run-49/checkpoint-1054/trainer_state.json

{
  "best_metric": 0.841743119266055,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-49/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 16.317338943481445,
      "learning_rate": 0.00023241919324802282,
      "loss": 1.235,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.841743119266055,
      "eval_loss": 1.2949353456497192,
      "eval_runtime": 2.8266,
      "eval_samples_per_second": 308.503,
      "eval_steps_per_second": 2.477,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.788784980773926,
      "learning_rate": 0.00011620959662401141,
      "loss": 0.565,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8302752293577982,
      "eval_loss": 1.419797420501709,
      "eval_runtime": 2.8151,
      "eval_samples_per_second": 309.763,
      "eval_steps_per_second": 2.487,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 1581,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4445543641350709,
    "learning_rate": 0.00034862878987203423,
    "num_train_epochs": 3,
    "temperature": 27
  }
}