run-3/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7984344422700587,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/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": 1.4839651584625244,
      "learning_rate": 0.00013176351778592356,
      "loss": 0.4981,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7651663405088063,
      "eval_f1": 0.781021897810219,
      "eval_loss": 0.4340898096561432,
      "eval_precision": 0.7316239316239316,
      "eval_recall": 0.837573385518591,
      "eval_runtime": 28.2114,
      "eval_samples_per_second": 36.227,
      "eval_steps_per_second": 1.134,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3078362941741943,
      "learning_rate": 0.0,
      "loss": 0.436,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7984344422700587,
      "eval_f1": 0.8134057971014492,
      "eval_loss": 0.4166872203350067,
      "eval_precision": 0.7571669477234402,
      "eval_recall": 0.8786692759295499,
      "eval_runtime": 28.4558,
      "eval_samples_per_second": 35.915,
      "eval_steps_per_second": 1.125,
      "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": 0.7300485455769454,
    "learning_rate": 0.0002635270355718471,
    "num_train_epochs": 2,
    "temperature": 7
  }
}