run-21/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.8111545988258317,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/checkpoint-1532",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1532,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.5445048809051514,
      "learning_rate": 0.0001354005618154073,
      "loss": 0.4902,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7710371819960861,
      "eval_f1": 0.7943760984182776,
      "eval_loss": 0.4425993263721466,
      "eval_precision": 0.7208931419457735,
      "eval_recall": 0.8845401174168297,
      "eval_runtime": 133.7206,
      "eval_samples_per_second": 7.643,
      "eval_steps_per_second": 1.914,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.882767677307129,
      "learning_rate": 0.00012035605494702872,
      "loss": 0.4409,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8111545988258317,
      "eval_f1": 0.8360237892948172,
      "eval_loss": 0.41539597511291504,
      "eval_precision": 0.7387387387387387,
      "eval_recall": 0.9628180039138943,
      "eval_runtime": 133.4678,
      "eval_samples_per_second": 7.657,
      "eval_steps_per_second": 1.918,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 7660,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7711149722361785,
    "learning_rate": 0.0001504450686837859,
    "num_train_epochs": 10,
    "temperature": 20
  }
}