255093002657219,
      "eval_loss": 0.43477192521095276,
      "eval_precision": 0.7540453074433657,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 25.942,
      "eval_samples_per_second": 39.396,
      "eval_steps_per_second": 1.234,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 672,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8907547983808969,
    "learning_rate": 0.00016164789583333842,
    "num_train_epochs": 7,
    "temperature": 2
  }
}