run-6/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7925636007827789,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-6/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": 1.7334731817245483,
      "learning_rate": 0.00019089508944155825,
      "loss": 0.4515,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.761252446183953,
      "eval_f1": 0.7676190476190476,
      "eval_loss": 0.40420547127723694,
      "eval_precision": 0.7476808905380334,
      "eval_recall": 0.7886497064579256,
      "eval_runtime": 28.5902,
      "eval_samples_per_second": 35.747,
      "eval_steps_per_second": 1.119,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9262505769729614,
      "learning_rate": 0.00015907924120129856,
      "loss": 0.3979,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7925636007827789,
      "eval_f1": 0.800376647834275,
      "eval_loss": 0.38181060552597046,
      "eval_precision": 0.7713248638838476,
      "eval_recall": 0.8317025440313112,
      "eval_runtime": 28.4377,
      "eval_samples_per_second": 35.938,
      "eval_steps_per_second": 1.125,
      "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": 0.6366473529685639,
    "learning_rate": 0.00022271093768181798,
    "num_train_epochs": 7,
    "temperature": 14
  }
}