run-1/checkpoint-428/trainer_state.json

{
  "best_metric": 0.6666666666666666,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-214",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 428,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.7990000247955322,
      "learning_rate": 1.1192818975380277e-05,
      "loss": 0.6296,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.609692394733429,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 3.1451,
      "eval_samples_per_second": 542.745,
      "eval_steps_per_second": 17.169,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7419017553329468,
      "learning_rate": 7.461879316920185e-06,
      "loss": 0.6096,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.6007484197616577,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 3.1123,
      "eval_samples_per_second": 548.47,
      "eval_steps_per_second": 17.351,
      "step": 428
    }
  ],
  "logging_steps": 500,
  "max_steps": 856,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1049551328880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9645989967572975,
    "learning_rate": 1.492375863384037e-05,
    "num_train_epochs": 4,
    "temperature": 9
  }
}