run-1/checkpoint-591/trainer_state.json

{
  "best_metric": 0.7759866592551418,
  "best_model_checkpoint": "tinybert-TG-HS-HX-parentpretrained\\run-1\\checkpoint-591",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 591,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.6037225390588216e-06,
      "loss": 0.2825,
      "step": 197
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7523624235686492,
      "eval_loss": 0.26902320981025696,
      "eval_runtime": 126.6156,
      "eval_samples_per_second": 56.833,
      "eval_steps_per_second": 0.45,
      "step": 197
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.4527919042941157e-06,
      "loss": 0.2724,
      "step": 394
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7718176764869372,
      "eval_loss": 0.2650889456272125,
      "eval_runtime": 126.546,
      "eval_samples_per_second": 56.865,
      "eval_steps_per_second": 0.45,
      "step": 394
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.3018612695294108e-06,
      "loss": 0.2692,
      "step": 591
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7759866592551418,
      "eval_loss": 0.2634750008583069,
      "eval_runtime": 126.5942,
      "eval_samples_per_second": 56.843,
      "eval_steps_per_second": 0.45,
      "step": 591
    }
  ],
  "logging_steps": 500,
  "max_steps": 985,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 133009215775884.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4394092153370026,
    "learning_rate": 5.7546531738235265e-06,
    "num_train_epochs": 5,
    "temperature": 23
  }
}