run-0/checkpoint-560/trainer_state.json

{
  "best_metric": 0.5524752475247525,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-560",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 560,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8820645213127136,
      "learning_rate": 2.4523034980957572e-05,
      "loss": 0.5259,
      "step": 280
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.502970297029703,
      "eval_f1": 0.07380073800738007,
      "eval_loss": 0.5183532238006592,
      "eval_precision": 0.5263157894736842,
      "eval_recall": 0.03968253968253968,
      "eval_runtime": 15.7192,
      "eval_samples_per_second": 32.126,
      "eval_steps_per_second": 1.018,
      "step": 280
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4514720439910889,
      "learning_rate": 0.0,
      "loss": 0.5176,
      "step": 560
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5524752475247525,
      "eval_f1": 0.37912087912087916,
      "eval_loss": 0.5147280693054199,
      "eval_precision": 0.6160714285714286,
      "eval_recall": 0.27380952380952384,
      "eval_runtime": 14.924,
      "eval_samples_per_second": 33.838,
      "eval_steps_per_second": 1.072,
      "step": 560
    }
  ],
  "logging_steps": 500,
  "max_steps": 560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1555686566280.0,
  "train_batch_size": 34,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6982252235036847,
    "learning_rate": 4.9046069961915145e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 34,
    "temperature": 6
  }
}