run-4/checkpoint-176/trainer_state.json

{
  "best_metric": 0.5430528375733855,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-88",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 176,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4811249077320099,
      "learning_rate": 6.084357840038042e-05,
      "loss": 0.24,
      "step": 88
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5430528375733855,
      "eval_f1": 0.2098138747884941,
      "eval_loss": 0.21115726232528687,
      "eval_precision": 0.775,
      "eval_recall": 0.12133072407045009,
      "eval_runtime": 31.2118,
      "eval_samples_per_second": 32.744,
      "eval_steps_per_second": 1.025,
      "step": 88
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9260004758834839,
      "learning_rate": 5.408318080033815e-05,
      "loss": 0.2091,
      "step": 176
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.541095890410959,
      "eval_f1": 0.20642978003384094,
      "eval_loss": 0.1988278329372406,
      "eval_precision": 0.7625,
      "eval_recall": 0.11937377690802348,
      "eval_runtime": 30.8172,
      "eval_samples_per_second": 33.163,
      "eval_steps_per_second": 1.038,
      "step": 176
    }
  ],
  "logging_steps": 500,
  "max_steps": 880,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 35,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.2634542499188758,
    "learning_rate": 6.760397600042269e-05,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 35,
    "temperature": 2
  }
}