run-11/checkpoint-138/trainer_state.json

{
  "best_metric": 0.7534246575342466,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-138",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 138,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.9220011234283447,
      "learning_rate": 6.350653635226098e-05,
      "loss": 0.5469,
      "step": 69
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7348336594911937,
      "eval_f1": 0.742150333016175,
      "eval_loss": 0.4790121018886566,
      "eval_precision": 0.7222222222222222,
      "eval_recall": 0.7632093933463796,
      "eval_runtime": 28.1103,
      "eval_samples_per_second": 36.357,
      "eval_steps_per_second": 1.138,
      "step": 69
    },
    {
      "epoch": 2.0,
      "grad_norm": 10.983757972717285,
      "learning_rate": 0.0,
      "loss": 0.4715,
      "step": 138
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7534246575342466,
      "eval_f1": 0.7640449438202248,
      "eval_loss": 0.45558783411979675,
      "eval_precision": 0.7324955116696589,
      "eval_recall": 0.7984344422700587,
      "eval_runtime": 28.896,
      "eval_samples_per_second": 35.368,
      "eval_steps_per_second": 1.107,
      "step": 138
    }
  ],
  "logging_steps": 500,
  "max_steps": 138,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 45,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7864097160999135,
    "learning_rate": 0.00012701307270452197,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 45,
    "temperature": 9
  }
}