run-2/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8140900195694716,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/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": 3.4323604106903076,
      "learning_rate": 0.0004667999543114754,
      "loss": 0.533,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7984344422700587,
      "eval_f1": 0.8049242424242424,
      "eval_loss": 0.4442267119884491,
      "eval_precision": 0.7798165137614679,
      "eval_recall": 0.8317025440313112,
      "eval_runtime": 28.6457,
      "eval_samples_per_second": 35.677,
      "eval_steps_per_second": 1.117,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.4163033962249756,
      "learning_rate": 0.0,
      "loss": 0.4374,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.8243992606284659,
      "eval_loss": 0.4299594759941101,
      "eval_precision": 0.7810858143607706,
      "eval_recall": 0.87279843444227,
      "eval_runtime": 28.93,
      "eval_samples_per_second": 35.327,
      "eval_steps_per_second": 1.106,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.88253398201078,
    "learning_rate": 0.0009335999086229508,
    "num_train_epochs": 2,
    "temperature": 19
  }
}