run-3/checkpoint-352/trainer_state.json

{
  "best_metric": 0.5148514851485149,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-176",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 352,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5431386232376099,
      "learning_rate": 0.00012088388613301416,
      "loss": 0.2133,
      "step": 176
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5148514851485149,
      "eval_f1": 0.14035087719298245,
      "eval_loss": 0.20053359866142273,
      "eval_precision": 0.6060606060606061,
      "eval_recall": 0.07936507936507936,
      "eval_runtime": 15.0376,
      "eval_samples_per_second": 33.582,
      "eval_steps_per_second": 1.064,
      "step": 176
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2198910117149353,
      "learning_rate": 0.00010745234322934593,
      "loss": 0.2028,
      "step": 352
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5069306930693069,
      "eval_f1": 0.1263157894736842,
      "eval_loss": 0.20091117918491364,
      "eval_precision": 0.5454545454545454,
      "eval_recall": 0.07142857142857142,
      "eval_runtime": 14.9354,
      "eval_samples_per_second": 33.812,
      "eval_steps_per_second": 1.071,
      "step": 352
    }
  ],
  "logging_steps": 500,
  "max_steps": 1760,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1555686566280.0,
  "train_batch_size": 54,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.239256796645401,
    "learning_rate": 0.0001343154290366824,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 54,
    "temperature": 13
  }
}