run-10/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8072407045009785,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-10/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.513129711151123,
      "learning_rate": 0.00015497977730708463,
      "loss": 0.5754,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7837573385518591,
      "eval_f1": 0.8122344944774851,
      "eval_loss": 0.46778297424316406,
      "eval_precision": 0.7177177177177178,
      "eval_recall": 0.9354207436399217,
      "eval_runtime": 24.9847,
      "eval_samples_per_second": 40.905,
      "eval_steps_per_second": 1.281,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.571012020111084,
      "learning_rate": 0.0,
      "loss": 0.4591,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8072407045009785,
      "eval_f1": 0.8267370272647317,
      "eval_loss": 0.4411018490791321,
      "eval_precision": 0.7507987220447284,
      "eval_recall": 0.9197651663405088,
      "eval_runtime": 25.063,
      "eval_samples_per_second": 40.777,
      "eval_steps_per_second": 1.277,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9845885728988275,
    "learning_rate": 0.00030995955461416927,
    "num_train_epochs": 2,
    "temperature": 12
  }
}