run-31/checkpoint-384/trainer_state.json

{
  "best_metric": 0.7808219178082192,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-31/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0108234882354736,
      "learning_rate": 0.0005125121117318405,
      "loss": 0.5275,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7720156555772995,
      "eval_f1": 0.8043660789252729,
      "eval_loss": 0.45868822932243347,
      "eval_mcc": 0.5764707637440674,
      "eval_precision": 0.7044117647058824,
      "eval_recall": 0.9373776908023483,
      "eval_runtime": 66.8111,
      "eval_samples_per_second": 15.297,
      "eval_steps_per_second": 0.479,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.709124565124512,
      "learning_rate": 0.00038438408379888036,
      "loss": 0.4786,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7808219178082192,
      "eval_f1": 0.7773359840954275,
      "eval_loss": 0.45048943161964417,
      "eval_mcc": 0.5619193527696978,
      "eval_precision": 0.7898989898989899,
      "eval_recall": 0.7651663405088063,
      "eval_runtime": 67.4944,
      "eval_samples_per_second": 15.142,
      "eval_steps_per_second": 0.474,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8160345152304423,
    "learning_rate": 0.0006406401396648006,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "temperature": 43
  }
}