run-14/checkpoint-1290/trainer_state.json

{
  "best_metric": 0.8914307871267934,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-1290",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1290,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9803465604782104,
      "learning_rate": 0.00027230837357094026,
      "loss": 0.327,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8887165568049632,
      "eval_f1": 0.19154929577464788,
      "eval_loss": 0.3063199520111084,
      "eval_mcc": 0.1990419339488759,
      "eval_precision": 0.48226950354609927,
      "eval_recall": 0.1195079086115993,
      "eval_runtime": 9.9261,
      "eval_samples_per_second": 519.642,
      "eval_steps_per_second": 16.321,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0405207872390747,
      "learning_rate": 0.00023340717734652024,
      "loss": 0.3025,
      "step": 1290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8914307871267934,
      "eval_f1": 0.14893617021276595,
      "eval_loss": 0.3057989478111267,
      "eval_mcc": 0.18620749231762276,
      "eval_precision": 0.550561797752809,
      "eval_recall": 0.08611599297012303,
      "eval_runtime": 9.9284,
      "eval_samples_per_second": 519.521,
      "eval_steps_per_second": 16.317,
      "step": 1290
    }
  ],
  "logging_steps": 500,
  "max_steps": 5160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3173885065200.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9932188265495786,
    "learning_rate": 0.0003112095697953603,
    "num_train_epochs": 8,
    "temperature": 21
  }
}