run-16/checkpoint-594/trainer_state.json

{
  "best_metric": 0.6,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-16/checkpoint-594",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 594,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.6842290163040161,
      "learning_rate": 0.00013702102722124262,
      "loss": 0.6194,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5069306930693069,
      "eval_f1": 0.023529411764705882,
      "eval_loss": 0.6244187951087952,
      "eval_mcc": 0.07745844229442526,
      "eval_precision": 1.0,
      "eval_recall": 0.011904761904761904,
      "eval_runtime": 0.9461,
      "eval_samples_per_second": 533.75,
      "eval_steps_per_second": 16.911,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.26532781124115,
      "learning_rate": 0.0001096168217769941,
      "loss": 0.6076,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6,
      "eval_f1": 0.6366906474820144,
      "eval_loss": 0.6058114767074585,
      "eval_mcc": 0.20470740045924352,
      "eval_precision": 0.5822368421052632,
      "eval_recall": 0.7023809523809523,
      "eval_runtime": 0.946,
      "eval_samples_per_second": 533.844,
      "eval_steps_per_second": 16.914,
      "step": 594
    }
  ],
  "logging_steps": 500,
  "max_steps": 1782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1461402531960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8675008492554994,
    "learning_rate": 0.00016442523266549115,
    "num_train_epochs": 6,
    "temperature": 44
  }
}